Protocol. This trial protocol has been provided by the authors to give readers additional information about their work.

Transcription

1 Protocol This trial protocol has been provided by the authors to give readers additional information about their work. Protocol for: Joura EA, Giuliano AR, Iversen O-E, et al. A 9-valent HPV vaccine against infection and intraepithelial neoplasia in women. N Engl J Med 2015;372: DOI: /NEJMoa

2 This supplement contains the following items: 1. Original protocol, final protocol, summary of changes. 2. Original statistical analysis plan, final statistical analysis plan, summary of changes

3 Merck s policy on posting of redacted study protocols on journal websites is described in Merck Guidelines for Publication of Clinical Trials in the Scientific Literature on the website. For publicly posted protocols, Merck redacts the background and rationale sections because these sections may contain proprietary information. Merck also redacts the names of any individuals due to privacy issues. The appendices generally are not provided because they may be lengthy and contain non-essential information. The publicly posted protocol includes all the key sections that are relevant to evaluating the study, specifically those sections describing the study objectives and hypotheses, the patient inclusion and exclusion criteria, the study design and procedures, the efficacy and safety measures, the statistical analysis plan, and amendments relating to those sections. This report may include approved and non-approved uses, formulations, or treatment regimens. The results reported may not reflect the overall profile of a product. Before prescribing any product mentioned this report, healthcare professionals should consult local prescribing information for the product approved in their country. Copyright 2014 Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. All Rights Reserved. Not for regulatory or commercial use.

4 A Randomized, International, Double-Blinded (With In-House Blinding), Controlled With GARDASIL, Dose-Ranging, Tolerability, Immunogenicity, and Efficacy Study of a Multivalent Human Papillomavirus (HPV) L1 Virus-Like Particle (VLP) Vaccine Administered to 16- to 26-Year-Old Women

5 V503, Protocol Issue Date: 15-Jun Product: V503 Protocol/Amendment No.: THIS PROTOCOL AND ALL OF THE INFORMATION RELATING TO IT ARE CONFIDENTIAL AND PROPRIETARY PROPERTY OF MERCK & CO., INC., WHITEHOUSE STATION, NJ, U.S.A. SPONSOR: Merck & Co., Inc. (hereafter referred to as the SPONSOR) One Merck Drive P.O. Box 100 Whitehouse Station, NJ, , U.S.A. Protocol-specific Sponsor Contact information can be found in the Administrative Binder. TITLE: A Randomized, International, Double-Blinded (With In-House Blinding), Controlled With GARDASIL, Dose-Ranging, Tolerability, Immunogenicity, and Efficacy Study of a Multivalent Human Papillomavirus (HPV) L1 Virus-Like Particle (VLP) Vaccine Administered to 16- to 26-Year-Old Women INVESTIGATOR: PRIMARY: CLINICAL PHASE: IIb US IND NUMBER: SITE: INSTITUTIONAL REVIEW BOARD/ETHICS REVIEW COMMITTEE: V503_001-00_ProtTitle VERSION 1.2 APPROVED 15-Jun-2007

6 V503, Protocol Issue Date: 15-Jun PROTOCOL A Randomized, International, Double-Blinded (With In-House Blinding), Controlled With GARDASIL, Dose-Ranging, Tolerability, Immunogenicity, and Efficacy Study of a Multivalent Human Papillomavirus (HPV) L1 Virus-Like Particle (VLP) Vaccine Administered to 16- to 26-Year-Old Women TABLE OF CONTENTS Contents Application Starting Page 1. SUMMARY Title Indication Summary of Rationale Summary of Study Design Sample Dosage/Dosage Form, Route, and Dose Regimen Study Flow Chart CORE PROTOCOL Objectives and Hypotheses Part A Analysis Primary Part B Analysis (Tolerability and Efficacy Analyses Include Part A subjects who received the selected 9-valent HPV L1 VLP vaccine dose or the comparator GARDASIL ) Primary Secondary Exploratory Subject/Patient Inclusion Criteria Subject/Patient Exclusion Criteria Study Design and Duration Summary of Study Design 21 14

7 V503, Protocol Issue Date: 15-Jun TABLE OF CONTENTS (CONT.) Contents Application Starting Page Vaccination Plan List of Immunogenicity and efficacy Measurements Immunogenicity Measurements Efficacy Measurements List of Safety Measurements Data Analysis Summary PROTOCOL DETAILS Rationale Introduction to Human Papillomavirus Disease Burden Biology of HPV Epidemiology of HPV Rationale for This Study Rationale for Part A Doses Summary of GARDASIL, 8-valent HPV L1 VLP Vaccine, and 9-valent HPV L1 VLP Vaccine Studies 3.2 Study Procedures Concomitant Medication(s)/Treatment(s) Prerequisites for Study Visits Summary of Scheduled Study Visit Procedures Calculation of Scheduled Visit Windows Informed Consent and Assignment of Baseline Numbers History & Medications Pregnancy Testing and Serum Collection Physical Examination & Vital Signs Gynecological Examination Genital/Cervical Swabs, Pap Test, and Sexually Transmitted Infection (STI) Testing 33 38

8 V503, Protocol Issue Date: 15-Jun TABLE OF CONTENTS (CONT.) Contents Application Starting Page External Genital Lesion Examination IVRS Assignment of Allocation Numbers and Vaccination Vials Study Vaccine Administration Clinical Follow-Up Summary of Unscheduled Study Visit Procedures Colposcopy, Cervical Biopsy, and Cervical Definitive Therapy External Genital Lesion and Vaginal Lesion Diagnosis and Follow-Up Processing of Tissue Specimens In the Context of the Study Pap Tests and Tissue Specimens Taken Outside the Context of the Study Procedures for Collection and Handling of Study Specimens Serum or Urine Specimen for Pregnancy Test Serum for Anti-HPV Measurements at Scheduled Visits Cord Serum and Maternal Serum for Anti-HPV Measurements (Applicable to Participating Study Sites) Labial/Vulvar/Perineal and Perianal (LVPP) Swabs for HPV PCR Endo/Ectocervical (EEC) Swab for HPV PCR Pap Test (ThinPrep ) Specimen Collection External Genital Lesion Biopsy and Vaginal Lesion Biopsy Colposcopy Guidelines Procedures for Cervical Biopsy Procedures for Endocervical Curettage (ECC) Procedures for Loop Electrosurgical Excision Procedure (LEEP) and Top Hat Method Procedure for Laser Conization

9 V503, Protocol Issue Date: 15-Jun TABLE OF CONTENTS (CONT.) Contents Application Starting Page Cold-Knife Conization Ablative Cervical Definitive Therapy Discontinuation/Withdrawal from Study Subject Relocation Conduct of the Clinical Trial Scientific Advisory Committee HPV Vaccine Program Pathology Panel Responsibility of the Data and Safety Monitoring Board (DSMB) Senior Management Committee for Dose Selection Efficacy/Immunogenicity Measurements Competitive Luminex Immunoassay (clia) - Anti-HPV Levels in Serum PCR Assays - Detection of HPV in Swabs and Tissue Specimens Multiplex PCR Assays Preparation and Disposition of Thinsections of Biopsy Tissue 3.4 Safety Measurements Clinical and Laboratory Measurements for Safety Recording Adverse Experiences Definition of an Overdose for This Protocol Reporting of Overdose to SPONSOR Reporting of Pregnancy/Breastfeeding Events to the SPONSOR Immediate Reporting of Adverse Experiences to the SPONSOR Serious Adverse Experiences Evaluating Adverse Experiences

10 V503, Protocol Issue Date: 15-Jun TABLE OF CONTENTS (CONT.) Contents Application Starting Page SPONSOR Responsibility for Reporting Adverse Experiences Data Analysis Responsibility for and Timing of Analyses Hypotheses Variables and Time Points of Interest Safety/Tolerability Efficacy Immunogenicity Analysis Populations Safety Efficacy Immunogenicity Statistical Methods Efficacy Immunogenicity Safety Multiplicity Considerations Sample Size and Power Calculations Interim Analyses Definition of Compliance Measure Labeling, Packaging, Storage, Dispensing, and Return of Clinical Supplies Subject and Replacements Information Product descriptions Primary Packaging and Labeling Information Secondary Packaging and Labeling Information- Supplies Will Not be Kitted Clinical Supplies-Vaccine Disclosure

11 V503, Protocol Issue Date: 15-Jun TABLE OF CONTENTS (CONT.) Contents Application Starting Page Storage Requirements Standard Policies / Return of Clinical Supplies Transport of the Vaccine Data Management Biological Specimens ADMINISTRATIVE AND REGULATORY DETAILS Confidentiality Confidentiality of Data Confidentiality of Subject/Patient Records Confidentiality of Investigator Information Compliance with Law, Audit, and Debarment Compliance with Financial Disclosure Requirements Quality Control and Quality Assurance Compliance with Information Program on Clinical Trials for Serious or Life Threatening Conditions 4.6 Publications LIST OF REFERENCES APPENDICES ATTACHMENTS SIGNATURES SPONSOR S REPRESENTATIVE INVESTIGATOR

12 V503, Protocol Issue Date: 15-Jun Product: V503 1 Protocol/Amendment No.: TITLE 1. SUMMARY A Randomized, International, Double-Blinded (With In-House Blinding), Controlled With GARDASIL, Dose-Ranging, Tolerability, Immunogenicity, and Efficacy Study of a Multivalent Human Papillomavirus (HPV) L1 Virus-Like Particle (VLP) Vaccine Administered to 16- to 26-Year-Old Women 1.2 INDICATION Prevention of cervical, vulvar, and vaginal cancers and related precancers, external genital lesions, Pap test abnormalities, and persistent infection caused by Human Papillomavirus (HPV) 6, 11, 16, 18, 31, 33, 45, 52, and SUMMARY OF RATIONALE Redacted V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

13 Product: V503 2 Protocol/Amendment No.: Redacted V503, Protocol Issue Date: 15-Jun SUMMARY OF STUDY DESIGN Part A (dose-ranging and tolerability): Approximately 1,240 healthy 16- to 26-yearold women will be randomized in equal numbers to one of three 9-valent HPV L1 VLP vaccine formulations or the comparator GARDASIL in a three dose regimen (Day 1, Month 2, and Month 6). The best 9-valent HPV L1 VLP vaccine dose, among those doses tested, will be selected for Part B after approximately 100% of the Part A post-dose 2 immunogenicity and adverse experience data are available in the clinical database. The dose for Part B will be chosen by a senior management committee at the SPONSOR who is not directly involved with study conduct. This committee will choose the dose after reviewing immunogenicity summaries by vaccination group, and will relay the dose selection to the Data Safety Monitoring Board (DSMB), who will approve the decision based on unblinded tolerability data. If the DSMB expresses a concern about the tolerability profile of the selected dose, the senior management committee will select another dose and again request approval from the DSMB. In the unlikely event that the SPONSOR is unable to determine an acceptable 9-valent HPV L1 VLP vaccine formulation for use in Part B, the study may not continue to Part B. Part B (tolerability, immunogenicity, and efficacy): Approximately 2,480 additional healthy 16- to 26-year-old women will be randomized in equal numbers to the 9-valent HPV L1 VLP vaccine dose selected from Part A or the comparator GARDASIL. The primary immunogenicity analyses for Part B will be the formal comparison of the selected 9-valent formulation to the GARDASIL control and will include approximately 1,500 (the first 60% enrolled at each site) of the 2,480 additional subjects enrolled in Part B. The overall tolerability and efficacy analyses, however, will also include subjects enrolled in Part A (those who received the selected 9-valent HPV L1 VLP vaccine dose or the comparator GARDASIL ) in addition to all subjects enrolled in Part B. The primary efficacy analysis is case driven and will be conducted after 43 primary efficacy cases have been observed. 1.5 SAMPLE This study will enroll healthy 16- to 26-year-old women. To minimize the number of enrolled subjects who are positive to study vaccine HPV types at baseline, subjects who have a history of 5 or more sexual partners, external genital or vaginal warts, a history of an abnormal Pap test, or a history of an abnormal cervical biopsy result will be excluded. V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

14 V503, Protocol Issue Date: 15-Jun Product: V503 3 Protocol/Amendment No.: DOSAGE/DOSAGE FORM, ROUTE, AND DOSE REGIMEN In Part A, subjects will be given 9-valent HPV L1 VLP vaccine or GARDASIL as summarized in Table 1-1. The rationale for these formulations can be found in Section Table 1-1 Part A Vaccine Formulations (0.5-mL Dose) Arm 1 HPV 6/ 11/ 16/ 18 20/ 40/ 40/ 20 (GARDASIL ) HPV 31 (mcg) HPV 33 (mcg) HPV 45 (mcg) HPV 52 (mcg) HPV 58 (mcg) Total VLP (mcg) AAHS (mcg) Estimated Subjects / 40/ 40/ / 40/ 60/ / 40/ 80/ AAHS=Merck Aluminum Adjuvant (amorphous aluminum hydroxyl phosphate sulfate) In Part B, approximately 2,480 additional subjects will be randomized in equal numbers to receive the selected 9-valent HPV L1 VLP vaccine or GARDASIL. Study vaccine will be administered as a 0.5-mL intramuscular injection at Day 1, Month 2, and Month 6. V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

15 Product: V503 4 Protocol/Amendment No.: STUDY FLOW CHART Key Scheduled Tests and Events Visit Windows : No visit window; Only applies to the first 150 subjects enrolled (for return of VRC data) Obtain Informed Consent, Assign Baseline Number + Review Inclusion/Exclusion Criteria + Collect Medical History (past year)/lifetime Gynecologic History + Update Medical History and Gynecologic History (new conditions not V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007 Day Months 1 Day months after Day 1, ±3 weeks 3 to 5 weeks after Month 2 6 months after Day 1, ±4 weeks 3 to 7 weeks after Month 6 12, 18, 24, 30, 36, 42 months after Day 1,±4 weeks; Mo. 7 is the final visit for 9-valent doses that are not selected already recorded as Medical History or Adverse Experiences) Review Medications and Non-Study Vaccines Pregnancy/Serum Specimen Collection: Pregnancy Test (serum or urine) Serum for Anti-HPV (including Retention Serum) Additional Serum for Anti-HPV Assay Development and Reference + Samples Perform Physical Examination Collect Vital Signs (sitting pulse, weight, blood pressure, respirations, and oral temperature) Perform Gynecologic Physical Examination # Pelvic Specimen Collection (obtain in serial order): Labial/Vulvar/Perineal & Perianal (LVPP) Swabs for HPV PCR Endo/Ectocervical (EEC) Swab for HPV PCR Pap Test (ThinPrep for cytology, Chlamydia/gonorrhea testing at Day 1, Month 7, Month 18, Month 30, Month 42) STI Testing (local laboratory testing, if clinically indicated) Perform External Genital Wart Inspection Assign Allocation Number + Perform Vaccination (intramuscular [i.m.]; prefer deltoid muscle of nondominant arm; do not give in buttocks) Provide Vaccination Report Card (VRC) Review and Collect VRC data Review Adverse Experiences, Clinical Follow-up for Safety V503, Protocol Issue Date: 15-Jun

16 Product: V503 5 Protocol/Amendment No.: Study Flow Chart Footnotes To calculate visit windows, assume 1 month equals 30 days and 1 week equals 7 days. With regard to protocol study visit windows, the following situations require consultation between the investigator and the SPONSOR and written documentation of the collaborative decision: a subject needs to be scheduled earlier than the start of a visit window, the study site is considering skipping a visit, or a study site needs significant guidance on scheduling visit windows. For post Month 7 visits, if a subject is very late for a visit such that she is in the next visit window, schedule future visits a minimum of 4 months apart until the visits are within the protocol-specified window. See the Administrative Binder for a summary of deviations that require documentation in this study. The Day 16 visit only applies to the first 150 subjects enrolled in Part A. There is no visit window on the Day 16 visit, because the purpose of the visit is to obtain 40 VRCs in a timely manner for an early tolerability evaluation. Although data from only 40 VRCs are needed for the tolerability evaluation, the Day 16 visit will be scheduled for more than 40 subjects to allow for lost VRCs, cancelled visits, or other scenarios that may delay the return of the VRCs. See the protocol details for prerequisites for medications and non-study vaccines. The serum pregnancy test or urine pregnancy test (sensitive to 25 miu/ml β-hcg) will be performed per the manufacturer s instructions. All materials used for serum and/or urine pregnancy testing will provided by the study sites. Serum for anti-hpv measurements (by competitive Luminex Immunoassay, clia) may be collected after the pelvic examination, but must be collected before vaccination. clia testing to be performed at Merck Research Laboratories (MRL). Serum (including Reference Serum) must be shipped as specified by the SPONSOR/Central Laboratory. The Retention Serum vial must remain at the site until the SPONSOR notifies the study site to ship the samples. If the subject has a fever (defined as an oral temperature of 100 F or 37.8 C) within the 24-hour period prior to receiving a study vaccination, the subject should not receive study vaccine, and the vaccination visit should be rescheduled until after the fever has resolved. # Collect pelvic study specimens prior to performing the bimanual pelvic examination. See the protocol details for prerequisites for pelvic sample collection. Type-specific HPV Polymerase Chain Reaction (PCR) testing to be performed at MRL. The 2 LVPP swabs are placed in one container of specimen transfer medium (STM) and 1 EEC swab is placed in a separate container of STM. Swabs must be shipped as specified by the SPONSOR/Central Laboratory. Pap test to be analyzed by the SPONSOR-designated Central Laboratory. The Pap test fluid will be evaluated for Chlamydia and gonorrhea at the specified visits, but additional local laboratory tests for Chlamydia and gonorrhea may be performed at other visits. Other sexually transmitted infection (STI) tests, including HSV, syphilis, and HIV, may be performed at any visit as needed. Observe subjects for 30 minutes after each vaccination for immediate untoward effects. The subject should use the Vaccination Report card (VRC) to document this safety information: oral temperature in the evening after each study vaccination and daily, at the same time of day, for 4 days after each study vaccination; injection-site or systemic adverse experience(s) starting after each study vaccination for a total of 15 days. The first 150 subjects enrolled will return the Day 1 VRC to the study site at Day 16, while the remaining subjects will return the Day 1 VRC at the Month 2 study visit. For all subjects, Month 2 VRCs will be returned at the Month 3 study visit and Month 6 VRCs will be returned at the Month 7 study visit. Serious adverse experiences (SAEs), pregnancy events (including non-randomized subjects with a positive Day 1 pregnancy test), and lactation events should be reported to the SPONSOR as instructed in the protocol and in the protocol Attachments. V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007 V503, Protocol Issue Date: 15-Jun

17 V503, Protocol Issue Date: 15-Jun Product: V503 6 Protocol/Amendment No.: CORE PROTOCOL 2.1 OBJECTIVES AND HYPOTHESES Part A Analysis Primary (1) Objective: To evaluate the tolerability of the 9-valent HPV L1 VLP vaccine when administered to 16- to 26-year-old women. Hypothesis: 9-valent HPV L1 VLP vaccine administered to 16- to 26-year-old women is generally well-tolerated. (2) Objective: To evaluate a formulation of 9-valent HPV L1 VLP vaccine for use in the efficacy evaluation in Part B. Hypothesis: 9-valent HPV L1 VLP vaccine selected for use in Part B generates anti-hpv 6, 11, 16, and 18 GMTs 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The GMTs for each of HPV types 6, 11, 16, and 18 will be tested separately. Noninferiority for GMTs is defined as statistically less than a 2-fold decrease.) Part B Analysis (Tolerability and Efficacy Analyses Include Part A subjects who received the selected 9-valent HPV L1 VLP vaccine dose or the comparator GARDASIL ) Primary (1) Objective: To evaluate the tolerability of the 9-valent HPV L1 VLP vaccine when administered to 16- to 26-year-old women. Hypothesis: 9-valent HPV L1 VLP vaccine administered to 16- to 26-year-old women is generally well-tolerated. (2) Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine will reduce the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection for a duration of 6 months (within ± 1 month windows) or longer and cervical, vulvar, and vaginal disease compared with GARDASIL in 16- to 26-year-old V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

18 V503, Protocol Issue Date: 15-Jun Product: V503 7 Protocol/Amendment No.: adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type. (3) Hypothesis: Administration of 9-valent HPV L1 VLP vaccine to 16- to 26-yearold adolescent and young-adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection for a duration of 6 months (within ± 1 month windows) or longer and cervical, vulvar, and vaginal disease compared with GARDASIL. (The criterion for success is defined as statistically greater than 20% efficacy. See section for the definition of HPV 31-,33-,45-,52-, and 58-related persistent infection and cervical, vulvar and vaginal disease.) Objective: To demonstrate that the 9-valent HPV L1 VLP vaccine induces noninferior GMTs for anti-hpv 6, 11, 16, and 18 compared to GARDASIL. (1) Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 6, 11, 16, and 18 GMTs 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The GMTs for each of HPV types 6, 11, 16, and 18 will be tested separately. Noninferiority for GMTs is defined as statistically less than a 1.5-fold decrease.) Secondary (1) Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine will reduce the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related documented infection for 12 months and cervical, vulvar, and vaginal disease, compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative and PCR negative at Day 1 to the relevant HPV type. Hypothesis: Administration of 9-valent HPV L1 VLP vaccine to 16- to 26-yearold adolescent and young-adult women who are seronegative and PCR negative at Day 1 to the relevant HPV type reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related documented infection for 12 months and cervical, vulvar, and vaginal disease, compared with GARDASIL. (The criterion for success is defined as statistically greater than 0% efficacy.) V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

19 V503, Protocol Issue Date: 15-Jun Product: V503 8 Protocol/Amendment No.: (2) Objective: To evaluate whether the administration of 9-valent HPV L1 VLP vaccine results in a combined incidence of persistent HPV 16 and 18 infection for a duration of 6 months (within ± 1 month windows) or longer and HPV 16- and 18-related cervical, vulvar, and vaginal disease that is observationally comparable to the combined incidence observed with GARDASIL. (3) Objective: To evaluate whether the administration of 9-valent HPV L1 VLP vaccine results in a combined incidence of HPV 6- and 11-related cervical, vulvar, and vaginal disease that is observationally comparable to the combined incidence observed with GARDASIL. (4) Objective: To demonstrate that the 9-valent HPV L1 VLP vaccine induces noninferior immune responses with respect to seroconversion percentages for HPV 6, 11, 16, and 18 compared to GARDASIL. (5) Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 6, 11, 16, and 18 seroconversion percentages by 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The seroconversion rates for each of HPV types 6, 11, 16, and 18 will be tested separately. Noninferiority is defined as statistically less than a decrease of 5 percentage points.) Objective: To demonstrate that 9-valent HPV L1 VLP vaccine formulation for use in Phase III trials is immunogenic with respect to HPV types 31, 33, 45, 52, and 58. (6) Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 31, 33, 45, 52 and 58 seroconversion rates 4 weeks postdose 3 that are acceptable in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The seroconversion rates for each of HPV types 31, 33, 45, 52, and 58 will be tested separately. Acceptability is defined as a seroconversion percentage statistically greater than 0.9 (90%).) Objective: To evaluate the persistence of anti-hpv 6, 11, 16, 18, 31, 33, 45, 52, and 58 responses generated by 9-valent HPV L1 VLP vaccine. V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

20 V503, Protocol Issue Date: 15-Jun Product: V503 9 Protocol/Amendment No.: (7) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of Pap test abnormalities (ASC-US [Positive for High Risk HPV] or worse) Exploratory (1) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the combined incidence of CIN, AIS, and cervical cancer caused by any HPV type. (2) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the combined incidence of vulvar and vaginal disease caused by any HPV type. (3) Objective: To characterize the titers of anti-hpv type 6/11 in both peripartum maternal blood and in cord blood of infants born to subjects for evaluating the potential impact of 9-valent HPV L1 VLP vaccine on recurrent respiratory papillomatosis. 4 (4) Objective: To evaluate the efficacy of the selected formulation of 9-valent HPV L1 vaccine against persistent HPV 35-, 39-, 51-, 56- and 59-related infection for a duration of 6 months (within ± 1 month windows) or longer and HPV 35-, 39-, 51-, 56-, and 59- related cervical, vulvar, and vaginal disease. (5) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of Pap test abnormalities (ASC-US [Positive for High Risk HPV] or worse) related to HPV Types 31, 33, 45, 52, & This objective is applicable to study sites that are able to collect and submit peripartum and cord serum samples for analysis. V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

21 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: (6) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of cervical biopsy and cervical definitive therapy treatments. 2.2 SUBJECT/PATIENT INCLUSION CRITERIA To be randomized and receive the first study vaccination, subjects should meet all inclusion criteria. For items with an asterisk (*), if the subject does not meet these inclusion criteria, the Day 1 visit may be rescheduled for a time when these criteria can be met. 1. Subject is female, between the ages of 16 years and 0 days and 26 years and 364 days on the day of randomization. 2. Subject has never had Pap testing or has only had normal Pap test results. 3. Subject (or, for minor subjects, parent/legal guardian and subject) fully understands study procedures, alternative treatments available, the risks involved with the study, and voluntarily agrees to participate by giving written informed consent. 4. Subject is able to read, understand, and complete the vaccination report card. 5. Subject is judged to be in good physical health on the basis of medical history, physical examination, and laboratory results. 6. The subject has the following lifetime sexual history at the time of enrollment: a) Subject has had 1 to 4 male and/or female sexual partners; or b) Subject has had 0 male and/or female sexual partners, is 18 years of age or older, and plans to become sexually active within the first 3 months of the study. Male partner is defined as someone with whom the subject has penile penetrative sexual intercourse. Female partner is defined as someone who has contacted, either by penetrative (with fingers or other objects) or non-penetrative means, the subject s genitalia during sexual activity. 7. *Subject has refrained from douching/vaginal cleansing and using vaginal medications or preparations for 2 calendar days prior to the Day 1 visit. Subject agrees to refrain from these activities for 2 calendar days prior to any future visit that includes collection of study specimens (cervical/genital swabs, Pap test, or biopsy/definitive therapy tissue). 8. *Subject has refrained from sexual activity (including anal, vaginal, or genital/genital contact whether same sex or opposite sex) for 2 calendar days prior to the Day 1 visit. Subject agrees to refrain from these sexual activities for 2 calendar days prior to any V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

22 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: future visit that includes collection of study specimens (cervical/genital swabs, Pap test, or biopsy/definitive therapy tissue). 9. *Since the first day of the subject s last menstrual period through Day 1, the subject has not had sex with males or has had sex with males and used effective contraception with no failures (an example of a failure is a male condom that ruptures during sexual intercourse). Effective contraception is defined as a marketed, approved contraceptive product that the subject has used per the manufacturer s instructions with every act of sexual intercourse. The subject understands and agrees that during the Day 1 through Month 7 period, she should not have sexual intercourse with males without effective contraception, and the uses of the rhythm method alone, withdrawal alone, and emergency contraception, are not acceptable methods per the protocol. 2.3 SUBJECT/PATIENT EXCLUSION CRITERIA To be randomized and receive the first study vaccination, subjects should not have any exclusion criteria. For items with an asterisk (*), if the subject meets these exclusion criteria, the Day 1 visit may be rescheduled for a time when these criteria are not met. 1. Subject has a history of an abnormal cervical biopsy result (showing cervical intraepithelial neoplasia [CIN] or worse). 2. Subject has a history of a positive test for HPV. 3. Subject is, at the time of signing informed consent, a user of recreational or illicit drugs or has had a recent history (within the last year) of drug or alcohol abuse or dependence. Alcohol abusers are defined as those who drink despite recurrent social, interpersonal, and/or legal problems as a result of alcohol use. 4. Subject has a history of severe allergic reaction (e.g., swelling of the mouth and throat, difficulty breathing, hypotension or shock) that required medical intervention. 5. Subject has known allergy to any vaccine component, including aluminum, yeast, or BENZONASE (nuclease, Nycomed [used to remove residual nucleic acids from this and other vaccines]). For the purpose of this exclusion criterion, an allergy to vaccine components is defined as an allergic reaction that met the criteria for serious adverse experiences defined in Section Subject is currently immunocompromised or has been diagnosed as having a congenital or acquired immunodeficiency, HIV infection, lymphoma, leukemia, systemic lupus erythematosus (SLE), rheumatoid arthritis, juvenile rheumatoid arthritis (JRA), inflammatory bowel disease, or other autoimmune condition. 7. Subject has had a splenectomy. V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

23 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Subject is receiving or has received in the year prior to enrollment the following immunosuppressive therapies: radiation therapy, cyclophosphamide, azathioprine, methotrexate, any chemotherapy, cyclosporin, leflunomide (Arava ), TNF-α antagonists, monocolonal antibody therapies (including rituximab [Rituxan ]), intravenous gamma globulin (IVIG), antilymphocyte sera, or other therapy known to interfere with the immune response. With regard to systemic corticosteroids, a subject will be excluded if she is currently receiving steroid therapy, has recently (defined as within 2 weeks of enrollment) received such therapy, or has received 2 or more courses of high dose corticosteroids (orally or parenterally) lasting at least 1 week in duration in the year prior to enrollment. Subjects using inhaled, nasal, or topical corticosteroids are considered eligible for the study. 9. Subject has received any immune globulin product (including RhoGAM [Ortho- Clinical Diagnostics]) or blood-derived product within the 3 months prior to the Day 1 vaccination, or plans to receive any such product during Day 1 through Month 7 of the study. 10. *Subject has received non-replicating (inactivated) vaccines within 14 days prior to the Day 1 vaccination or has received replicating (live) vaccines within 21 days prior to the Day 1 vaccination. 11. Subject has thrombocytopenia or other coagulation disorder that would contraindicate intramuscular injections. 12. *Subject has donated blood within 1 week prior to the Day 1 vaccination, or intends to donate during Day 1 through Month 7 of the study. 13. Subject is expecting to donate eggs during Day 1 through Month 7 of the study. 14. Subject is concurrently enrolled in clinical studies of investigational agents or studies involving collection of cervical specimens. 15. Subject has received a marketed HPV vaccine, or has participated in an HPV vaccine clinical trial and has received either active agent or placebo. 16. Subject has a history or current evidence of any condition, therapy, lab abnormality or other circumstance that might confound the results of the study, or interfere with the subject s participation for the full duration of the study, such that it is not in the best interest of the subject to participate. 17. Subject is unlikely to adhere to the study procedures, keep appointments, or is planning to relocate during the study. 18. *Subject has had a fever (defined as an oral temperature of 100 F or 37.8 C) within the 24-hour period prior to the Day 1 vaccination. V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

24 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Subject is pregnant (as determined by a serum pregnancy test or urine pregnancy test that is sensitive to 25 miu/ml β-hcg). 20. *Subject has clinical evidence of gross purulent cervicitis. 21. *Subject is having menses. 22. Subject has a history of or clinical evidence at the Day 1 pelvic examination of HPVrelated external genital lesions (e.g., condyloma acuminata or vulvar intraepithelial neoplasia [VIN]) or external genital cancer, HPV-related vaginal lesions (e.g., condyloma acuminata or vaginal intraepithelial neoplasia [VaIN]) or vaginal cancer. 23. Subject does not have an intact cervix uteri or has more than one cervix uteri. 2.4 STUDY DESIGN AND DURATION Summary of Study Design This is a randomized, international, multi-centered, double-blinded (with in-house blinding procedures), dose-ranging, tolerability, immunogenicity, and efficacy study of the 9-valent HPV L1 VLP vaccine. This study will be conducted in two parts (Part A and Part B). Enrollment in each part of the study is expected to be complete approximately 3 months after the first subject has been enrolled in each part. Part A (dose-ranging and tolerability): In Part A (dose-ranging and tolerability), approximately 1,240 healthy 16- to 26-year-old women will be randomized in equal numbers to one of three 9-valent HPV L1 VLP vaccine formulations or the comparator GARDASIL in a three dose regimen (Day 1, Month 2, and Month 6). The primary goal in Part A is to identify the best 9-valent HPV L1 VLP vaccine dose for the V503 program which maintains the immunogenicity of GARDASIL against HPV types 6, 11, 16 and 18, while extending prophylactic vaccine coverage to include new HPV types 31, 33, 45, 52 and 58. A selected 9-valent HPV L1 VLP vaccine formulation should be generally well-tolerated, have HPV 6, 11, 16, and 18 GMTs that are comparable to the GARDASIL control, and be highly immunogenic for the new HPV vaccine types. Based on results for GARDASIL and other previously studied 3-dose (Day 1, Month 2, and Month 6) HPV L1 VLP vaccines, the ratio of the HPV 6, 11, 16, or 18 GMTs for the 9-valent HPV L1 VLP vaccine versus GARDASIL 1 month after the second dose of vaccine (post-dose 2) is highly indicative of the ratio after the completion of the 3-dose series. The 9-valent HPV L1 VLP vaccine is an important medical advance, offering the potential of significant prophylactic cancer coverage in addition to that already provided by GARDASIL. Therefore, in order to be able to make a decision about the immunogenicity of 9-valent HPV L1 VLP vaccine formulations in Part A and to move efficiently to evaluate a chosen formulation for efficacy in Part B of the study, the decision about which 9-valent HPV L1 VLP vaccine formulation to proceed with will occur after approximately 100% of the Part A post-dose 2 immunogenicity and adverse V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

25 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: experience data are available in the clinical database. The dose for Part B will be chosen by a senior management committee at the SPONSOR who is not directly involved with study conduct. This committee will choose the dose after reviewing immunogenicity summaries by vaccination group, and will relay the dose selection to the Data Safety Monitoring Board (DSMB). The DSMB will review the immunogenicity data and all tolerability data, and will communicate to the SPONSOR senior management committee if, in the opinion of the DSMB, there is concern with respect to the incidence of serious adverse experiences, severe injection-site adverse experiences, or severe systemic adverse experiences for the selected dose formulation as compared to the incidence observed in the GARDASIL group. If no tolerability concerns are raised, the senior management committee will inform the study team of the selected dose formulation and the study will proceed to Part B with the selected formulation and the control arm GARDASIL. If the DSMB expresses a concern about the tolerability profile of the selected dose, the senior management committee will select another dose and again request approval from the DSMB. Once the senior management committee and the DSMB agree on a chosen dose, the senior management committee at the SPONSOR will inform the study team of the 9-valent HPV L1 VLP vaccine formulation selected for Part B. The selected dose will be communicated to the study sites by letter before Part B vaccine supplies are shipped to the sites. In the unlikely event that the SPONSOR is unable to determine an acceptable 9-valent HPV L1 VLP vaccine formulation for use in Part B, the study may not continue to Part B. After a dose is selected, the unblinded statistician will generate a list of subjects who received the formulations that were not chosen for further evaluation in Part B. This list will be given to the study sites, and these subjects will complete their regimens with the same formulation they received in the first and second doses, and they will be followed for tolerability and immunogenicity through Month 7. Subjects who receive a 9-valent HPV L1 VLP dose with a lower total VLP amount than the formulation that is selected for use in Part B may be offered a single dose of GARDASIL. For the subjects in Part A who receive the selected 9-valent HPV L1 VLP vaccine dose or the comparator GARDASIL, the total duration of follow-up will be approximately 42 months from enrollment. Even though some subjects will stop their visits at Month 7 while others continue to Month 42, the integrity of the blinding will be maintained because no vaccination group information will be released to the study sites, and the subjects continuing beyond Month 7 will remain randomized between the two continuing vaccination groups. The primary immunogenicity analysis for all subjects included in Part A will be performed when all Part A post-dose three, Month 7 serology data are available in the clinical database. The purpose of the primary immunogenicity analysis in Part A is to provide confirmation of the dose chosen for continuation into Part B of the study, so the primary immunogenicity analysis will be performed for the selected 9- valent HPV L1 VLP vaccine formulation and the GARDASIL control. An important goal in Part A of the study is to evaluate the tolerability of the 9-valent formulations relative to that of GARDASIL. HPV L1 VLP vaccines formulated with Merck s adjuvant AAHS have been well-studied clinically, have been found to be generally well-tolerated, and are currently approved for use as a vaccine in the form of V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

26 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: GARDASIL. Because, however, this is the first time these specific formulations have been studied clinically, Merck & Co., Inc. personnel will perform an early and expedited review of blinded Vaccine Report Card (VRC) data once data for the first 40 subjects (approximately 10 subjects per arm) in Part A are available in the clinical database, and will request an unblinded review by the chairman of the Data Safety Monitoring Board (DSMB) if the rates of severe and/or serious adverse experiences are not consistent with the known adverse experience profile in the GARDASIL program. Subsequently, the DSMB, which consists of members that are independent of Merck Research Laboratories and a non-voting Merck & Co., Inc. statistician that is not associated with the study, will review VRC data when approximately 25% of Part A Day 1 vaccination visits are available in the study database, and will perform additional reviews at milestones specified in this protocol (See Section ). Part B (tolerability, immunogenicity, and efficacy): In Part B (tolerability, immunogenicity, and efficacy), approximately 2,480 additional healthy 16- to 26-year-old women will be randomized in equal numbers to the 9-valent HPV L1 VLP vaccine dose selected from Part A or the comparator GARDASIL. The primary immunogenicity analyses for Part B will be the formal comparison of the selected 9-valent formulation to the GARDASIL control and will include approximately 1,500 (the first 60% enrolled at each site) of the 2,480 additional subjects enrolled in Part B. Because the immunogenicity data for Part A will have been analyzed previously, the Part A immunogenicity data for the selected 9-valent HPV L1 VLP vaccine dose and the comparator GARDASIL will not be used in the Part B immunogenicity analysis. However, since neither the unblinded safety or efficacy data for Part A subjects will have been viewed previously by Merck personnel involved in the conduct of the study, the overall tolerability and efficacy analyses during Part B will combine the data from subjects enrolled in Part A (those who received the selected 9- valent HPV L1 VLP vaccine dose or the comparator GARDASIL ) and subjects enrolled in Part B. The primary efficacy analysis is case driven and will be conducted after 43 primary efficacy cases have been observed. For the subjects in Part B, the total duration of follow-up will be approximately 42 months from enrollment. Additional Study Design Considerations: In Part A or Part B, subjects will receive an allocation number from an allocation schedule generated by the Clinical Biostatistics department of the SPONSOR. There will be separate allocation schedules for Part A and Part B. The randomization will be balanced within sites. At the time of enrollment, subjects must meet all inclusion criteria and must not meet any exclusion criteria. Subjects should not be enrolled from settings in which the typical clinical population has a high risk for HPV infection at baseline, such as sexually-transmitted disease clinics. Eligible study sites include community health centers, academic health centers, and the offices of primary health care providers. The subjects, investigators (and his/her staff), laboratory staff, members of the Scientific Advisory Committee, and HPV Vaccine Program Pathology Panel will remain blinded to V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

27 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: subject vaccination group allocations for the duration of the study. The SPONSOR will remain blinded to subject vaccination allocations until the required number of cases of the primary efficacy endpoint have been observed and the database is unblinded for the primary efficacy analysis, with the exception of unblinded personnel who will provide data summaries for dose selection and DSMB meetings, and those who will determine when the required number of cases of the primary efficacy endpoint have been observed. These unblinded personnel will not be associated with the conduct of the study or the design of any of the statistical analyses for the study (other than those requested by the DSMB). For subjects who are enrolled in countries that have appropriate centralized registry infrastructures, results for Pap tests and biopsies taken outside of the context of the study will be obtained from the registries and certain data from the V database may be sent to these registries. Also, subjects enrolled in countries with these registries may be asked to participate in a sub-study for long-term follow-up. If this sub-study is performed, it will occur after a subject has completed her Protocol 001 scheduled study visits and will use cervical cancer screening registries to capture Pap test and biopsy results to assess the long-term duration of vaccine efficacy Vaccination Plan In Part A, approximately 1,240 healthy 16- to 26-year-old women will be randomized in equal numbers to one of 3 dose formulations of 9-valent HPV L1 VLP vaccine or the comparator GARDASIL (See Table 1-1, Section 1.6). In Part B, approximately 2,480 additional subjects will randomized in equal numbers to receive the 9-valent HPV L1 VLP vaccine selected in Part A or GARDASIL. At Day 1, Month 2, and Month 6, subjects will receive 9-valent HPV L1 VLP vaccine or GARDASIL as a series of three 0.5-mL intramuscular injections (see Section for injection procedures). Study vaccine must be stored in a refrigerator that has a temperature between 2 C and 8 C (35.6 F and 46.4 F). If the refrigerator in which the study vaccine is stored deviates from the 2 C and 8 C (35.6 F and 46.4 F) range, study vaccinations should be suspended and the SPONSOR should be contacted immediately so that an investigation can be conducted. For a sample of the label and for more information on packaging, labeling, and storage of clinical supplies, see Section LIST OF IMMUNOGENICITY AND EFFICACY MEASUREMENTS Immunogenicity Measurements Serum will be collected for analysis of anti-hpv 6, 11, 16, 18, 31, 33, 45, 52, and 58 responses by competitive Luminex Immunoassay (clia). Serum will be analyzed to support study objectives and to support assay development work. At Day 1, serum specimens will be collected before the first study vaccination to identify subjects who have been exposed to study vaccine-specific HPV types prior to enrollment. V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

28 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: After Day 1, serum specimens will be collected at visits specified in the Study Flow Chart. All subjects in Part A will have clia measurements summarized for dose selection and the Part A primary immunogenicity analysis, but only approximately 1,500 of the 2,480 subjects enrolled in Part B will be included in the Part B immunogenicity analysis. Specifically, the first 60% of subjects randomized at each site in Part B will be included in the Part B primary immunogenicity analyses. In addition to collection of serum for anti-hpv measurements to support the study objectives, an extra vial of serum will be collected at Month 7 for use in anti-hpv assay development and for use as a standard reference for anti-hpv assays. The assay development may include, but is not limited to, identifying potential biomarkers (especially a surrogate marker for protection), improving laboratory methods for future HPV antibody assays evaluating the isotypes or total anti-hpv immunoglobulin levels, or assessing the ability of HPV study vaccines to offer cross-protection against types of HPV that are not included in the vaccine. An experimental objective for the study is to characterize the levels of anti-hpv type 6/11 in both peripartum blood and in cord blood of infants born to subjects for evaluating the potential impact of 9-valent HPV L1 VLP vaccine on recurrent respiratory papillomatosis. Only some sites will be able to provide serum samples for this exploratory objective, because an investigator or sub-investigator associated with this study must provide complete obstetrical care for the subject, must deliver the infant, and approval must be obtained for collection of the samples at the facility where the infant is delivered Efficacy Measurements Labial/vulvar/perineal and perianal (LVPP) and endo/ectocervical (EEC) swabs will be tested for detection of HPV types 6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59 by Polymerase Chain Reaction (PCR) assay. PCR analysis of the swabs will be used to identify subjects who have an active HPV infection at enrollment and to determine persistent HPV infection endpoints. A subject with abnormal Pap Tests will be referred to colposcopy using a protocolmandated triage algorithm. The colposcopist will obtain cervical biopsies of areas of the most severe abnormalities and may choose to perform an endocervical curettage (ECC) if no cervical dysplasia is observed. A subject may need subsequent cervical definitive therapy per the protocol-mandated cervical definitive therapy guidelines. Vaginal lesions will be biopsied if they are observed during a Pap test, colposcopy, or at any time. A thorough external genital wart/lesion inspection will be performed at routine visit intervals. External genital warts/lesions noted after Day 1 will be biopsied. In addition, subjects with histologically confirmed HPV-related external genital warts (e.g., condyloma acuminata, VIN, cancer) or vaginal warts (e.g., condyloma acuminata, VaIN, cancer) will be referred to colposcopy if the biopsies were not obtained during a colposcopy. V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

29 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Tissue obtained from biopsy and cervical definitive therapy procedures will be analyzed by MRL HPV Thinsection PCR assay (detection of HPV types 6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59) and by a consensus diagnosis from the HPV Vaccine Program Pathology Panel to determine clinical disease efficacy endpoints. The Day 1 LVPP/EEC swabs and tissue specimens will include PCR analysis for nonvaccine HPV types (35, 39, 51, 56, 59, and potentially others). This data may be used for calculation of incidence rates for disease associated with non-vaccine HPV types and for cross-protection analyses. 2.6 LIST OF SAFETY MEASUREMENTS Each subject will receive a Vaccination Report Card (VRC) at the Day 1, Month 2, and Month 6 study vaccination visits. On the VRC, the subject will be asked to record her oral temperatures in the evening after each study vaccination and daily, at the same time of day, for 4 days after each study vaccination for the purpose of identifying febrile events. Also, beginning after each study vaccination and for a total of 15 days including the day of vaccination, the subject will be asked to record injection-site and systemic adverse experiences, concomitant medications, and concomitant vaccinations on the VRC. Serious adverse experiences, pregnancy information, and lactation information will be collected as described in Section DATA ANALYSIS SUMMARY Part A The dose formulation for Part B will be selected based on an interim analysis of the immunogenicity data from Part A that will be conducted when ~100% of the 310 enrolled subjects in each vaccination group have data available through Week 4 Postdose 2. Anti-HPV 6, 11, 16, 18, 31, 33, 45, 52, and 58 geometric mean titers (GMTs) and seropositivity percentages for each of the 9-valent HPV L1 VLP vaccine formulations and the GARDASIL comparator will be summarized. All available tolerability information will be summarized. The interim analysis will be conducted in a subset of subjects who 1) are seronegative at Day 1 to the relevant HPV type, without regard to PCR status at Day 1; 2) receive the first 2 doses of 9-valent HPV L1 VLP vaccine or GARDASIL in acceptable day ranges; and 3) have post-vaccination serum samples collected in acceptable day ranges. Group-level immunogenicity summaries will be provided by the unblinded statistician to a senior management committee at the SPONSOR who is unrelated to the trial. No serology or randomization data for individual study participants will be disclosed to the senior management committee. Unblinded summaries of all tolerability data will be provided to the DSMB for detailed safety review. The senior management committee and the DSMB will select the dose using the process outlined in Section The primary immunogenicity analysis in Part A will be conducted in type-specific per protocol immunogenicity populations, as defined in section The analysis will be performed by the unblinded statistician when all Week 4 Postdose 3 serology data are available from subjects enrolled in Part A. The primary hypothesis test will be V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

30 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: conducted to compare the 9-valent HPV L1 VLP vaccine formulation that is selected for use in Part B and GARDASIL. For each of HPV types 6, 11, 16 and 18, the primary immunogenicity hypothesis of noninferiority of the selected 9-valent HPV L1 VLP vaccine to GARDASIL with respect to GMTs at Week 4 Postdose 3 will be tested by constructing a confidence interval for the fold difference in GMTs between the groups (9- valent/gardasil ). In order to declare that the formulation of 9-valent HPV L1 VLP vaccine selected for use in Part B is noninferior to GARDASIL, noninferiority needs to be established for each of the HPV types 6, 11, 16, and 18. The primary immunogenicity hypothesis will be tested at the nominal 1-sided α level of to adjust for the informal Postdose 2 interim summary. Because Part A is considered a pilot study, the statistical criterion for noninferiority with respect to GMTs requires that the lower bound of the confidence interval for the fold difference in GMTs exclude 0.5 (that the confidence interval rules out a decrease of 2-fold or more). With approximately 310 subjects per group, Part A has an overall power of >99%. Part B - The primary efficacy analysis in Part B will be conducted in type-specific per protocol efficacy populations. Subjects enrolled in Part A in the selected 9-valent HPV L1 VLP vaccine arm or GARDASIL arm will also be included in the safety and efficacy analyses in Part B. There will be a single primary analysis of efficacy after at least 43 cases of the primary efficacy endpoint have been observed. A primary efficacy case is defined as a subject with an incident persistent HPV 31, 33, 45, 52, or 58 infection for a duration of 6 months (within ±1 month windows) or longer or HPV 31-, 33-, 45-, 52-, or 58-related cervical, vulvar, or vaginal cancer, AIS, CIN 2/3, VIN 2/3, VaIN 2/3, CIN 1, VIN 1, VaIN 1, or genital warts after 4 weeks postdose 3. The primary efficacy hypothesis will be addressed by constructing a 2-sided exact confidence interval for vaccine efficacy. The statistical criterion for success requires that the lower bound of the 95% confidence interval for vaccine efficacy exclude 20%. The primary efficacy hypothesis will be tested at the α=0.025 level (1-sided). Assuming a true vaccine efficacy relative to GARDASIL of 75% (accounting for the potential for GARDASIL to provide a ~40% reduction in the incidence of the primary endpoint due to cross-protection), the study will have at least 90% power to achieve success for the primary hypothesis. The primary immunogenicity analysis will be performed in type-specific per-protocol immunogenicity populations in Part B, as defined in section No subjects from Part A will be used in the immunogenicity analyses in Part B. The immunogenicity data from the first ~60% of the subjects enrolled in each site in Part B will be used in the primary immunogenicity analysis. The analysis will be performed at the time when the database is unblinded for the primary efficacy analysis. For each of HPV types 6, 11, 16 and 18, the primary immunogenicity hypothesis of noninferiority of 9-valent HPV L1 VLP vaccine to GARDASIL with respect to GMTs at 4 weeks postdose 3 will be tested by constructing a confidence interval for the ratio of GMTs between the groups (9- valent/gardasil ). The statistical criterion for noninferiority with respect to GMTs requires that the lower bound of the confidence interval for the GMT ratio exclude 0.67 (that the confidence interval rules out a decrease of 1.5-fold or more). The first primary immunogenicity hypothesis will be tested at the α=0.025 level (1-sided). V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

31 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: With approximately 750 subjects per vaccination group, this study will have >99% power for the primary immunogenicity hypothesis. Success in this study will be declared if the primary efficacy hypothesis and the primary immunogenicity hypothesis in Part B are achieved. The secondary efficacy analysis in Part B will be conducted in modified intention-to-treat population after at least 30 cases of the secondary endpoint of persistent HPV 31/33/45/52/58-related documented infection for 12 months or clinical disease have been observed. The secondary efficacy hypothesis will be addressed using the same methodology as for the primary hypothesis, and the statistical criterion for success requires that the lower bound of the 95% confidence interval for vaccine efficacy exclude 0%. Assuming a true vaccine efficacy relative to GARDASIL of 75%, the study will have at least 90% power to achieve success for the secondary hypothesis. The sample size for the efficacy evaluation is driven by the secondary efficacy hypothesis of persistent HPV 31/33/45/52/58-related infection for 12 months. To ensure at least 30 cases persistent HPV 31/33/45/52/58-related infection for 12 months or clinical disease are observed by Month 24, approximately 3,100 subjects (2,480 subjects enrolled for Part B combined with the ~620 subjects in the GARDASIL arm and selected dose formulation arm from Part A) will be required. It is anticipated that the secondary efficacy analysis will be conducted at the time of the primary efficacy analysis. If it appears that fewer than 30 cases of the secondary endpoint will be observed at the time of the primary efficacy analysis, the vaccine efficacy against the secondary endpoint will be summarized; however, the secondary hypothesis will not be tested until at least 30 cases are observed. The secondary immunogenicity hypothesis of noninferiority of 9-valent HPV L1 VLP vaccine to GARDASIL with respect to percentage of subjects who seroconvert to each of HPV 6, HPV 11, HPV 16, and HPV 18 by 4 weeks Postdose 3 will be tested at the α=0.025 level (1-sided) for each HPV type by constructing a 95% confidence interval for the difference in the percentage of subjects who seroconverted between the 9-valent HPV L1 VLP vaccine group and the GARDASIL group. The analysis will be conducted in type-specific per protocol immunogenicity populations, as defined in section Seroconversion is defined as changing serostatus from seronegative to seropositive. A subject with a clia titer at or above the serostatus cutoff for a given HPV type is considered seropositive for that type. The statistical criterion for noninferiority requires that the lower bound of the confidence interval for the difference in seroconversion percentages is greater than For HPV 31, 33, 45, 52 and 58, the secondary hypothesis of acceptability of the 9-valent HPV L1 VLP vaccine with respect to seroconversion rates at 4 weeks postdose 3 will be tested by constructing a confidence interval for the seroconversion rates for the 9-valent HPV L1 VLP vaccine group. The statistical criterion for acceptability requires that the lower bound of the confidence interval exclude 0.9 (90%). The secondary immunogenicity hypothesis will be tested at α= sided. V503_001-00_ProtCore VERSION 6.1 APPROVED 15-Jun-2007

32 V503, Protocol Issue Date: 15-Jun Product: V503 1 Protocol/Amendment No.: Redacted 3.1 RATIONALE 3. PROTOCOL DETAILS V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

33 V503, Protocol Issue Date: 15-Jun Product: V503 2 Protocol/Amendment No.: Redacted V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

34 Product: V503 3 Protocol/Amendment No.: Redacted V503, Protocol Issue Date: 15-Jun V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

35 Redacted V503, Protocol Issue Date: 15-Jun Product: V503 4 Protocol/Amendment No.: V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

36 Product: V503 5 Protocol/Amendment No.: Redacted V503, Protocol Issue Date: 15-Jun STUDY PROCEDURES Concomitant Medication(s)/Treatment(s) See the exclusion criteria for specific restrictions for prior and concomitant medications at Day 1. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

37 V503, Protocol Issue Date: 15-Jun Product: V503 6 Protocol/Amendment No.: Use of medicines and non-study vaccines should be documented in the data collection system in the following manner: Special Medications (corticosteroids, immunosuppressives, immune globulins, and blood products) from 3 days prior to Day 1 through Month 7; Other Medications from 3 days prior to each study vaccination through 14 days after each study vaccination; Non-Study Non-Replicating (Inactive) Vaccines for 14 days prior to each study vaccination through 14 days after each study vaccination; and Non-Study Replicating (Live) Vaccines for 21 days prior to each study vaccination through 14 days after each study vaccination. If possible, the subject should not receive Special Medications or Non-Study Vaccines within the time periods given above. Subjects may receive allergen desensitization therapy and tuberculin skin testing while participating in the study Prerequisites for Study Visits This section summarizes prerequisites for visits with pelvic specimen collection and visits with study vaccinations. Deviations from these prerequisites require consultation between the investigator and the SPONSOR and written documentation of the collaborative decision. See the Administrative Binder for a summary of deviations that require documentation in this study Prerequisites for Study Visits With Pelvic Specimen Collection See the inclusion/exclusion criteria for specific restrictions at Day 1. For visits that include collection of study specimens, study personnel should verify by questioning the subject and/or by examination that: 1. The subject has refrained from douching/vaginal cleansing and using vaginal medications or preparations for 2 calendar days prior to any visit that includes collection of labial/vulvar/perineal and perianal (LVPP) and endo/ectocervical (EEC) swabs, a Pap test and/or biopsies/definitive therapy specimens. 2. The subject has refrained from sexual activity (including anal, vaginal, or genital/genital contact whether same sex or opposite sex) for 2 calendar days prior to any visit that includes collection of LVPP/EEC swabs, a Pap test and/or biopsies/definitive therapy specimens. 3. The subject does not have visible blood in the vagina. As a guide, a study visit that includes collection of LVPP/EEC swabs, a Pap test and/or biopsies/definitive therapy specimens should not be scheduled between the time period of 2 days prior to menses V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

38 V503, Protocol Issue Date: 15-Jun Product: V503 7 Protocol/Amendment No.: (if date of menses can be predicted) to 2 days after menses. If, despite using these collection timeframe guidelines, visible blood is noted in the vagina, then the specimens may be collected. 4. The subject does not have gross purulent cervicitis at a study visit that includes collection of LVPP/EEC swabs, a Pap test and/or cervical biopsies/definitive therapy specimens. If the subject does not meet the requirements listed above, the study visit (including specimen collection and/or study vaccination) should be re-scheduled. If the subject has gross purulent cervicitis, the re-scheduled study visit should occur after the condition is diagnosed and treated according to the study site s standards and practices. Prerequisites for Study Vaccination Visits See the inclusion/exclusion criteria for specific restrictions at Day 1. For visits with study vaccinations, study personnel should verify by questioning the subject and/or by examination that: 1. The subject has not had a fever (defined as an oral temperature of 100 F or 37.8 C) within the 24-hour period prior to any study vaccination visit. 2. The subject has not received any systemic (oral or parenternal) corticosteroids in the 2 weeks prior to the Month 2 and Month 6 study vaccination visits. 3. The subject has not received a Non-Study Non-Replicating (Inactive) Vaccine within 14 days prior to any study vaccination visit or a Non-Study Replicating (Live) Vaccine within 21 days prior to any study vaccination visit. If the subject does not meet the requirements listed above, the study visit (including specimen collection and study vaccination) should be re-scheduled Summary of Scheduled Study Visit Procedures The Study Flow Chart summarizes procedures for scheduled study visits and items appear in the order they should be performed. This section provides clarifications to the scheduled study visit procedures (presented in order of the Study Flow Chart) Calculation of Scheduled Visit Windows The visit windows provided in this protocol are for site scheduling purposes. The visits windows used for exclusion from analyses will be provided in the Statistical Analysis Plan (SAP). The Day 1 visit is defined as the day that the first study vaccination is given. The Month 2 visit is based on Day 1 and has a 3 week visit window. The Month 3 visit has a visit window from 3 weeks after Month 2 through 5 weeks after Month 2. The Month 6 visit is based on Day 1 and has a 4 week visit window. The Month 7 visit has a visit window from 3 weeks after Month 6 through 7 weeks after Month 6. Post Month 7 V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

39 V503, Protocol Issue Date: 15-Jun Product: V503 8 Protocol/Amendment No.: visits are each based on Day 1 and have a 4 week visit window. To calculate visit windows, assume 1 month equals 30 days and 1 week equals 7 days Informed Consent and Assignment of Baseline Numbers Written consent will be obtained for each subject (or, for minors, from the parent/legal guardian and the subject) prior to enrollment, study data collection, or study procedures. A copy of the signed informed consent form will be given to each subject for her records. Verification of the subject s identity and age is to be determined prior to obtaining written consent. Any government-issued photo identification will suffice for verification purposes and should be documented in the subject s file. If the date of the Day 1 visit is later than the consent date, the interval between the date of the consent and the date of the Day 1 visit should be no more that 14 days apart. If this interval is 15 days or longer, then the subject must be re-consented. Baseline numbers will be supplied by the SPONSOR via the electronic data capture (EDC) system. Each subject will be assigned a unique baseline number immediately after the informed consent is signed. Baseline numbers will be assigned sequentially from the lowest number to the highest number at each site. Baseline numbers will not be skipped or reassigned for any reason. A single subject cannot be assigned more than 1 baseline number. During the Day 1 visit, one may discover a condition that will make the subject ineligible for the study. For example, exclusionary medical conditions mentioned in the inclusion/exclusion questions may be found during medical history review or during the pelvic examination. Thus, all pre-vaccination data must be collected and all study procedures must be completed before a subject is randomized (i.e. assigned an allocation number using Interactive Voice Response System [IVRS]) and vaccinated History & Medications At the Day 1 visit, the subject s lifetime gynecologic history and medical history for the year prior to Day 1 will be collected. After the Day 1 visit, any new gynecologic history and new medical history that have not been previously documented (either as adverse experiences [AEs] or as medical history conditions) will be collected. In addition to documentation of new medical history and AEs, concomitant and prior medications will be handled as described in Section Other documentation, such as demographics, sexual history, pregnancy history, and contraceptive use, will be collected in the data collection system, as discussed in the ecrf (Electronic Case Report Form) Entry Guidelines Pregnancy Testing and Serum Collection A serum pregnancy test or urine pregnancy test (sensitive to 25 miu/ml beta human chorionic gonadotropin [β-hcg]) will be performed at each Day 1 through Month 7 visit. The pregnancy test results will be obtained prior to each study vaccination on the day the subject is vaccinated. Any subject found to be pregnant at the Day 1 visit will not be V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

40 V503, Protocol Issue Date: 15-Jun Product: V503 9 Protocol/Amendment No.: randomized and will not participate in the study. For randomized subjects who become pregnant after receiving one or two study vaccinations, study visits and study vaccinations will be paused until resolution of the pregnancy (e.g., term, elective termination, spontaneous abortion). Study visits and study vaccinations in pregnant subjects will be handled as described in Table 3-1. Table 3-1 Guidelines for Pregnant Subjects: Managing Study Visits and Study Vaccinations Visit Where Pregnancy is Action Detected Day 1 Do not randomize subject Between Day 1 and Month 2 No scheduled visits until resolution of the pregnancy (e.g., term, elective termination, spontaneous abortion). The Month 2 study vaccination should be administered 4 weeks following resolution of pregnancy and after normalization of β- hcg levels. The Month 6 study vaccination should be administered 4 months after the Month 2 study vaccination. The Month 7 visit should be conducted 1 month after the Month 6 study vaccination. Between Month 2 and Month 6 No scheduled visits until resolution of the pregnancy (e.g., term, elective termination, spontaneous abortion). The Month 6 study vaccination should be administered 4 weeks following resolution of pregnancy and after normalization of β- hcg levels. The Month 7 visit should be conducted 1 month after the Month 6 study vaccination. After Month 6 Continue with scheduled study visits during the pregnancy. Completion of protocol-specified study procedures in pregnant subjects is at the investigator s discretion. The Month 7 visit should be conducted 1 month after the Month 6 study vaccination. After resolution of the pregnancy (e.g., term, elective termination, spontaneous abortion), the subject will undergo all protocolspecified procedures. The use of a cytobrush (Cytobrush ) for ThinPrep Pap test collection is contraindicated in women who are 10 weeks pregnant or more, according to the current manufacturer s label. The use of a broom (Wallach Papette ) for ThinPrep Pap test collection is not contraindicated in pregnancy, according to the current manufacturer s label. The use of a Wallach Papette instead of Cytobrush will be allowed if the investigator decides to perform a ThinPrep Pap test during pregnancy. Breastfeeding is not a contraindication to enrollment or to receiving study vaccinations. Pregnancy and breast-feeding in study subjects and infant serious adverse experiences (SAEs) must be reported as described in Section 3.4 and the Attachments. If allowed at the study site, an umbilical cord blood sample must be collected within 15 minutes of birth and a blood sample from the subject must be collected shortly before or at the time of delivery for anti-hpv testing (see Section ). V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

41 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Serum for anti-hpv measurements must be collected at Day 1, and at Months 3, 7, 12, 24, 36, and 42. For collection of serum samples, the study site must adhere to the procedures described in this protocol, must follow instructions provided by the SPONSOR/Central Laboratory, and must use the materials provided by the SPONSOR/Central Laboratory (see Section 3.2.5) Physical Examination & Vital Signs A complete physical examination will be conducted at Day 1 and Months 7, 18, 30, & 42. Physical examination details will be documented in the subject s chart and any medical conditions will be documented in the data collection system. Vital signs (sitting pulse, weight, blood pressure, respirations, and oral temperature) will be taken before each study vaccination. The pre-vaccination oral temperature will be documented in the data collection system and the other vital signs will be documented in the subject s chart. If the subject has a fever (defined as an oral temperature of 100 F or 37.8 C) within the 24-hour period prior to receiving a study vaccination, the subject should not receive study vaccine, and the vaccination visit should be rescheduled Gynecological Examination A complete gynecological examination will be performed at Day 1 and Months 7, 18, 30, & 42. A sterile, non-lubricated, single use, individually wrapped, plastic speculum should be used. If lubrication of the speculum is needed, only use sterile saline or sterile water. All study samples (See Section ) must be taken from the pelvic area before the bimanual pelvic examination is performed to prevent contamination of study samples. If the subject has gross purulent cervicitis at a visit with specimen collection (LVPP/EEC swabs, a Pap test, and/or cervical biopsy/definitive therapy specimens), the visit should be re-scheduled for a time after the condition is diagnosed and treated according to the study site s standards and practices. Gynecologic conditions other than gross purulent cervicitis, such as vaginitis, bacterial vaginosis, and vaginal yeast infections, do not require a visit to be re-scheduled. These conditions should be diagnosed and treated according to the study site s standards and practices and treatment should be given after study sample collection and/or after study vaccination Genital/Cervical Swabs, Pap Test, and Sexually Transmitted Infection (STI) Testing Two (2) labial/vulvar/perineal and perianal (LVPP) swabs for HPV Polymerase Chain Reaction (PCR) assay, 1 endo/ectocervical (EEC) swab for PCR assay, and a ThinPrep Pap test for liquid-based cytology Papanicolaou testing will be collected at Day 1 and Months 7, 12, 18, 24, 30, 36 & 42. At Day 1 and Months 7, 18, 30, & 42, tests for Chlamydia and gonorrhea are performed on the Pap test fluid. For collection of swabs and the ThinPrep Pap test, the study site must adhere to the procedures described in this protocol, must follow instructions provided by the SPONSOR/Central Laboratory, V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

42 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: and must use the materials provided by the SPONSOR/Central Laboratory (see Section 3.2.5). A sterile, non-lubricated, single use, individually wrapped, plastic speculum should be used for pelvic specimen collection. If lubrication of the speculum is needed, only use sterile saline or sterile water. During the Pap test, an inspection for vaginal lesions should be performed. Per the exclusion criteria, a clinical impression of HPV-related vaginal lesions (e.g., condyloma acuminata or VaIN) will exclude a subject at Day 1. However, if these types of vaginal lesions are identified after Day 1, the lesions should be classified and managed as described in Section ThinPrep Pap tests will be submitted to the SPONSOR-designated Central Laboratory. Cytology specimens will be evaluated using The Bethesda System For a diagnosis of atypical squamous cells of undetermined significance (ASC-US), the Central Laboratory will perform reflex testing for high-risk/low-risk HPV probes (Digene Hybrid Capture II Assay) on residual ThinPrep material. Site staff, subjects, and MRL personnel will remain blinded to the identity of the positive probe. The Pap test diagnoses generated by the Central Laboratory will be reported to the study sites for subject management, and subjects will be referred to colposcopy according to a protocolmandated triage algorithm (see Table 3-2). All Pap test reports must be reviewed and signed by a physician (M.D./D.O.) investigator/sub-investigator. At Day 1 and Months 7, 18, 30, & 42, tests for Chlamydia and gonorrhea are performed on the Pap test fluid and processed through the SPONSOR-designated Central Laboratory, but additional local laboratory tests for Chlamydia and gonorrhea may be performed at other visits at the discretion of the investigator if clinically indicated. For these additional local laboratory tests for Chlamydia and gonorrhea, the study site will collect samples using their own materials. Tests for other sexually transmitted infections (STIs), including HSV, syphilis, Hepatitis B, and HIV, may be performed at any visit at the discretion of the investigator if clinically indicated. However, per the exclusion criteria for this study, known HIV-positive subjects should not be enrolled. Any STI sampling of the cervix or external genital region should occur after the swabs and Pap test are obtained. Subjects who are enrolled into the study and test positive for STIs may remain in the study. All subjects who test positive for STIs should be referred for counseling and treatment. It is important to document all tests for STIs on the appropriate ecrf as discussed in the ecrf Entry Guidelines External Genital Lesion Examination After obtaining all study specimens (swabs and Pap test), and after a change of gloves, an external genital examination should be performed using a good light source and a handheld magnifying glass (4x to 5x power) or, optionally, a colposcope with low-power magnification. The following guidance should be used for all external genital lesion examinations performed within the study: 1. The examiner should ask the subject if she has noticed any bumps, lesions, or unusual symptoms (e.g., itching, discomfort, dyspareunia, dysuria). V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

43 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: The labia should be spread, and the condition of the hymen and vulvovaginal skin and clitoris are to be examined and noted. Note abnormalities, such as abnormal hair growth distributions, abnormalities of the skin, rashes, lacerations, or bruises. 3. The examination should include a systematic inspection of the entire external genitalia (periurethral, perineal, perianal, and vulvar regions) for apparent cutaneous and subcutaneous lesions or wart-like growths. 4. Acetic acid is not to be used routinely. It may be used for confirmation of a suspected lesion. 5. All observations should be recorded into the subject s chart. Medical conditions and lesions should be recorded on the appropriate ecrfs. Per the exclusion criteria, a clinical impression of any HPV-related external genital lesions (e.g., condyloma acuminata or VIN) will exclude a subject at Day 1. However, if these types of external genital lesions are identified after Day 1, the lesions should be classified and managed as described in Section IVRS Assignment of Allocation Numbers and Vaccination Vials The Interactive Voice Response System (IVRS) will assign the subject an allocation number and then subsequently assign a unique vial identification number for the vial of clinical material the subject should receive at that visit. The IVRS will automatically assign the appropriate clinical material based on the subject s treatment allocation. Randomization will be balanced within each study site. The study personnel will access IVRS at each subsequent visit when administration of vaccine is to occur for assignment of a unique vial identification number for the clinical material to be administered to the subject. A single patient/subject cannot be assigned more than 1 allocation number. Randomization deviations (e.g., a subject is assigned two allocation numbers by mistake through IVRS) require written documentation. See the Administrative Binder for a summary of deviations that require documentation in this study Study Vaccine Administration Preparation for Administration The study vaccine must be used as supplied (no dilution before administration). Prior to administration, mix the contents of the vial thoroughly by rolling the vial between the palms of both hands for 30 seconds. Withdraw a 0.5-mL dose from the vial, which contains approximately 0.75 ml of study vaccine. The study vaccine should be a whitish, semi-translucent suspension when thoroughly mixed. If the appearance is otherwise, do not administer, and contact the SPONSOR immediately V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

44 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Study Vaccine Administration Study vaccine will be administered at Day 1, Month 2, and Month 6. At each visit, subjects will receive 9-valent HPV L1 VLP vaccine or GARDASIL as a 0.5-mL intramuscular injection. The deltoid muscle of the nondominant arm is the preferred site of vaccination. Study vaccinations should not be administered into the buttocks area. If the subject has received injectable contraceptives (e.g., DEPO-PROVERA TM ) or implanted contraceptives (e.g., NORPLANT ) in both arms in the past 9 months, the injection may be given in the thigh rather than in an arm. If the subject has an implanted contraceptive (e.g., NORPLANT ) at the time of study vaccination, avoid administering the study vaccination in the arm with the implanted contraceptive. Injections should not be given within 2 cm of a tattoo, scar, or skin deformity. Study vaccine should be administered using a 1.0-mL syringe. Injections should be administered at a 90 angle into the muscle tissue using a needle long enough to ensure intramuscular deposition of vaccine. For injection into the arm muscle, use the following needle length and gauge specifications: 1-inch needle, 22 to 23 gauge, for women weighing <200 pounds (<90.9 kg.) or 1 ½-inch needle, 22 to 23 gauge for women weighing 200 pounds ( 90.9 kg). For injection into the thigh muscle, a 1 ½-inch needle, 22 to 23 gauge, should be used Clinical Follow-Up Please refer to Section 3.4 for Vaccine Report Card (VRC) information and information on AE reporting Summary of Unscheduled Study Visit Procedures Colposcopy, Cervical Biopsy, and Cervical Definitive Therapy Subjects with abnormal ThinPrep Pap tests will be referred to colposcopy and biopsy according to the protocol-mandated triage algorithm. In addition, subjects with histologically confirmed HPV-related external genital warts (e.g., condyloma acuminata, VIN, cancer) or vaginal warts (e.g., condyloma acuminata, VaIN, cancer) will be referred to colposcopy if the biopsies were not obtained during a colposcopy. If a colposcopy is required per the protocol-mandated triage algorithm, the colposcopy must be performed within 2 months of receipt of the abnormal result by the study investigator. If abnormalities of the cervix are found during colposcopy, cervical biopsies will be taken of the areas with the most severe abnormalities. If no cervical dysplasia is observed, the colposcopist may choose to perform an endocervical curettage (ECC). A vaginal exam is required during all study colposcopies. Additional Pap testing during the unscheduled colposcopy study visit is not allowed and is considered a deviation from the protocolmandated triage algorithm. A summary of colposcopy referral is given in Table 3-2. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

45 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Table 3-2 Protocol-Mandated Triage Algorithm for Referral to Colposcopy ThinPrep Pap Result Negative for intraepithelial lesion or malignancy (includes reactive, reparative, inflammatory, etc.) Atypical Squamous Cells Undetermined Significance (ASC-US) Atypical Squamous Cells cannot exclude HSIL (ASC-H) Low-grade Squamous Intraepithelial Lesion (LSIL) High-grade Squamous Intraepithelial Lesion (HSIL) Atypical Glandular Cells (to include atypical endocervical, endometrial, NOS; Adenocarcinoma in Situ, adenocarcinoma, etc.) Unsatisfactory Action Routine visit interval as specified by the protocol. Central laboratory performs reflex HPV testing on residual ThinPrep material (High Risk and Low Risk Probe, Hybrid Capture II, DIGENE ). If at least 1 probe is positive or if no result is obtained, the subject will be referred for colposcopy. If both probes are negative, the subject will return for Pap screening at the routine visit interval. Referral to colposcopy. Referral to colposcopy. Referral to colposcopy. Referral to colposcopy. Repeat ThinPrep Pap test as soon as possible, but not less than 4 weeks from the Pap test that had the unsatisfactory finding. Action External Genital Biopsy Result HPV-related (e.g., condyloma acuminata, VIN, Referral to colposcopy if the biopsy was not cancer) obtained at a colposcopy. Vaginal Biopsy Result Action HPV-related (e.g., condyloma acuminata, VaIN, Referral to colposcopy if the biopsy was not cancer) obtained at a colposcopy. Notes: Additional guidance for the protocol-mandated triage algorithm, including follow-up for discordant cases, is given in the Mandatory Regimen for Triage Attachment. Deviations to the protocol-mandated triage algorithm for colposcopy, including repeat Pap tests at unscheduled colposcopy study visits, require consultation between the investigator and the SPONSOR and written documentation of the collaborative decision. Deviations that require this documentation include a colposcopy that is performed for another medical reason, such as unresolved post-coital bleeding that persists after completion of the appropriate work-up. In addition, if a deviation is discovered by the SPONSOR or study investigator after an error is made, retrospective written documentation is required. See the Administrative Binder for a summary of deviations that require documentation in this study. If the Pap test diagnosis is AGC (to include AIS, atypical endocervical cells, adenocarcinoma, etc.), ASC-H, or HSIL and the subsequent biopsy diagnosis is CIN 1 or less, then this is defined as a discordant case. A discordant diagnosis will require a review of the results by the Sponsor-designated Central Laboratory. Upon completion of the review, a consultation report will be sent to the investigator. If the results are V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

46 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: amended upon completion of this review, the protocol-mandated colposcopy triage algorithm (Table 3-2) and guidelines for cervical definitive therapy (Table 3-3) should be followed. If the discordant results are reviewed but not amended, see Table 3-3 and the Mandatory Regimen for Triage Attachment (Figure 4) for guidelines. Subjects will be referred for cervical definitive therapy according to Table 3-3. If cervical definitive therapy is performed, on the day of the procedure, pre-definitive cervical biopsies must be taken of the areas with the most severe abnormalities prior to performing the cervical definitive therapy. Loop Electrosurgical Excision Procedure (LEEP) is the preferred method for cervical definitive therapy. LEEP has contraindications for subjects with the following conditions: allergy to all local anesthetics; pregnancy; severe acute cervicitis (severe acute cervicitis should be diagnosed and treated, and excision conducted after the infection has resolved); obvious invasive cancer; significant glandular neoplasia (AGUS favor neoplasia, AIS, Adenocarcinoma); and microinvasive cancer. In addition to LEEP, laser conization is also acceptable if it is the standard of care at the study site. Cold-knife conization and ablative cervical definitive therapy (e.g., cervical cryotherapy or cervical laser vaporization) should be reserved for rare instances where cervical definitive therapy is required and LEEP or laser conization is not indicated. LEEP, laser conization, and coldknife conization are study procedures, but ablative cervical definitive therapy is not a study procedure because this procedure is not tissue preserving. Table 3-3 Protocol-Mandated Guidance for Referral to Cervical Definitive Therapy Biopsy Results that Require Cervical Definitive Therapy Cervical biopsy result or ECC result of CIN 2, CIN 3, Adenocarcinoma in Situ (AIS), or cervical cancer. Cervical biopsy (including ECC, if taken) result of CIN 1 on at least 2 consecutive biopsies obtained over a period of 18 months or longer. Discordant Case Guidance for Cervical Definitive Therapy (After Discordant Case Review Confirms Discrepancy) If colposcopic examination was unsatisfactory- Cervical Definitive Therapy required. High Grade Pap Result (AGC [to include AIS, atypical endocervical cells, adenocarcinoma, etc.)], ASC-H, or HSIL), negative vaginal examination, cervical biopsy/ecc is CIN 1 (or less) or if no lesion is seen and cervical biopsy/ecc is not taken: If colposcopic examination was satisfactory- See Attachment (Mandatory Regimen for Triage, Figure 4) for guidance. Two repeat High Grade Pap diagnoses (regardless of whether they are consecutive or not) without a confirmed cervical biopsy or ECC diagnosis of CIN 2, CIN 3, or cervical cancer will be referred to definitive therapy. General Notes: Pre-definitive cervical biopsies must be obtained on the day of the cervical definitive therapy procedure and must be taken before the cervical definitive therapy procedure is performed. Ablative cervical definitive therapy is not a study procedure. Deviations to the protocol-mandated guidance for referral to cervical definitive therapy and the footnotes in this table require consultation between the investigator and the SPONSOR and written documentation of the collaborative decision. In addition, if a deviation is discovered by the SPONSOR or study investigator after an error is made, retrospective written documentation is required. See the Administrative Binder for a summary of deviations that require documentation in this study. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

47 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: If cervical biopsies, pre-definitive cervical biopsies, endocervical curettage (ECC) specimens, or cervical definitive therapy specimens are obtained, they must be sent to the SPONSOR-designated Central Laboratory for analysis. Separate biopsy forceps must be used for each of the discrete lesions that are biopsied. See Section for detailed guidance on colposcopy, biopsy, ECC, and cervical definitive therapy procedures. Subjects who undergo cervical biopsy and/or cervical definitive therapy will continue to be followed through the completion of scheduled study visits. The interval between a cervical biopsy and/or cervical definitive therapy and the next scheduled visit that includes a pelvic exam must be at least 2 months. Follow-up care after cervical definitive therapy will occur according to the study site s standards and practices. Colposcopy, biopsy, and ECC procedures must be performed by an experienced health care professional (>50 colposcopies per year for at least 2 years). The cervical definitive therapy provider should be an experienced physician (defined as performing 20 LEEP procedures per year for at least 2 years) and must have documented formal instruction [i.e., residency or postgraduate course or American Society for Colposcopy and Cervical Pathology (ASCCP) training course]. To ensure that colposcopy practices and skill levels are standardized across study sites, and that colposcopists at all sites develop a uniform approach to all study colposcopies, study colposcopists will be required to participate in colposcopic standardization training. The purpose of this training is to evaluate the colposcopic practices at all study sites, to standardize colposcopy and biopsy practices, and to provide a guide for all aspects of the protocol related to colposcopy and histological sampling of the cervix. An alternative to the SPONSOR-organized colposcopy training, the Comprehensive Colposcopy Course given by the American Society for Colposcopy and Cervical Pathology (ASCCP), will be accepted as equivalent for the purpose of standardization to participate in this study, provided it has been taken within the year prior to study start External Genital Lesion and Vaginal Lesion Diagnosis and Follow-Up External genital lesions or vaginal lesions may be identified during a scheduled visit exam, Pap test, colposcopic examination, or at any other time during the study. For the purposes of this study, lesions that require access via the introitus (with the exception of the cervix, which is considered a separate location) are to be defined as vaginal lesions. Lesions that are external to the introitus are to be defined as external genital lesions. To classify the location of external genital lesions, the external genital area will be further divided into the following 4 regions: periurethral (includes the clitoris) perineal perianal vulvar (includes all areas other than the periurethral, perineal or perianal regions) V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

48 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: It is important to make an accurate evaluation of any identified external genital lesions. The practitioner must classify a lesion as condyloma acuminata, other HPV-related lesion (e.g., a flat wart, a papular wart, a keratotic wart, VIN, Bowenoid papulosis, Bowen s disease) or a non HPV-related lesion. If lesions are identified, they must be documented on the appropriate ecrf. External genital lesions suspected to be possibly, probably, or definitely HPV-related (i.e., clinical impression of condyloma acuminata or other HPV-related lesion) or of unknown etiology are to be biopsied. Separate biopsies should be performed if external genital lesions are at different locations (periurethral, perineal, perianal, or vulvar region) or if more than one external genital lesion is at the same location and has a different morphology. An additional external genital lesion biopsy should be obtained if a new HPV-related lesion (or lesion of unknown etiology) appears and is of differing morphology and/or differing location from the initial lesion(s). A recurrent lesion should not be biopsied. A recurrent external genital lesion is defined as a lesion that has both of the following characteristics: (1) appears within 2 months of the complete resolution of an initial lesion(s) and (2) has the same location and a similar morphology as the initial lesion(s). If an HPV-related lesion or a lesion of unknown etiology appears after 2 months of complete resolution of the initial lesion(s), a biopsy of the lesion should be taken. Vaginal lesions suspected to be possibly, probably, or definitely HPV-related (i.e., condyloma acuminata or other HPV-related lesion) or of unknown etiology are to be biopsied. Separate biopsies should be performed for each vaginal lesion with a different morphology. An additional vaginal biopsy should be obtained if a new potentially or definitely HPV-related lesion (or lesion of unknown etiology) appears and is of differing morphology from the initial lesion(s). A recurrent lesion should not be biopsied. A recurrent vaginal lesion is defined as a lesion that has both of the following characteristics: (1) appears within 2 months of the complete resolution of an initial lesion(s) and (2) has a similar morphology as the initial lesion(s). If an HPV-related lesion or a lesion of unknown etiology appears after 2 months of complete resolution of the initial lesion(s), a biopsy of the lesion should be taken. External genital lesion biopsies and vaginal lesion biopsies must be sent to the SPONSOR-designated Central Laboratory for analysis. Separate biopsy forceps must be used for each of the discrete lesions that are biopsied. Biopsy of an external genital lesion or vaginal lesion should ideally be performed on the same day that the lesion is first observed, because the lesion may resolve before the subject returns for a biopsy procedure if it is performed at a later time. See Section for detailed guidance on these procedures. Subjects with histologically confirmed HPV-related external genital warts (e.g., condyloma acuminata, VIN, cancer) or vaginal warts (e.g., condyloma acuminata, VaIN, cancer) will be referred to colposcopy if the external genital or vaginal biopsies were not obtained during a colposcopy (see Section 3.2.5). Subjects who undergo external genital lesion biopsy, vaginal lesion biopsy, and any subsequent treatment for the lesions will V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

49 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: continue to be followed through the completion of scheduled study visits. After a biopsy is obtained, management and treatment of external genital lesions and vaginal lesions are at the investigator s discretion and according to the study site s standards and practices. If excision is chosen as the method of treatment, all excised tissue is to be submitted to the SPONSOR-designated Central Laboratory for analysis. The interval between the external genital lesion biopsy/excision or vaginal lesion biopsy/excision and the next scheduled visit involving a pelvic exam must be at least 2 months. External genital lesion biopsy/excision and vaginal lesion biopsy/excision must be performed by an experienced health care professional Processing of Tissue Specimens In the Context of the Study Extra visits that are associated with tissue excision (cervical, vaginal, external genital, and cervical definitive therapy samples) will be considered part of the study. Any specimens collected will be obtained as described in the protocol and collected and shipped as described in the Laboratory Manual provided by the SPONSOR/Central Laboratory. Slides of tissue specimens will be prepared at the SPONSOR-designated Central Laboratory and reviewed by a pathologist, and reports will be provided to the study sites for subject management. All tissue slides will be reviewed by the HPV Vaccine Program s expert Pathology Panel, which will consist of up to 5 pathologists. The consensus diagnosis of HPV Vaccine Program Pathology Panel will represent the final diagnosis for study purposes and not for subject management. Also, all tissue will be tested at MRL for detection of HPV types by PCR assay (See Section 3.3.2). The SPONSOR and the HPV Vaccine Program Pathology Panel will follow established guidelines for review of the slides. The investigator will be notified of any histological diagnosis made by the HPV Vaccine Program Pathology Panel that is more severe than the one made by the SPONSOR-designated Central Laboratory Pap Tests and Tissue Specimens Taken Outside the Context of the Study Cervical, vaginal, and external genital tissue samples and Pap tests collected outside the context of the study are strongly discouraged. For procedures with sample collection, outside the context of the study is defined as processing of samples at a local laboratory rather than through the SPONSOR-designated Central Laboratory. For procedures without sample collection, such as a colposcopy without biopsy, outside the context of the study means a procedure performed by a non-study clinician. If a subject undergoes a study procedure outside the context of the study, all efforts will be made to obtain and send the following to the SPONSOR: (1) Pap or colposcopy/operative report; (2) diagnostic slides (for tissue biopsies only); (3) local pathology report; and (4) tissue block for diagnostic slide preparation and HPV analysis Procedures for Collection and Handling of Study Specimens For scheduled study visits, consult the Study Flow Chart for the specific samples needed for each visit and the order of sample collection. The following Sections describe the V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

50 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: step-by-step procedures for collection of study specimens, a description of supplies needed, and the guidelines for handling specimens. Samples should be shipped, labeled, and handled as instructed by the SPONSOR/Central Laboratory. Specimen collection supplies provided by the SPONSOR/Central Laboratory must be used by the site without substitution. Within 30 minutes of collection, place serum and LVPP/EEC swabs in a freezer at -20 C (or lower) until the samples are shipped on dry ice. If the samples thaw, contact the SPONSOR. Thawed serum/swab samples require written documentation, including details such as allocation number, visit interval, date of collection, and sample accession numbers (see the Administrative Binder for a summary of deviations that require documentation in this study). For collection of pelvic specimens (EEC swab, Pap test, cervical biopsy/definitive therapy), a sterile, non-lubricated, single use, individually wrapped, plastic speculum should be used. If lubrication of the speculum is needed, only use sterile saline or sterile water may be used Serum or Urine Specimen for Pregnancy Test The serum pregnancy test or urine pregnancy test (sensitive to 25 miu/ml β-hcg) will be performed per the manufacturer s instructions. All materials used for serum and urine pregnancy testing will provided by the study sites Serum for Anti-HPV Measurements at Scheduled Visits For each visit that requires a serum specimen for anti-hpv measurements, a 10-mL (nonheparinized, non-serum separator, red-top tube provided by the SPONSOR-designated Central Laboratory) blood specimen will be collected and should be separated to avoid hemolysis. A minimum of 3.0 ml of serum should be aliquoted to a vial provided by the SPONSOR-designated Central Laboratory. An additional 1.5 ml of serum ( Retention Serum ) should be aliquoted to a vial provided by the SPONSOR-designated Central Laboratory. At Month 7, an additional 10-mL (non-heparinized, non-serum separator, red-top tube provided by the SPONSOR-designated Central Laboratory) blood specimen will be collected and should be separated to avoid hemolysis. The serum ( Reference Serum, approximately 4.5 ml) should be transferred to a vial provided by the SPONSORdesignated Central Laboratory. Serum and Reference Serum vials will be stored at the site at -20 C (or lower) until shipped on dry ice. Retention Serum serum vials will remain stored at the site at -20 C (or lower) until the SPONSOR notifies the study site to ship the vials. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

51 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Cord Serum and Maternal Serum for Anti-HPV Measurements (Applicable to Participating Study Sites) The collection of cord blood should not affect the care of the mother or child, and there should be no significant deviation from normal procedures. Blood from the mother should be collected by venipuncture shortly before or at the time of delivery. Cord blood should be collected according to local standard procedures. Collections of cord blood are generally more successful if initiated within 15 minutes of the birth of the infant. The placenta need not be delivered in order to collect cord blood. Clean the site of the cord that will be punctured with antiseptic swabs. The needle of a syringe can be inserted into the cleansed area of the umbilical vein. For the cord blood or peripartum maternal blood, a 10-mL (non-heparinized, non-serum separator, red-top tube provided by the SPONSOR-designated Central Laboratory) blood specimen will be collected and serum should be separated to avoid hemolysis. A minimum of 3.0 ml of cord serum or peripartum maternal serum should be aliquoted to a vial provided by the SPONSOR-designated Central Laboratory. An additional 1.5 ml ("Retention Serum" vials) of cord serum or peripartum maternal serum should be aliquoted to a vial provided by the SPONSOR-designated Central Laboratory. Serum vials will be stored at the site at -20 C (or lower) until shipped on dry ice. The Retention Serum serum vials will remain stored at the site at -20 C (or lower) until the SPONSOR notifies the study site to ship the vials Labial/Vulvar/Perineal and Perianal (LVPP) Swabs for HPV PCR Use DACRON swabs and Specimen Transport Medium (STM) vials supplied by the SPONSOR-designated Central Laboratory. When collecting swab specimens, remember to keep all swab tips and swab shafts (i.e. the portion that will be placed in the STM) untouched. 1. Remove DACRON swabs from packaging. 2. Using the first swab, swab back and forth in a tight zigzag motion from the clitoral prepuce down to the posterior fourchette on first one side and then the other so two parallel zigzag paths down the perineum will allow collection between the folds of the labia minora and majora. 3. With the second swab, swab the perianal area. 4. Place both swabs in the appropriately labeled collection/transport tube containing STM. Break off the end of the shafts protruding from the tube by bending them sharply against the rim of the tube. The shafts are pre-scored to facilitate breakage. 5. Securely cap the collection/transport tube containing the specimen. 6. Ensure proper labeling, store the sample at -20 C (or lower), and ship the sample as specified by the SPONSOR/Central Laboratory. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

52 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Endo/Ectocervical (EEC) Swab for HPV PCR Use DACRON swabs and Specimen Transport Medium (STM) vials supplied by the SPONSOR-designated Central Laboratory. When collecting swab specimens, remember to keep all swab tips and swab shafts (i.e. the portion that will be placed in the STM) untouched. 1. Remove one DACRON swab from its packaging. 2. Introduce the swab into the cervical os with enough pressure to maintain contact with the epithelium but not to induce bleeding. Twirl the swab only 1 to 2 times, and then use a back-and-forth swiping motion across the ectocervix from anterior to posterior cervical lip (top to bottom). 3. Place the swab into the collection/transport tube containing the STM. Break off the end of the shaft protruding from the tube by bending it sharply against the rim of the tube. The shaft is pre-scored to facilitate breakage. 4. Securely cap the collection/transport tube containing the specimen. 7. Ensure proper labeling, store the sample at -20 C (or lower), and ship the sample as specified by the SPONSOR/Central Laboratory Pap Test (ThinPrep ) Specimen Collection Use PreservCyt supplied by the SPONSOR-designated Central Laboratory. 1. Using a plastic spatula, scrape the ectocervix in a 360 arc for 2 full rotations, being sure to include the squamocolumnar junction in the portion sampled. Place the spatula into the open PreservCyt vial, using vigorous shaking and swishing of the spatula to rinse all of the cellular debris from the spatula. It is acceptable to leave the spatula in the open PreservCyt vial until the cytobrush collection is completed. 2. Insert a cytobrush 1 into the cervical os, and slowly rotate 1/2 turn one direction. DO NOT OVER-ROTATE. Do not insert deeper than the length of the brush head. Place the cytobrush immediately into the PreservCyt vial. 3. Use the spatula to rub against the cytobrush in order to dislodge as much cellular debris from the cytobrush and spatula as possible. Swish and rinse the cytobrush, and knock it against the side of the PreservCyt vial at least 10 times to release enough epithelial cells for an adequate Pap test sample. In order to minimize adherence of 1 The use of a cytobrush (Cytobrush ) for ThinPrep Pap test collection is contraindicated in women who are 10 weeks pregnant or more, according to the current manufacturer s label. The use of a broom (Wallach Papette ) for ThinPrep Pap test collection is not contraindicated in pregnancy, according to the current manufacturer s label. The use of a Wallach Papette instead of Cytobrush will be allowed if the investigator decides to perform a ThinPrep Pap test during pregnancy. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

53 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: cellular debris to the collection devices, the rinsing and swishing procedures should be performed within 2 minutes of collection. 4. Remove the spatula and the cytobrush from the PreservCyt vial and twist the cap on the vial securely. 5. Ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. Store PreservCyt vials containing the specimens at room temperature until they are shipped to the Central Laboratory. According the manufacturer s instructions, ThinPrep specimens must be processed within 21 days of collection to preserve sample integrity. To meet this requirement, specimens must be shipped according to instructions and time intervals specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. 6. Prior to removal of the speculum, inspect the lateral walls of the vagina for vaginal lesions. Then collapse the speculum and rotate 90 degrees and re-open the blades to inspect the anterior and posterior walls of the vagina for vaginal lesions. If vaginal lesions are present, use the guidance given in the protocol to classify vaginal lesions and to obtain required biopsies (see Sections and ). Collapse and remove the speculum. 7. Change gloves and perform a systematic inspection of the entire external genitalia using a good light source and a hand-held magnifying glass (4x to 5x power) or, optionally, a colposcope with low-power magnification. If external genital lesions are present, use the guidance given in the protocol to classify external genital lesions and to obtain required biopsies (see Sections and ) External Genital Lesion Biopsy and Vaginal Lesion Biopsy Use the appropriate ecrfs to note locations, clinical impressions, and biopsy numbers during the procedure. Use the specimen collection kits (labels/requisition) and formalin fixative container provided by the SPONSOR-designated Central Laboratory. The external genital biopsy kit should be used for external genital biopsies and the vaginal biopsy kit should be used for vaginal biopsies. 1. Cleanse the biopsy area with antiseptic solution. Using a 30-gauge needle and a syringe of 1 to 2 ml of 1% lidocaine or bacteriostatic saline, infiltrate below the epidermis of the lesion. 2. Elevate the lesion and remove tangentially with fine (iris) scissors or a scalpel blade to obtain the specimen. 3. Place the biopsy piece in the fixative container and ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

54 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: If multiple biopsies are obtained, each specimen must be placed in a separate container of fixative. To prevent cross-contamination, a different set of instruments is to be used for each biopsy taken. A maximum of 6 external genital lesion biopsies and 3 vaginal biopsies should be obtained. 5. If the biopsy procedure does not remove the entire lesion(s) and surgical excision is chosen as a treatment method, the excised tissue is to be submitted to the Central Laboratory in a separate container of fixative and labeled as an additional biopsy. 6. To promote hemostasis after the procedure, apply gentle pressure. Monsel s solution may be used. For larger areas, a single interrupted suture may be used. Electrocauterization is to be avoided, but the decision is left to the discretion of the practitioner. 7. Advise the patient to keep the area clean and to expect spotting for 3 days with healing by 1 week. 8. On the day of collection, ship the sample at room temperature as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory Colposcopy Guidelines 1. Assist patient to dorsal lithotomy position. 2. Insert speculum and visualize the entire cervix through the colposcope at low power. 3. Use optional visualization adjuncts (i.e., condom or rubber glove finger tube over speculum, lateral side-wall retractor) if necessary to enhance colposcopic visualization. 4. Remove mucus and debris by liberally applying 5% acetic acid to the cervix using cotton balls and ring forceps, large cotton swabs, or by spray technique. Avoid use of 4 4 gauze pads. 5. Reapply acetic acid (or Lugol s solution) to the cervix for a minimum of 60 seconds. Thereafter, reapply acetic acid every 3 to 5 minutes or when the columnar epithelium is no longer blanched white. 6. Identify the entire squamocolumnar junction (360 ), if able. 7. Identify acetowhite cervical lesions if present. 8. Assess the severity of each cervical lesion using green filter as needed to examine the vascular patterns. 9. As a guide, use the Reid Colposcopic Index (RCI), which is provided in the Administrative Binder, to classify any identified cervical lesions. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

55 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Obtain cervical biopsies and, if medically indicated, performed an endocervical curettage (ECC) sample collection (see below for cervical biopsy and ECC procedures). 11. Prior to removal of the speculum, inspect the lateral walls of the vagina with the colposcope. If clinically indicated, apply 5% acetic acid or Lugol s solution to the entire epithelium and then view by low power colposcopic magnification, noting acetowhite (or brownish if Lugol s solution used) vaginal lesions. Collapse the speculum and rotate 90 degrees and re-open the blades to inspect the anterior and posterior walls of the vagina for vaginal lesions (again using the 5% acetic acid or Lugol s solution if clinically indicated). If vaginal lesions are present, use the guidance given in the protocol to classify vaginal lesions and to obtain required biopsies (see Sections and ). Collapse and remove the speculum. 12. Change gloves and perform a systematic inspection of the entire external genitalia using a good light source and a hand-held magnifying glass (4x to 5x power) or, optionally, a colposcope with low-power magnification. If external genital lesions are present, use the guidance given in the protocol to classify external genital lesions and to obtain required biopsies (see Sections and ) Procedures for Cervical Biopsy Use the appropriate ecrfs to note locations, clinical impressions, and biopsy numbers during the procedure. Use the cervical biopsy kit (labels/requisition) and formalin fixative provided by the SPONSOR-designated Central Laboratory. 1. Each discrete abnormal area that is observed on colposcopy should be biopsied. The most severe area of abnormality of a lesion observed on colposcopy should be biopsied. 2. Apply local anesthesia, if this is the local standard of care. 3. If taking multiple biopsies, start with the most posterior (i.e. lower) lesion first. This will prevent contamination of subsequent biopsy sites by the flow of blood from the previous biopsy sites. 4. Place biopsy forceps over the abnormal lesion (usually near the squamocolumnar junction). 5. Open forceps jaws to sufficient extent. 6. Check to make sure the forceps are properly aligned (fixed end of forceps jaw towards cervical os). 7. Rotate biopsy handles to place lesion in the center of biopsy jaw angle if necessary. 8. Exert moderate pressure with biopsy forceps to push cervix backwards until cervix is fixed in position. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

56 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Squeeze handles together quickly and firmly to close forceps jaws and excise lesion. 10. Lock biopsy jaws to secure tissue (if locking mechanism available) or just hold tightly, then pass the forceps to the assistant. 11. Confirm by colposcopic visualization that an appropriate and adequate biopsy was collected. The biopsy should be perpendicular to the epithelium and deep enough to sample the entire epithelium along with a small amount of stroma (at least 2 mm) for histology. 12. Tamponade biopsy site with cotton swab. 13. Place the biopsy piece in the fixative container and ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. 14. Repeat the biopsy procedure for each additional biopsy performed using a different pair of forceps. If multiple biopsies are obtained, each specimen must be placed in a separate fixative container. A maximum of 4 cervical biopsies should be obtained. Collect all biopsy specimens prior to establishing hemostasis, if possible. 15. Obtain complete hemostasis with directed silver nitrate stick or Monsel s (ferric subsulfate) paste application. 16. Instruct subject to: (1) take nonsteroidal anti-inflammatory drugs for uterine cramping (provided no allergy), (2) report significant bleeding immediately, and (3) refrain from use of vaginal products, douching, tampons, and sexual intercourse for 3 to 7 days. 17. On the day of collection, ship the sample at room temperature as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. If no lesion is noted on colposcopy, and the subject was referred based on a cytological abnormality, endocervical curettage (ECC) may be performed Procedures for Endocervical Curettage (ECC) Use the cervical biopsy kit (labels/requisition) and formalin fixative provided by the SPONSOR-designated Central Laboratory. A curet must be used (not a cytobrush) when obtaining the ECC specimen. The decision to perform an ECC will be made by the study investigator according to local standards and practices. 1. Obtain an endocervical curettage specimen either before or after endocervical biopsies (based on local standard of care). 2. Insert open basket endocervical curet into cervical os. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

57 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Apply gentle pressure at the distal tip of curet against the endocervical canal and pull the curet from inside to outside with pressure, while simultaneously rotating the curet in a circular direction. The curet should advance no more that 20 mm up the canal and should be rotated completely 2 to 3 times. Avoid sampling ectocervical lesions that extend proximally into the endocervical canal, if possible. 4. Spin the curet rapidly to trap epithelium in the basket and remove the curet from the canal. Scraped material remaining in the canal can be retrieved with small forceps. 5. Place the ECC specimen into the fixative vial (or place the ECC specimen on paper or Telfa and transfer it to the fixative vial) and ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. 6. On the day of collection, ship the sample at room temperature as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory Procedures for Loop Electrosurgical Excision Procedure (LEEP) and Top Hat Method The LEEP is a cervical definitive therapy method that uses a shallow pass to remove the cervical transformation zone. It may be a single or multiple pass and is ~8 mm deep. A top hat cone implies a deeper wedge resection, ~15 to 25 mm deep. It is performed by the top hat method of successively more internal and smaller loops. Use the appropriate ecrfs to note clinical impressions and biopsy numbers during the procedure. Use the definitive therapy kit (labels/requisitions) and formalin fixative provided by the SPONSOR-designated Central Laboratory. The LEEP should be done under colposcopic control. All instruments used for the procedure must be nonconductive. 1. Assist subject into dorsal lithotomy position. 2. Place dispersive pad near operative site (thigh) and plug into electrosurgical unit (ESU). 3. Place vaginal speculum and secure smoke evacuation tubing. 4. Set electrosurgical unit parameters (power, cut/blend) and test safety systems. 5. Identify cervical pathology by colposcopic examination. 6. If appropriate, apply half-strength Lugol s solution to cervix and identify transformation zone limits. 7. Give local anesthesia if no allergy 4 quadrant (locations 3, 6, 9, 12 o clock) intracervical injection of lidocaine with epinephrine. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

58 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Obtain pre-definitive cervical biopsies of the areas with the most severe abnormalities (2- to 3-mm size per location) and use the cervical biopsy labels/requisition provided within the definitive therapy kit (see Section for cervical biopsy guidelines). 9. Select appropriate loop electrode size to remove transformation zone and lesion, insert into hand piece and plug hand piece into electrosurgical unit. 10. Activate smoke evacuator. 11. Initiate excision by depressing cut switch if using handpiece, cut down (perpendicular to tissue), across transformation zone and straight out at opposite side of transformation zone (depth of 8 mm) and avoid vaginal sidewall contact. 12. Repeat cut procedure (reduced power) along endocervical canal with smaller 10 x 10 loop electrode if top hat conization is necessary. 13. Hand excised tissue to assistant and dictate its orientation. Orient the tissue by placing a suture at the 12 o clock location. Place tissue in the fixative container and ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. Note: If the top hat procedure is performed, the tissue from each loop or pass should be placed in separate labeled fixative containers. 14. Inspect endocervical canal opening and perform ECC, if medically indicated. Use the ECC label/requisition provided within the definitive therapy kit (see Section for ECC guidelines). 15. Fulgurate base of the excision with coagulation using the ball or paddle electrode until adequate hemostasis and fulgurate 5 mm of the ectocervical margin. If lesion extends beyond the excision, ablate the area and record the ablation on the appropriate ecrf and source document. 16. If hemostasis is inadequate, pack base of excision with Monsel s (ferric subsulfate) paste or in rare event suture may be required. 17. Remove blood from posterior fornix. 18. Deactivate ESU and smoke evacuator. 19. Remove dispersive pad and vaginal speculum. 20. Assist subject up from table. 21. Instruct subject to: (1) take nonsteroidal anti-inflammatory drugs as needed (provided no allergy), (2) refrain from use of vaginal products, douching, tampons, and sexual intercourse for 2 to 4 weeks, and (3) report significant bleeding and signs of infection (fever, pain) immediately. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

59 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: On the day of collection, ship the sample at room temperature as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory Procedure for Laser Conization Laser Conization is a cervical definitive therapy procedure that uses a laser beam to excise tissue, using small depressor instruments or hooks to position the cone for optimal excision. Use the appropriate ecrfs to note clinical impressions and biopsy numbers during the procedure. Use the definitive therapy kit (labels/requisitions) and formalin fixative provided by the SPONSOR-designated Central Laboratory. 1. Assist subject into dorsal lithotomy position. 2. Place vaginal speculum and secure smoke evacuation tubing. 3. Set laser unit parameters (power, optical focusing of the laser beam) and test safety systems. 4. Identify cervical pathology by colposcopic examination. 5. If appropriate, apply half-strength Lugol s solution to cervix and identify transformation zone limits. 6. Give local anesthesia if no allergy 4 quadrant (3, 6, 9, 12) intracervical injection of lidocaine with epinephrine. If needed paracervical injection of the anesthetic may also be used. General anesthesia may be provided if so preferred. If general anesthesia is preferred, intracervical injection of lidocaine with epinephrine may still be used if there is no allergy to the components of the injection. 7. Obtain pre-definitive cervical biopsies of the areas with the most severe abnormalities (2- to 3-mm size per location) and use the cervical biopsy labels/requisition provided within the definitive therapy kit (see for cervical biopsy guidelines). 8. Activate smoke evacuator. 9. Initiate excision by marking the border of the area to be excised with the laser beam. 10. Cut the cone by the laser beam, using small depressor instruments or hooks to position the cone for optimal excision with the laser beam. 11. Hand excised tissue to assistant and dictate its orientation. Orient the tissue by placing a suture at the 12 o clock location. Place tissue in the fixative container and ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

60 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Inspect endocervical canal opening and perform ECC, if medically indicated. Use the ECC label/requisition provided within the definitive therapy kit (see Section for ECC guidelines). 13. Use the defocused laser beam in the wound area until adequate hemostasis, and vaporize 5 mm of the ectocervical margin. If lesion extends beyond the excision, ablate the area and record the ablation on the appropriate ecrf and source document. 14. If hemostasis is inadequate, inject more lidocaine with adrenaline or in rare event suture may be required. 15. Remove blood from posterior fornix. 16. Remove dispersive pad and vaginal speculum and deactivate smoke evacuator. 17. Assist subject up from table. 18. Instruct subject to: (1) take nonsteroidal anti-inflammatory drugs as needed (provided no allergy), (2) refrain from use of vaginal products, douching, tampons, and sexual intercourse for 2 to 4 weeks, and (3) report significant bleeding and signs of infection (fever, pain) immediately. 19. On the day of collection, ship the sample at room temperature as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory Cold-Knife Conization Cervical cold-knife conization should be performed only when the LEEP or laser conization procedures are not indicated. A cervical cold-knife conization specimen represents a conically shaped section of the cervix that will vary in size according to the lesion. A broad, shallow conization would be performed for a predominantly exocervical lesion and a narrow, deep conization is appropriate for a predominantly endocervical lesion. Pre-definitive biopsies should be obtained prior to the cervical cold-knife conization. Use the appropriate ecrfs to note clinical impressions and biopsy numbers during the procedure. Use the definitive therapy kit (labels/requisitions) for pre-definitive biopsy/ecc (if indicated)/definitive therapy tissue and use the formalin fixative provided by the SPONSOR-designated Central Laboratory. The samples should be labeled/shipped as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory (ship at room temperature, ship on day of collection) Ablative Cervical Definitive Therapy Ablative cervical definitive therapy is not a study procedure, because this procedure is not tissue preserving. If a subject undergoes ablative cervical definitive therapy (e.g., cervical cryotherapy or cervical laser vaporization) or a hysterectomy during the study, obtain pre-definitive biopsy specimens. Use the definitive therapy kit (labels/requisitions) for pre-definitive biopsy/ecc (if indicated) and use the formalin fixative provided by the SPONSOR-designated Central Laboratory. The samples should V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

61 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: be labeled/shipped as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory (ship at room temperature, ship on day of collection) Discontinuation/Withdrawal from Study Subjects/patients may withdraw at any time or be dropped from the study at the discretion of the investigator should any untoward effects occur. In addition, a subject/patient may be withdrawn by the investigator or the SPONSOR if he/she violates the study plan or for administrative and/or other safety reasons. The investigator or study coordinator must notify the SPONSOR immediately when a subject/patient has been discontinued/withdrawn due to an adverse experience (telephone or FAX). When a subject/ patient discontinues/withdraws prior to study completion, all applicable activities scheduled for the final study visit should be performed at the time of discontinuation. Any adverse experiences which are present at the time of discontinuation/withdrawal should be followed in accordance with the safety requirements outlined in section 3.4 SAFETY MEASUREMENTS - DETAILS. If a subject must discontinue/withdraw after the Month 7 study visit, the subject should be asked to return for a final discontinuation visit if it has been at least 4 months since the last study visit. This visit would consist of the same specimen collections and tests conducted at the last study visit and the subject would be formally discontinued from the study at the end of this visit. The discontinuation visit should not be done if it is medically contraindicated or if the subject refuses. If no discontinuation visit is performed, the subject should be formally discontinued from the study on the day the decision to discontinue is made. All attempts must be made to contact a subject who is lost to follow-up (a certified letter must be sent at the final attempt). Subjects who are lost to follow-up should be formally discontinued from the study on the day of the last unsuccessful attempt at contact. At a minimum, the cover sheet and subject status ecrfs must be completed and submitted to the SPONSOR for all discontinuations Subject Relocation Given the duration of the study and the age of the study population, it can be expected that subjects may relocate during the study. The SPONSOR must be contacted for each temporary and permanent relocation as soon as the situation is known. Every effort should be made to adjust study visits around a subject s temporary absence (e.g., college breaks, summer vacation) so that the visits will be within the visit windows. Every effort should be made to have a relocated subject seen at another site participating in this study in order to keep the visits within the visit windows and to allow the subject to complete the study Conduct of the Clinical Trial The Scientific Advisory Committee will provide scientific input and direction. Clinical efficacy endpoints will be adjudicated by the HPV Vaccine Program Pathology Panel, V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

62 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: who is responsible for the definitive pathologic diagnoses in all clinical conditions which may be considered as possible endpoints in subjects in this trial. The Data and Safety Monitoring Board will be responsible for monitoring tolerability and, in conjunction with the Senior Management Committee, will participate in the selection of the best Part A 9- valent HPV L1 VLP vaccine dose for use in Part B of the trial. The operation of the Scientific Advisory Committee, the HPV Vaccine Program Pathology Panel, the Data and Safety Monitoring Board, and the Senior Management Committee are described below Scientific Advisory Committee The SAC (Scientific Advisory Committee) is a body of internal representatives and external scientific leaders in the field of gynecology, oncology, and/or HPV research. The SAC will convene to provide advice on the protocol direction, Statistical Analysis Plan, and to interpret and publish the study results of confirmatory studies. The SAC will be comprised of external leaders and internal representatives HPV Vaccine Program Pathology Panel The HPV Vaccine Program Pathology Panel will be responsible for providing the definitive pathologic diagnoses of cervical biopsies, vaginal biopsies, external genital lesion biopsies, endocervical curettage, and cervical definitive therapy specimens for study purposes (not for medical management). Cervical histology slides and slides from external genital lesion biopsies and vaginal lesion biopsies will be evaluated by HPV Vaccine Program Pathology Panel. The HPV Vaccine Program Pathology Panel will prepare reports on each tissue specimen without knowing the vaccination groups of the subjects. A separate guideline that details the HPV Vaccine Program Pathology Panel process has been approved by the HPV Vaccine Program Pathology Panel Responsibility of the Data and Safety Monitoring Board (DSMB) The Data and Safety Monitoring Board (DSMB) assesses the effects of the study vaccine during the trial. The DSMB will review all available tolerability data at the following milestones during the Part A and Part B vaccination periods: 1. Milestone 1: VRC data from approximately 25% of Part A Day 1 vaccination visits are available in the study database, 2. Milestone 2: VRC data from approximately 100% of Part A Day 1 vaccination visits are available in the study database, 3. Milestone 3: VRC data from approximately 25% of Part A Month 2 vaccination visits are available in the study database, 4. Milestone 4: VRC data from approximately 100% of Part A Month 2 vaccination visits are available in the study database (this milestone will involve review of the dose selected by the senior management committee, as discussed in Section ), V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

63 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Milestone 5: VRC data from approximately 100% of Part A Month 6 vaccination visits are available in the study database (review will include all available VRC data from Part B), 6. Milestone 6: VRC data from approximately 100% of Part B Month 2 vaccination visits are available in the study database In addition to above milestones, the DSMB may also meet on an ad-hoc basis at the request of the DSMB chair, the protocol team, the site, or a study investigator. With the exception of a non-voting Merck & Co., Inc. statistician, the members of the DSMB are independent of Merck Research Laboratories and clinical investigators participating in this trial, and will not have any other involvement in the study, nor will they have any relation to the study participants. The DSMB monitors the trial for evidence of adverse effects of the study vaccine using the guidelines proposed by the protocol. The DSMB may recommend any steps to ensure the safety and integrity of the trial. Furthermore, the DSMB may recommend that the trial be terminated or that specific high-risk patient groups be withdrawn from the study, if any subgroup manifests serious or widespread side effects. To guarantee the unrestricted performance of its task, the DSMB may receive the individual study morbidity and mortality data from a designated MRL statistician. Unblinded reports of all serious adverse experiences in this study from the MRL Worldwide Adverse Experience database and the clinical database will be presented to the DSMB. The monitoring guidelines that the DSMB will follow will be described in the Statistical Analysis Plan. The DSMB minutes will summarize the actions and deliberations of the DSMB and will be made available to Merck & Co., Inc. at the conclusion of the trial. A full manual of operations for the DSMB will be written and approved by the DSMB Senior Management Committee for Dose Selection As outlined in section 2.4.1, when ~100% postdose 2 Part A data are available in the clinical database, a senior management committee will choose the dose for Part B after reviewing immunogenicity summaries by vaccination group, and will relay the dose selection to the DSMB, who will approve the decision based on unblinded tolerability data. Details of the process for dose selection and the procedures used to maintain blinding will be provided in a guidance document, which will be approved before the senior management committee or the DSMB convenes to review the dose-selection data. The senior management committee for dose selection will include a small number of Merck & Co., Inc. personnel who are experts in medical, regulatory, statistical and epidemiological aspects of clinical trials. Members will not be involved in study conduct. The senior management committee for dose selection will follow a protocolspecific charter of operations, which the committee will approve before convening to review dose-selection data. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

64 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: EFFICACY/IMMUNOGENICITY MEASUREMENTS Competitive Luminex Immunoassay (clia) - Anti-HPV Levels in Serum The purpose of the HPV competitive Luminex immunoassay (clia) is to detect neutralizing antibodies to L1 virus-like particles (VLPs) of specific HPV Types (6, 11, 16, 18, 31, 33, 45, 52, and 58), before and after vaccination with GARDASIL or the 9- valent HPV L1 VLP vaccine. This is the primary assay used by the Vaccines and Biologics Research Serology Laboratory of Merck Research Laboratories (MRL) to evaluate the serological response to the vaccine, to measure HPV infection induced antibody, and to exclude subjects with evidence of a current or past HPV infection from the primary analysis. Yeast-derived VLPs are coupled to a set of 9 distinct fluorescent Luminex microspheres. Antibody titers are determined in a competitive format in which known, type-specific phycoerythrin (PE)-labeled, neutralizing monoclonal antibodies (mabs) compete with the subject s serum antibodies for binding to conformationally sensitive, neutralizing epitopes on the VLPs. The fluorescent signal detected from the bound HPV-specific detection mabs is inversely proportional to the subject s neutralizing antibody titer. Results for the assay will be reported as concentration of antibody in milli-merck Units per milliliter (mmu/ml) of neutralizing monoclonal antibody equivalent. The seropositivity cutoffs for HPV types were assessed using a panel of sera from subjects highly likely to be HPV naïve (children), and from subjects who were highly likely to be seropositive. Any sample with a value less than the cutoffs are considered serostatus negative PCR Assays - Detection of HPV in Swabs and Tissue Specimens Cervicovaginal swabs and all Thinsection microtomy biopsy specimens will be tested for detection of HPV types 6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59. In addition to this testing, cervicovaginal swabs and Thinsection microtomy biopsy specimens may tested for other HPV types. HPV types 6, 11, 16, 18, 31, 33, 45, 52, and 58 will be analyzed by type-specific multiplex (L1, E6, E7 gene detection) PCR assay (described in Section ). HPV types other than 6, 11, 16, 18, 31, 33, 45, 52, and 58 will be analyzed by the duplex (E6, E7 gene detection) PCR assay (using the preparation method described in Section ) Multiplex PCR Assays The following procedures will be done at Merck Research Laboratories (MRL), West Point, PA, for the detection of HPV Types 6, 11, 16, 18, 31, 33, 45, 52, and 58 in frozen swabs and Thinsection microtomy biopsy samples. Specimens are received and then prepared for multiplex PCR using a DNA purification method (Qiagen Technology Kit). Multiplex PCR (based on real-time fluorescent PCR) allows the simultaneous detection of 3 gene products (L1, E6, and E7) for a given HPV type in 1 reaction. The HPV typespecific primer pairs based on the published HPV L1, E6, and E7 sequences, are used to V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

65 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: specifically amplify a portion of each gene simultaneously. The specific amplicons are detected in real-time by fluorescently-labeled oligonucleotide probes. The gene-specific oligonucleotide probes are each labeled with a different fluorescent label, and the fluorescent emission is captured during PCR cycling. After analysis of the raw fluorescent data by the Sequence Detection Software, a threshold cycle (Ct), which represents the PCR cycle at which an increase in reporter fluorescence above a baseline signal can first be detected, is determined. Each genespecific assay (i.e., gene-specific dye layer) is considered positive if the Ct is <45 cycles. A gene-specific assay is considered negative if the Ct 45. A sample is called positive when 2 or 3 genes are positive or when the same single gene scores positive on consecutive tests Preparation and Disposition of Thinsections of Biopsy Tissue The following procedures will be performed at the SPONSOR-designated Central Laboratory. The procedures will be performed by an experienced, qualified histotechnologist according to the Central Laboratory s (Standard Operating Procedure (SOP). The histotechnologist will assure that the microtome and work areas are clean and free of contaminants. All Thinsection microtomy for PCR will be performed at a time when all other routine work has been completed, so that potential contaminations can be minimized. Prior to sectioning each block, a new blade will be installed in the microtome. The blade will only be positioned so that it is at the left margin of the blade surface. Technicians sectioning study blocks will utilize biologically clean gloves while handling the blocks (new gloves for each block). First, the histotechnologist will face the block by removing two 4-micron sections from the face of the block. These sections will be discarded. Using sterile plastic forceps, the next two 4-micron paraffin sections are collected and floated in a water bath for the preparation of 1 H&E slide (Slide 1, with 2 sections). Nine additional, consecutive sections will then be cut to be used for Thinsection PCR. There will be 9 individual tubes (Tube 1, 2, 3, 4, 5, 6, 7, 8, 9), and one 4-micron section will be placed in each tube using a sterile disposable plastic forceps. The pair of sterile plastic forceps used is then discarded after placing the cut section in each tube. Each tube is then placed inside a plastic sleeve and sealed. Two additional, consecutive 4-micron sections will then be cut and the 2 sections floated in the water bath for preparation of the second H&E slide both sections to be placed on one slide (Slide 2 with 2 sections each). All H&E slides (Slides 1 and 2) will have a histopathologic review by the central laboratory s pathologist. Slides and tubes should be labeled with subject s allocation number. The specimen tubes are collated with the appropriate specimen requisition and prepared for shipping to the SPONSOR-designated Central Laboratory and then in turn, shipped on to MRL. The microtome is cleaned in preparation for the next block and the process above is repeated. The microtome blade is replaced with a new blade and adjusted for each new V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

66 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: biopsy block and the same procedure is to be followed. A new pair of clean gloves and a new pair of clean, disposable forceps will be used for each block being sectioned. The used blade may be retained for cutting non-pcr blocks. The total number of sections to be cut from each block is 13. A total of 2 slides and 9 tubes: 1. Slide 1 (H&E), with 2 sections each, stained. 2. Tubes 1, 2, 3, 4, 5, 6, 7, 8, 9 (HPV PCR Analysis), 1 section per tube. 3. Slide 2 (H&E), with 2 sections each, stained. 3.4 SAFETY MEASUREMENTS Clinical and Laboratory Measurements for Safety All subjects will be observed for at least 30 minutes after each study vaccination for any untoward effects, including allergic reactions. This observation period will be documented in the subject s study chart. Each subject will receive a Vaccination Report Card (VRC) at the Day 1, Month 2, and Month 6 study vaccination visits. On the VRC, the subject will be asked to record her oral temperatures in the evening after each study vaccination and daily for 4 days after each study vaccination for the purpose of identifying febrile events. The subject should be instructed to collect temperatures at the same time of day whenever possible. Also, beginning after each study vaccination and for a total of 15 days including the day of vaccination, the subject will be asked to record injection-site and systemic adverse experiences, concomitant medications, and concomitant vaccinations on the VRC. The information on the VRC should be generated only by the subject and must be signed and dated by the subject to confirm the accuracy of the recorded information. The subject will be asked to bring the VRC to the study site at the next scheduled visit (with the exception of the first 150 subjects enrolled in Part A, who will submit the VRC at a Day 16 visit). When the VRC is returned, the VRC should be reviewed for completeness by study site personnel. If clarification is needed, the study site personnel will discuss the VRC with the subject. Original information on the VRC should never be altered by study personnel, although comments can be written in the designated area for study site personnel on the front of the VRC. Only the subject can make corrections to her information on the VRC. Any corrections to the VRC should be made by using an ink pen to add the omitted data and/or to draw a single line through the error and add the correct information. The subject should initial/date all VRC corrections. All VRC information will be recorded in the ecrfs. The physician investigator/subinvestigator will determine causality of systemic and injection-site adverse experiences recorded on the VRC using the reporting guidelines given in the protocol and will classify each event as a serious adverse experience (SAE) or non-serious adverse experience (NSAE). If an oral temperature indicates a fever (defined as an oral temperature of 100 F or 37.8 C), the adverse experience of fever must be documented in the ecrfs. At the time of VRC review at the next scheduled visit, V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

67 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: subjects will be questioned regarding any new medical conditions that occurred beyond Postdose Day 15. The physician investigator/sub-investigator will determine if the medical condition is to be reported as an SAE using the reporting guidelines provided below Recording Adverse Experiences An adverse experience is defined as any unfavorable and unintended change in the structure, function, or chemistry of the body temporally associated with the use of the SPONSOR s product, whether or not considered related to the use of the product. Any worsening (i.e., any clinically significant adverse change in frequency and/or intensity) of a preexisting condition which is temporally associated with the use of the SPONSOR s product, is also an adverse experience. Changes resulting from normal growth and development which do not vary significantly in frequency or severity from expected levels are not to be considered adverse experiences. Examples of this may include, but are not limited to, teething, typical crying in infants and children, and onset of menses or menopause occurring at a physiologically appropriate time. All adverse experiences will be collected from the time the consent form is signed through 14 days following the first vaccination(s) and from the time of any subsequent vaccination(s) through 14 days thereafter, and such events will be recorded at each examination on the Adverse Experience Case Report Forms/Worksheets Definition of an Overdose for This Protocol In this study, an overdose is defined as a subject receiving >3 doses (0.5 ml each dose) or receiving 0.75 ml of vaccine in any one dose Reporting of Overdose to SPONSOR If an adverse experience(s) is associated with ( results from ) the overdose of test drug or vaccine, the adverse experience(s) is reported as a serious adverse experience, even if no other criteria for serious are met. If a dose of test drug or vaccine meeting the protocol definition of overdose is taken without any associated clinical symptoms or abnormal laboratory results, the overdose is reported as a non-serious Event of Clinical Interest (ECI), using the terminology accidental or intentional overdose without adverse effect. All reports of overdose with and without an adverse experience must be reported within 24 hours to one of the individuals listed on the sponsor contact information page found in the Administrative Binder Reporting of Pregnancy/Breastfeeding Events to the SPONSOR Although not considered an adverse experience, it is the responsibility of the investigators or their designees to report any pregnancy in a subject (spontaneously V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

68 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: reported to them or detected by urine or serum pregnancy test per protocol). If a subject becomes pregnant within the study and the last menstrual period date is on or before the Month 7 visit, then an Adverse Experience Report (AER) form must be submitted. This reporting of pregnancies using the AER form includes subjects who were never randomized and had a positive pregnancy test at Day 1. For randomized subjects who become pregnant after receiving one or two study vaccinations, study visits and study vaccinations will be paused until resolution of the pregnancy (e.g., term, elective termination, spontaneous abortion). Study visits and study vaccinations in pregnant subjects will be handled as described in Table 3-1. All randomized subjects who receive study vaccine, including discontinued subjects who agree to provide further information, will be followed to the completion/termination of the pregnancy. In addition, if the pregnancy continues to term, the outcome (health of the infant) must be reported. If a subject receives study vaccine while breastfeeding during the Day 1 through Month 7 period, an AER form must be submitted for the lactation event and its outcome must be reported. Infant serious adverse experiences (SAEs) for all infants born to subjects who received study vaccine must be reported to the SPONSOR. The reporting of pregnancy, lactation, and infant SAE events involves completing AER forms, various ecrfs, and/or questionnaires. For detailed guidelines, see the Attachments entitled Pregnancy Reporting and Follow-Up HPV Vaccine Clinical Program, and Lactation Reporting and Follow-Up HPV Vaccine Clinical Program. All initial and follow-up AER forms for pregnancy and lactation must be submitted to the SPONSOR within 5 business days. All SAEs, including pregnancy-related SAEs, lactation-related SAEs, and infant SAEs must be reported within 24 hours Immediate Reporting of Adverse Experiences to the SPONSOR Serious Adverse Experiences Any serious adverse experience, including death due to any cause, which occurs to any subject from the time the consent is signed through 14 days following the first vaccination(s) and from the time of any subsequent vaccination(s) through 14 days thereafter, whether or not related to the investigational product, must be reported within 24 hours to one of the individual(s) listed on the sponsor contact information page found in the Administrative Binder. Additionally, any serious adverse experience brought to the attention of an investigator who is a qualified physician at any time outside of the time period specified in the previous paragraph also must be reported immediately to one of the individuals listed on the sponsor contact information page (found in the administrative binder) if the event is either: 1. A death which resulted in the subject/patient discontinuing the study or 2. A serious adverse experience that is considered by an investigator who is a qualified physician to be possibly, probably, or definitely vaccine related. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

69 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: or 3. A serious adverse experience that is considered by an investigator who is a qualified physician to be possibly, probably, or definitely related to a study procedure. All subjects/patients with serious adverse experiences must be followed up for outcome Evaluating Adverse Experiences Refer to Table 3-4 for instructions in evaluating adverse experiences. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

70 Product: V Protocol/Amendment No.: Table 3-4 An investigator who is a qualified physician, will evaluate all adverse experiences as to: Maximum Mild awareness of sign or symptom, but easily tolerated (for pediatric studies, awareness of symptom, but easily tolerated) Intensity Moderate discomfort enough to cause interference with usual activity (for pediatric studies, definitely acting like something is wrong) Severe incapacitating with inability to work or do usual activity (for pediatric studies, extremely distressed or unable to do usual activities) Injection site redness or swelling from the day of vaccination through Day 4 post-vacc will be evaluated by maximum size. Seriousness A serious adverse experience is any adverse experience occurring at any dose that: Results in death; or Is life threatening; or places the subject/patient, in the view of the investigator, at immediate risk of death from the experience as it occurred [Note: This does not include an adverse experience that, had it occurred in a more severe form, might have caused death.]; or Results in a persistent or significant disability/incapacity (substantial disruption of one s ability to conduct normal life functions); or Results in or prolongs an existing inpatient hospitalization (hospitalization is defined as an inpatient admission, regardless of length of stay, even if the hospitalization is a precautionary measure for continued observation. (Note: Hospitalization [including hospitalization for an elective procedure] for a preexisting condition which has not worsened does not constitute a serious adverse experience.); or Is a congenital anomaly/birth defect (in offspring of subject/patient taking the product regardless of time to diagnosis);or Is a cancer; or Is an overdose (Whether accidental or intentional.) Any overdose whether or not associated with an adverse experience must be reported within 24 hours to one of the individuals on the Contact Information Page found in the Administrative Binder. Other important medical events that may not result in death, not be life threatening, or not require hospitalization may be considered a serious adverse experience when, based upon appropriate medical judgment, the event may jeopardize the subject/patient and may require medical or surgical intervention to prevent one of the outcomes listed previously (designated above by a ). Duration Action taken Relationship to test vaccine Record the start and stop dates of the adverse experience. If less than 1 day, indicate the appropriate length of time and units Did the adverse experience cause the test vaccine to be discontinued? Did the test vaccine cause the adverse experience? The determination of the likelihood that the test vaccine caused the adverse experience will be provided by an investigator who is a qualified physician. The investigator s signed/dated initials on the source document or worksheet, that supports the causality noted on the AE form, ensures that a medically qualified assessment of causality was done. This initialed document must be retained for the required regulatory time frame. The criteria below are intended as reference guidelines to assist the investigator in assessing the likelihood of a relationship between the test vaccine and the adverse experience based upon the available information. The following components are to be used to assess the relationship between the test vaccine and the AE; the greater the correlation with the components and their respective elements (in number and/or intensity), the more likely the test vaccine caused the adverse experience (AE): Exposure Time Course Likely Cause Is there evidence that the subject/patient was actually exposed to the test vaccine such as: reliable history, acceptable compliance assessment (e.g. diary), seroconversion or identification of vaccine virus in bodily specimen? Did the AE follow in a reasonable temporal sequence from administration of the test vaccine? Is the time of onset of the AE compatible with a vaccine-induced effect? Is the AE not reasonably explained by another etiology such as underlying disease, other drug(s)/vaccine(s), or other host or environmental factors V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007 V503, Protocol Issue Date: 15-Jun

71 Product: V Protocol/Amendment No.: Relationship to test vaccine (continued) The following components are to be used to assess the relationship between the test vaccine and the AE: (continued) Dechallenge (not applicable for vaccines) Rechallenge Consistency with Study Vaccine Profile Was the subject/patient reexposed to the test vaccine in this study? If yes, did the AE recur or worsen? If yes, this is a positive rechallenge. If no, this is a negative rechallenge. (Note: This criterion is not applicable if: (1) the initial AE resulted in death or permanent disability, or (2) the study is a single-dose vaccine study.) NOTE: IF A RECHALLENGE IS PLANNED FOR AN ADVERSE EVENT WHICH WAS SERIOUS AND WHICH MAY HAVE BEEN CAUSED BY THE TEST VACCINE, OR IF REEXPOSURE TO THE TEST VACCINE POSES ADDITIONAL POTENTIAL SIGNIFICANT RISK TO THE SUBJECT/PATIENT, THEN THE RECHALLENGE MUST BE APPROVED IN ADVANCE BY THE U.S. CLINICAL MONITOR AND THE INSTITUTIONAL REVIEW BOARD/INDEPENDENT ETHICS COMMITTEE. Is the clinical/pathological presentation of the AE consistent with previous knowledge regarding the test vaccine or vaccine class pharmacology or toxicology? The assessment of relationship will be reported on the case report forms /worksheets by an investigator who is a qualified physician according to his/her best clinical judgment, including consideration of the above elements. Use the following scale of criteria as guidance (not all criteria must be present to be indicative of a vaccine relationship). Definitely related Probably related Possibly related Probably not related Definitely not related There is evidence of exposure to the test vaccine. The temporal sequence of the AE onset relative to administration of the test vaccine is reasonable. The AE is more likely explained by the test vaccine than by another cause. Dechallenge is positive. Rechallenge (if feasible) is positive. The AE shows a pattern consistent with previous knowledge of the test vaccine or test vaccine class. There is evidence of exposure to the test vaccine. The temporal sequence of the AE onset relative to administration of the test vaccine is reasonable. The AE is more likely explained by the test vaccine than by another cause. Dechallenge (if performed) is positive. There is evidence of exposure to the test vaccine. The temporal sequence of the AE onset relative to administration of the test vaccine is reasonable. The AE could have been due to another equally likely cause. Dechallenge (if performed) is positive. There is evidence of exposure to the test vaccine. There is another more likely cause of the AE. Dechallenge (if performed) is negative or ambiguous. Rechallenge (if performed) is negative or ambiguous. The subject/patient did not receive the test vaccine. OR Temporal sequence of the AE onset relative to administration of the test vaccine is not reasonable. OR There is another obvious cause of the AE. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007 V503, Protocol Issue Date: 15-Jun

72 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: SPONSOR Responsibility for Reporting Adverse Experiences All adverse experiences will be reported to regulatory agencies, IRB/IECs, and investigators in accordance with all applicable global laws and regulations. 3.5 DATA ANALYSIS Responsibility for and Timing of Analyses The data collected under this protocol will be analyzed by the Vaccine Clinical Biostatistics Department of the SPONSOR. A comprehensive Statistical Analysis Plan will be provided prior to the interim analysis of Part A. This study will be conducted using in-house blinding procedures. This study will have a data and safety monitoring board (DSMB). An unblinded statistician at the SPONSOR who is otherwise unrelated to this protocol will serve as a non-voting member of the DSMB and will be responsible for providing summaries of safety results to the rest of the DSMB. The unblinded statistician will also be responsible for generating grouped summaries of immunogenicity data for the Week 4 Postdose 2 (Month 3) interim analysis for distribution to a senior management committee at the SPONSOR who is not directly involved with study conduct for the purpose of selecting a dose for the safety, immunogenicity, and efficacy evaluations in Part B. Part A The interim analysis of immunogenicity will be conducted when ~100% of all serology data are available through Week 4 Postdose 2 (Month 3) for subjects in Part A. At that time, the database will be made available only to the unblinded statistician in order to carry out the immunogenicity analysis. All available data related to the immunogenicity summaries will be screened and cleaned in a blinded manner before the database is made available to the unblinded statistician. Group-level summaries of immune responses (GMTs, seroconversion rates) will be provided to a senior management committee at the SPONSOR. No serology or randomization data for individual study participants will be disclosed. Unblinded summaries of all available tolerability data in Part A will be provided to the DSMB for review. No efficacy results will be summarized. The senior management committee with the assistance provided by the DSMB will select the dose using the process outlined in Section All subjects in Part A will be followed through at least Month 7, and subjects who are not in the vaccination groups selected for Part B will be discontinued from the study after the Month 7 visit. The unblinded statistician will provide a list of allocation numbers by site for such subjects to the Clinical Operations team, who will subsequently notify the applicable study sites. No vaccination group information will be released to the study sites. The primary immunogenicity analysis in Part A will be conducted based on the Week 4 Postdose 3 (Month 7) immunogenicity results of the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B and the GARDASIL comparator, and will be conducted by the unblinded statistician once all Month 7 data are available. At that time, V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

73 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: the critical database fields for conducting the primary immunogenicity analyses for Part A and for identifying protocol violators will be screened and cleaned in a blinded manner by the designated clinical, statistical, and data management team. After all protocol violators are identified, the database will be audited, and a copy of the database will be frozen for use by the unblinded statistician. Group-level summaries of the immune responses (GMTs, seroconversion rates) for the selected 9-valent HPV L1 VLP formulation and GARDASIL will be provided to the senior management committee, as well as the results of the hypothesis tests. Subjects who were enrolled in a 9-valent HPV L1 VLP vaccine formulation with a lower total VLP concentration than the formulation that was selected for use in Part B may be offered a single dose of GARDASIL. Although identifying subjects who will be offered a single dose of GARDASIL could potentially unblind the clinical team to their vaccination group assignment, these subjects will be discontinued after the Month 7 visit in Part A and thus will not participate in the planned efficacy component of the study. Thus, the integrity of the blind will be preserved for the subjects in the selected 9-valent HPV L1 VLP vaccine formulation and GARDASIL arms who will proceed to Part B. The unblinded statistician will provide a list of allocation numbers of subjects who may be offered a single dose of GARDASIL to the study to the Clinical Operations group, if applicable. In the unlikely event that the SPONSOR is unable to determine an acceptable 9-valent HPV L1 VLP vaccine formulation for use in Part B, the study may not continue to Part B. Part B If the formulation selected by the senior management committee based on the post-dose 2 immunogenicity data is confirmed to have a favorable tolerability profile by the DSMB, then an additional 2,480 subjects will be randomized in a 1:1 ratio to receive the selected 9-valent HPV L1 VLP vaccine formulation or GARDASIL. Subjects from Part A enrolled in the GARDASIL arm and the 9-valent HPV L1 VLP formulation selected for Part B will remain in the trial for the evaluation of efficacy and safety. Because Merck & Co., Inc. personnel (senior management committee) will view summaries of immunogenicity data from Part A, no immunogenicity data from Part A will be used in the primary immunogenicity analyses in Part B. Part B employs a fixed event design. Although the study will continue for 42 months, the study conclusions regarding vaccine efficacy with respect to the primary endpoint will be based on the analyses conducted when at least 43 cases of the primary efficacy endpoint (HPV 31/33/45/52/58-related persistent infection for a duration of 6 months [within ±1 month windows] or longer or related clinical disease) have been observed. This is expected to occur when all subjects have completed 24 months in the study. Inference with respect to the primary efficacy hypothesis will be based on this analysis. The counting of efficacy cases will be done by the same unblinded statistician at the SPONSOR who will provide summaries of safety to the DSMB. If the primary efficacy analysis is conducted prior to the end of the study (i.e. before the Month 42 follow-up V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

74 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: visit of the last subject enrolled), the remainder of study follow-up still to be accrued will be considered an extension, the purpose of which is to collect additional information in order to assess vaccine efficacy over the entire follow-up period. Although the immunogenicity results in Part B may be available prior to the primary efficacy analysis, the primary immunogenicity analysis will be conducted at the time the database is unblinded for the primary efficacy analysis. The official clinical database for the primary efficacy analysis and primary immunogenicity analysis in part B will be unblinded only after medical/scientific review has been completed, and the database has been declared complete. After the primary efficacy analysis and the primary immunogenicity analysis are performed, data cleaning, data screening, and medical/scientific review of the database will continue unblinded until follow-up is complete. However, investigators, laboratory staff (including SPONSOR s laboratory staff), the members of the HPV Vaccine Program Pathology Panel, and study subjects will remain blinded until the conclusion of the study. All investigators and technicians who are responsible for the ascertainment and confirmation of efficacy endpoints will remain blinded for the duration of the study. If changes are made after the start of the study to the statistical analysis plan presented in this Data Analysis section, the changes made, along with an explanation as to why they occurred, will be listed in the Statistical Analysis Plan and/or the Clinical Study Report of this study, as appropriate Hypotheses The study hypotheses are listed in Section 2.1 of the protocol. The study will be considered a success if the primary efficacy hypothesis and the primary immunogenicity hypothesis in Part B are demonstrated Variables and Time Points of Interest Safety/Tolerability The important variables for safety/tolerability are the incidences of severe injection-site reactions and any vaccine-related serious adverse experiences Efficacy The primary efficacy endpoint is the combined incidence of persistent HPV 31, 33, 45, 52, and 58 infection for a duration of 6 months (within ±1 month windows) or longer and HPV 31-, 33-, 45-, 52-, and 58-related cervical, vulvar, and vaginal cancer, AIS, CIN 2/3, VIN 2/3, VaIN 2/3, CIN 1, VIN 1, VaIN 1, and condyloma acuminata. To be classified as an endpoint case for the primary analysis, a subject must develop at least one of the following after the completion of the Month 7 visit: 1. Persistent HPV 31, 33, 45, 52 or 58 Infection. This endpoint is defined to have occurred if a subject is positive for the same HPV type by the HPV 31/33/45/52/58 PCR assay to at least 1 common gene in 2 or more consecutive V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

75 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: cervicovaginal/external genital swab, biopsy, or definitive therapy samples obtained at least 6 months apart (within ± 1 month windows). Although persistent infection is defined by 2 consecutive samples at least 6 months apart with ±1 month windows, it is possible for a subject to attend a visit 1 month late and a subsequent visit 1 month early, thereby the minimum length of time between the visits will be 4 months for a subject to be counted as a case. If a subject has 2 consecutive samples that are PCR positive to at least one common gene for the same HPV type and at least one of the samples is a biopsy or definitive therapy sample showing pathologic evidence of HPV disease, then the subject will be considered a case of persistent infection without regard to the length of time between the samples. 2. External genital warts, vulvar intraepithelial neoplasia (VIN), vaginal intraepithelial neoplasia (VaIN), vulvar cancer, or vaginal cancer related to HPV 31, 33, 45, 52 or 58. This endpoint is defined to have occurred if on a single biopsy or excised tissue, there is: (a) a HPV Vaccine Program Pathology Panel consensus diagnosis of genital wart, VIN (any grade), VaIN (any grade), vulvar cancer, or vaginal cancer; AND (b) detection of at least 1 of HPV types 31, 33, 45, 52 or 58 by Thinsection PCR in an adjacent section from the same tissue block. 3. Cervical intraepithelial neoplasia (CIN), Adenocarcinoma in Situ (AIS), or cervical cancer related to HPV 31, 33, 45, 52 or 58. This endpoint is defined to have occurred if on a single cervical biopsy, ECC, LEEP or Conization (cold knife/laser) specimen, there is: (a) a HPV Vaccine Program Pathology Panel consensus diagnosis of CIN (any grade), AIS, or cervical cancer; AND (b) detection of at least 1 of HPV types 31, 33, 45, 52 or 58 by Thinsection PCR in an adjacent section from the same tissue block. The primary efficacy endpoint includes all of the above. The incidence of each individual infection and disease component of the primary endpoint will also be reported. An additional efficacy endpoint of interest that is similar to the primary endpoint of HPV 31/33/45/52/58-related persistent infection for a duration of 6 months (within ±1 month windows) or longer or clinical disease will be evaluated in a sensitivity analysis. The difference between the endpoint to be used in this sensitivity analysis and the primary efficacy endpoint is in the definition of HPV 31/33/45/52/58-related cervical, vaginal, or vulvar disease (#2 and #3 of the endpoint definition). In the sensitivity analysis, a case of disease will require a consensus diagnosis of HPV disease by the pathology panel, the appropriate HPV type detected in an adjacent section of the same tissue block AND at least 1 specimen immediately prior to or subsequent to the biopsy or definitive therapy sample that is positive to the same HPV type that is found in the biopsy. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

76 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: A secondary efficacy case of HPV 31-, 33-, 45-, 52-, or 58-related documented infection for 12 months or HPV 31-, 33-, 45-, 52-, or 58-related disease is defined in the same manner as the primary endpoint with the following exceptions: 1. a subject may be defined as a case at any time after the Day 1 visit; and 2. the documented infection for 12 months component is defined to have occurred if a subject is positive for the same HPV type by the HPV 31/33/45/52/58 PCR assay to at least 1 common gene in 2 or more consecutive cervicovaginal/external genital swab, biopsy, or definitive therapy samples obtained for over a period of at least 12 months. A secondary efficacy case of HPV 16/18-related persistent infection for 6 months (within ±1 month windows) or longer or related clinical disease is defined as a subject who, after the completion of the Month 7 visit, is found to have an incident case of persistent HPV 16 or 18 infection, or HPV 16/18-related disease. The definitions of occurrence of HPV infection or disease are the same as those given for the primary efficacy case, except in relation to HPV types 16 and 18 instead of types 31, 33, 45, 52 and 58. A secondary case of HPV 6/11-related disease is defined as a subject who, after the completion of the Month 7 visit, is found to have a biopsy or definitive therapy specimen with (a) a HPV Vaccine Program Pathology Panel consensus diagnosis of cervical, vulvar, or vaginal disease (as detailed in #2 and #3 of the primary endpoint definition); and (b) detection of at least 1 of HPV types 6 or 11 by Thinsection PCR in an adjacent section from the same tissue block. An additional secondary efficacy endpoint of interest is the incidence of Pap test abnormalities, a case of which is defined as a subject who, after completion of the Day 1 visit, is found to have a Pap test result of atypical squamous cells of undetermined significance (ASC-US) with positive HPV probe or worse (ASC-US [positive for High Risk HPV], atypical squamous cells suggestive of a high-grade intraepithelial lesion [ASC-H], HSIL, LSIL, atypical glandular cells, or adenocarcinoma in situ [AIS]). An exploratory efficacy endpoint of interest is the combined incidence of HPV 35/39/51/56/59-related persistent infection for a duration of 6 months (within ±1 month windows) or longer and related clinical disease. To be classified as a case for this exploratory analysis, a subject must develop an incident case of persistent HPV 35-, 39-, 51-, 56-, or 59-related infection or HPV 35/39/51/56/59-related disease after completion of the Day 1 visit. Other exploratory efficacy endpoints of interest include (a) the combined incidence of CIN, AIS, and cervical cancer caused by any (vaccine or non-vaccine) HPV type; (b) the combined incidence of vulvar and vaginal disease caused by any HPV type; (c) the combined incidence of Pap test abnormalities (ASC-US [positive for High Risk HPV] or worse) potentially related to HPV 31/33/45/52/58; and (d) the incidence of cervical biopsy and cervical definitive therapy procedures. A case of all CIN, AIS, or cervical V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

77 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: cancer caused by any HPV type is defined as a subject who, after completion of the Day 1 visit, is found to have a consensus diagnosis of CIN (any grade), AIS, or cervical cancer on a single cervical biopsy, ECC, LEEP or Conization (cold knife/laser) specimen. A subject is defined to have a case of vulvar or vaginal disease caused by any HPV type if, after completion of the Day 1 visit, she is found to have a consensus diagnosis of genital wart, VIN (any grade), VaIN (any grade), vulvar cancer, or vaginal cancer on a single biopsy or excised tissue. A subject is defined to have a case of ASC-US [positive for High Risk HPV] or worse potentially related to HPV 31/33/45/52/58 if, after completion of the Month 7 visit, she is found to have a Pap test diagnosis of ASC-US [with positive for High Risk HPV ], atypical squamous cells suggestive of a high-grade intraepithelial lesion [ASC-H], HSIL, LSIL, atypical glandular cells, or adenocarcinoma in situ [AIS], and at least one cervicovaginal/external genital swab that is PCR positive for HPV 31, 33, 45, 52, or 58 at the same study visit. A subject is defined to have an incident cervical biopsy or definitive therapy procedure if she has a cervical biopsy or definitive therapy after completion of the Day 1 visit. For all efficacy analyses, if a subject has experienced one or more of the components of the respective composite endpoint, she will be classified as a single case for the analysis at the time of detection of the first endpoint. She will be classified as a noncase only if she has experienced none of the components of the respective composite endpoint. A subject will be counted as a case at most once in each efficacy analysis. For analyses in which cases of the composite endpoint are further classified by each individual infection or disease component, a subject will be counted as a case at most once in each applicable sub-category but may appear in multiple sub-categories (e.g., persistent HPV 31 infection and HPV 31-related CIN 2) Immunogenicity Part A The primary immunogenicity endpoints in Part A are geometric mean titers (GMTs) to HPV 6, 11, 16, and 18 at Week 4 Postdose 3. Other important endpoints are geometric mean titers (GMTs) to HPV 31, 33, 45, 52 and 58 at Week 4 Postdose 3. Part B The primary immunogenicity endpoints in Part B are geometric mean titers (GMTs) to HPV 6, 11, 16, and 18 at Week 4 Postdose 3. The secondary immunogenicity endpoints are a) the seroconversion percentages to each of HPV 6, 11, 16, and 18 by Week 4 Postdose 3; and b) the percentages of subjects who seroconvert for each HPV type (31, 33, 45, 52 and 58) by Week 4 Postdose 3. Seroconversion is defined as changing serostatus from seronegative to seropositive, by Week 4 Postdose 3. A subject with a clia titer at or above the serostatus cutoff for a given HPV type is considered seropositive for that type. Additional secondary immunogenicity endpoints include the geometric mean titers (GMTs) for HPV 31, 33, V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

78 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: , 52 and 58 at Week 4 Postdose 3 and the immune responses for HPV 6, 11, 16, 18, 31, 33, 45, 52, and 58 at the persistence time points (Months 12, 24, 36, and 42). The exploratory immunogenicity endpoints are the geometric mean titers (GMTs) to HPV 6 and HPV 11 measured in peripartum maternal blood and cord blood of infants born to women vaccinated at sites participating in cord blood data collection Analysis Populations Safety All subjects who received at least 1 study vaccination and have follow-up data will be included in the primary analysis of safety Efficacy Primary Per Protocol Efficacy Analysis The approach for the primary efficacy hypothesis will be per-protocol. To be eligible for endpoints related to HPV 31, 33, 45, 52, and/or 58, subjects must be seronegative to the respective HPV type at Day 1 and PCR negative to the respective HPV type on all cervicovaginal swabs and biopsies from Day 1 through Month 7 and must have Month 7 PCR swab samples collected within an acceptable day range. In addition, to be included in the primary efficacy evaluation, subjects must have received all 3 vaccinations with the correct clinical material within 1 year, and have 1 or more follow-up visits following Month 7. The primary approach for the secondary efficacy analysis of the combined incidence of HPV 16/18-related persistent infection for a duration of 6 months (within ±1 month windows) or longer and clinical disease will also be per protocol. To be eligible for the secondary analysis of efficacy of endpoints related to HPV 16 and/or 18, subjects must be seronegative to the respective HPV type at Day 1 and PCR negative to the respective HPV type on all cervicovaginal swabs and biopsies from Day 1 through Month 7. To be eligible for the secondary per protocol efficacy analysis of HPV 6/11-related clinical disease, subjects will be required to be seronegative at Day 1 and PCR negative from Day 1 through Month 7 for both HPV 6 and HPV 11. Modified Intent-to-Treat Analyses Two modified intent-to-treat (MITT) analyses will be performed as supporting analyses to the primary per protocol efficacy analysis, to explore the robustness of the vaccine efficacy and to evaluate the secondary efficacy hypothesis. Subjects who received at least 1 vaccination and have any follow-up visit following the first vaccination will be included in both populations. An efficacy case will be defined for the primary analysis as a subject with an incident persistent HPV 31, 33, 45, 52 or 58 infection, or HPV 31/33/45/52/58-related disease after the Day 1 visit (i.e. post dose 1). The populations are defined as follows: V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

79 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: HPV Type-Specific Naïve Population - This population will include only subjects who are seronegative and PCR-negative at enrollment to the appropriate HPV types, as defined in the Primary Per Protocol Analysis. 2. Full Analysis Set - This population will not be restricted to subjects who are seronegative and PCR-negative at enrollment to the appropriate HPV types, as defined in the Primary Per Protocol Analysis. This analysis in essence includes all subjects randomized, except those who did not receive any vaccination or those without any follow-up. The HPV Type-Specific Naïve Population approach will be the primary approach for the secondary efficacy analysis of HPV 31/33/45/52/58-related documented infection for 12 months and clinical disease. The MITT analyses will also be performed as supportive analyses for both (a) the secondary efficacy analysis of the combined incidence of HPV 16/18-related persistent infection for 6 months (within ±1 month windows) and clinical disease; and (b) the secondary efficacy analysis of HPV 6/11-related clinical disease. Additional secondary and exploratory analyses will be conducted in the All HPV Naïve Population, defined as subjects who (a) are seronegative and PCR-negative at enrollment to all 9 vaccine HPV types; (b) are PCR-negative at enrollment to all other non-vaccine HPV types for which PCR assays will be available; and (c) have a normal Pap test result at enrollment. The All HPV Naïve Population approach will be the approach for (a) the secondary efficacy analysis of the incidence of Pap test abnormalities; (b) the exploratory efficacy analysis of the combined incidence of all CIN, AIS, and cervical cancer caused by any HPV type; (c) the exploratory efficacy analysis of the combined incidence of all vulvar and vaginal disease caused by any HPV type; (d) the exploratory efficacy analysis of the combined incidence of HPV 35/39/51/56/59-related persistent infection for a duration of 6 months (within ±1 month windows) or longer and clinical disease; and (e) the exploratory efficacy analysis of the incidence of cervical biopsy and cervical definitive therapy procedures. The primary approach for the exploratory efficacy analysis of the incidence of Pap test abnormalities (ASC-US [positive for High Risk HPV] or worse) potentially related to HPV 31/33/45/52/58 will be per protocol Immunogenicity The primary approach to the analyses of immunogenicity will also be per protocol (immunogenicity). Each vaccine component will be analyzed separately. To be included in the primary immunogenicity analysis for the HPV 6 and HPV 11 components, subjects must be seronegative to both HPV 6 and 11 at Day 1 and must be PCR negative to HPV 6 and 11 from Day 1 through Month 7. To be included in the primary immunogenicity analysis for the other vaccine HPV types, subjects are required to be seronegative at Day 1 and PCR negative from Day 1 through Month 7 only for the HPV type being analyzed. In addition, subjects must receive all 3 doses of the correct clinical material within V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

80 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: acceptable day ranges, and must have at least 1 serology result Postdose 3 within acceptable day ranges. A supportive immunogenicity analysis will be carried out on the all type-specific naïve subjects with serology population. To be included in this analysis subjects must be seronegative and PCR negative to the relevant HPV type(s) as described above, and must have provided serology data Statistical Methods Efficacy To address the primary efficacy hypothesis, a one-sided test of the null hypothesis that the vaccine efficacy is 20% will be conducted. The alternative hypothesis states that the vaccine efficacy is > 20%. The statistical criterion for success with respect to the primary efficacy hypothesis requires that the lower bound of the confidence interval for vaccine efficacy exclude 20% for the primary efficacy endpoint. The primary efficacy hypothesis will be tested at the α=0.025 (1-sided) level. Vaccine efficacy is defined as: VE = 100%*{1-(r N /r G )} where r N, the incidence rate among 9-valent HPV L1 VLP vaccine recipients, is defined as r N = C N /τ N. C N = number of primary efficacy cases among 9-valent HPV L1 VLP vaccine recipients and τ N = total person-years of follow-up among 9-valent HPV L1 VLP vaccine recipients. Similarly, r G = C G /τ G is the incidence rate among GARDASIL recipients, where C G is the number of primary efficacy cases among GARDASIL recipients and τ G is the total person-years of follow-up among GARDASIL recipients The null hypothesis that vaccine is not efficacious (i.e., VE 20%) will be tested by constructing a two-sided exact confidence interval for VE. Under the assumption that r N and r G are the means of independent Poisson processes, and given that there are a total of n = C N + C G primary efficacy cases observed on all subjects, the number of primary efficacy cases C N among 9-valent HPV L1 VLP vaccine recipients is distributed as Binomial(n,p), where the binomial probability p is defined as p= τ N r N /(τ N r N +τ G r G ). The probability p is a person-years-adjusted estimate of the probability that a given primary efficacy case is a 9-valent HPV L1 VLP vaccine recipient. The lower bound of the 100*(1 α)% exact confidence interval for the probability p is obtained by searching for the proportion p L such that the probability of observing C N or more primary efficacy cases out of n total primary efficacy cases is α/2. Similarly, the upper bound of the 100*(1 α)% exact confidence interval for the probability p is obtained by searching for the proportion p U such that the probability of observing C N or fewer primary efficacy cases out of n total primary efficacy cases is α/2 [26]. The upper and lower bounds of the 100*(1 α)% confidence interval for the vaccine efficacy can be computed from the upper and lower bounds of the confidence interval for p. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

81 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: No statistical testing will be carried out for the incidence of individual infection or disease components, although descriptive statistics, including 95% confidence intervals will be provided. To address the secondary efficacy hypothesis concerning the combined incidence of HPV 31/33/45/52/58-related documented infection for 12 months and clinical disease, a onesided test of the null hypothesis that the vaccine efficacy is 0% will be conducted using the same methodology as for the primary efficacy hypothesis. The statistical criterion for success with respect to the secondary efficacy hypothesis requires that the lower bound of the 95% confidence interval for vaccine efficacy exclude 0% for the secondary efficacy endpoint. To address the secondary objective concerning the combined incidence of infection and disease related to HPV types 16 and 18 in subjects who receive 9-valent HPV L1 VLP vaccine as compared to GARDASIL, the incidence of HPV 16/18-related infection and disease and corresponding 95% confidence interval will be summarized for each vaccination group. For comparison, the incidence of HPV 16/18-related infection and disease will also be presented for the GARDASIL and placebo groups from a historic cohort of subjects in Protocols 007 and 012 from the Phase II/III placebo-controlled efficacy studies for GARDASIL. The incidence of HPV 16/18-related persistent infection and disease in the 9-valent HPV L1 VLP vaccine group will be observationally compared to the incidence in the placebo group in the historic cohort. Within the historic cohort of Protocols 007 and 012, the incidence of HPV 16/18-related persistent infection and disease will also be presented for the GARDASIL and placebo groups for a subset of subjects whose age, HPV 6, 11, 16, and 18 infection status at baseline, and HPV 6, 11, 16, 18 serostatus at baseline are similar to the those in the current study. This comparison group will be used to estimate the vaccine efficacy of 9- valent HPV L1 VLP vaccine relative to placebo. Additional baseline characteristics may be considered for defining this comparison group, and further details will be provided in the Statistical Analysis Plan. To provide an estimate of the vaccine efficacy of 9-valent HPV L1 VLP vaccine relative to placebo, an indirect method will be used [27]: VE N/P = 100%* (1-RR N/P ) = 100% {1-(r N /r G * r G /r P )} where r N /r G = 1-VE[current study], is the observed relative risk of 9-valent HPV L1 VLP vaccine to GARDASIL from the current study, and r G /r P = 1-VE[historic cohort] is the observed relative risk of GARDASIL to placebo from the historic cohort. A 95% confidence interval for VE N/P will be estimated from the 95% confidence interval for the natural log of the relative risk RR N/P. For the confidence interval calculation, the following estimate of the variance of RR N/P will be used: Var(Ln(RR N/P ))= Var(Ln(r N /r G )) + Var(Ln(r G /r P )) V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

82 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: A similar approach will be used to address: a) the secondary objective evaluating the impact of the 9-valent HPV L1 VLP vaccine on the incidence of HPV 6/11-related disease; b) the exploratory objective evaluating the impact of the 9-valent HPV L1 VLP vaccine on the combined incidence of all CIN, AIS, and cervical cancer due to any HPV type, and c) the combined incidence of all vulvar and vaginal disease due to any HPV type. An approach that is similar to the primary and secondary efficacy analyses will be used to address (a) the secondary efficacy objective concerning the incidence of Pap test abnormalities; (b) the exploratory objective concerning the combined incidence of HPV 35/39/51/56/59-related persistent infection and disease; (c) the exploratory efficacy objective concerning the incidence of Pap test abnormalities (ASC-US [positive for High Risk HPV] or worse) potentially related to HPV 31/33/45/52/58; and (d) the exploratory efficacy objective concerning the incidence of cervical biopsy and cervical definitive therapy procedures. No hypothesis testing will be conducted Immunogenicity Part A The hypotheses of noninferiority of GMTs for HPV types 6, 11, 16 and 18 will be based on one-sided tests of non-inferiority comparing GMTs for each component. Four ANOVA models (one per HPV type) with a response of log individual titers and a fixed effect for vaccination group will be used. The hypotheses to be tested are: H 0 : GMT N /GMT G 1/2 versus H 1 : GMT N /GMT G > 1/2, where GMT G represents the GMTs in subjects receiving GARDASIL and GMT N represents the GMTs in subjects receiving the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B. Part A is considered a pilot study, and the statistical criterion for noninferiority in these tests corresponds to the lower bound of the confidence interval for the fold-difference in GMTs between the 2 groups, (9-valent HPV L1 VLP vaccine group/gardasil group), excluding a decrease of 2-fold or more for each component. Part B The hypotheses of noninferiority of GMTs for HPV types 6, 11, 16 and 18 will be based on one-sided tests of non-inferiority comparing GMTs for each component. Four ANOVA models (one per HPV type) with a response of log individual titers and a fixed effect for vaccination group will be used. The hypotheses to be tested are: H 0 : GMT N /GMT G 0.67 versus H 1 : GMT N /GMT G > 0.67, where GMT G represents the GMTs in subjects receiving GARDASIL and GMT N represents the GMTs in subjects receiving the 9-valent HPV L1 VLP vaccine. Each hypothesis will be tested at the α=0.025 level (1-sided). Because the confirmatory immunogenicity evaluation of the selected 9-valent HPV L1 VLP vaccine formulation will be conducted in Part B, the statistical criterion for non-inferiority in these tests corresponds to the lower bound of the 95% confidence interval for the fold-difference in GMTs between the 2 groups, (9-valent HPV L1 VLP vaccine group/gardasil group), excluding a decrease of 1.5-fold or more for each component. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

83 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: The secondary hypothesis of noninferiority of seroconversion percentages for each of HPV types 6, 11, 16, and 18 will be addressed by 4 one-sided tests of noninferiority (one corresponding to each HPV type) conducted at the α=0.025 level (1-sided). For each HPV type, the hypotheses to be tested are H 0 : p 1 -p H a : p 1 -p 2 > where p 1 is the proportion of subjects who seroconvert at Week 4 Postdose 3 in recipients 9-valent HPV L1 VLP vaccine and p 2 is the proportion of subjects who seroconvert at Week 4 Postdose 3 in GARDASIL recipients. The tests above will be conducted based on methods developed by Miettinen and Nurminen [28] for testing the equivalence of 2 proportions. The secondary hypothesis of acceptability of anti-hpv seroconversion rates (for HPV types 31, 33, 45, 52 and 58) will be tested by one-sided tests of the proportions of subjects seroconverting (1 test per HPV type). These tests will be conducted based on the exact 95% confidence interval for a binomial proportion. The hypotheses to be tested are: H 0 : P N 0.9 versus H 1 : P N > 0.9 (for each of the HPV Types 31, 33, 45, 52, and 58), where P N represents the true response rate of subjects receiving a 3-dose regimen of a formulation of 9-valent HPV L1 VLP vaccine. Rejection of the null hypothesis in these tests corresponds to the lower bound of the 95% confidence interval for the percentage of subjects seroconverting in the 9-valent HPV L1 VLP vaccine group being greater than 90% for the given HPV type. The primary time point at which immune response will be evaluated is Week 4 Postdose 3. Anti-HPV responses for each of the 9 vaccine HPV types will be summarized by vaccination groups in terms of geometric mean titers (GMTs) with 95% confidence intervals at each time point when serum samples are collected. Anti-HPV responses during the period of vaccination and persistence from Month 7 onwards will be investigated using longitudinal plots of the GMTs. Reverse cumulative distribution (RCD) plots and a summary of seropositivity percentages, including 95% confidence intervals, will be presented for each of the 9 vaccine HPV types at all time points at which serum samples are collected. The geometric mean titers (GMTs) to HPV 6 and HPV 11 in peripartum maternal blood and cord blood of infants born to women vaccinated at sites participating in cord blood data collection will be summarized descriptively Safety Safety and tolerability will be assessed by statistical and clinical review of all safety data collected throughout the study. All subjects who are vaccinated and who have safety follow-up data will be included in the safety analyses and summaries. All safety analyses and summaries will be provided separately for: (1) all subjects in Part A; and (2) all V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

84 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: subjects in Part B combined with subjects from Part A who received the selected 9-valent HPV L1 VLP vaccine formulation or GARDASIL. To provide an overall assessment, safety measures such as the incidence of 1. any adverse experiences; 2. any injection-site experiences; 3. any systemic adverse experiences; and 4. any vaccine-related adverse experiences that occurred throughout the study will be summarized in both groups. Adverse experiences will be summarized as frequencies and percentages by vaccination group by vaccination visit and across all vaccination visits. To assess the risks of adverse experiences temporally associated with vaccination, a multi-tiered approach will be used for the analysis of safety parameters. Tier-1 adverse experiences include (1) injection-site adverse experiences prompted for on the VRC, such as redness, swelling, and pain/tenderness/soreness occurring Day 1 through Day 5 following any vaccination, and (2) elevated temperature ( F [ 37.8ºC]), from Day 1 to Day 5 following any vaccination. For Tier-1 adverse experiences, the risk difference between each 9-valent HPV L1 VLP vaccination group and GARDASIL in Part A, the corresponding two-sided 95% CI on the risk difference, and the p-value for the test of significance of the risk difference will be provided. For the analyses combining data from Part A and Part B, the risk difference between the selected 9-valent HPV L1 VLP vaccination group and the GARDASIL group, 95% CI on the risk difference, and the p-value for the test of significance will also be provided. The corresponding p-values will be based on the asymptotic normal approximation for testing two independent binomial proportions at the two-sided 0.05 level. All risk differences and 95% CIs will be calculated using the methods proposed by Miettinen and Nurminen [28] The Tier-2 adverse experience summaries include (1) specific systemic adverse experiences within 14 days following any vaccination occurring in 1% of subjects in any vaccination group, (2) injection-site adverse experiences not prompted for on the VRC occurring Day 1 to Day 5 following any vaccination in 1% of subjects in any vaccination group, (3) serious adverse experiences occurring within 14 days following any vaccination, (4) serious vaccine-related adverse experiences observed at any time during the study, and (5) severe injection-site adverse experiences Day 1 through Day 5 following any vaccination visit. Risk differences and 95% confidence intervals between each 9-valent HPV L1 VLP vaccination group compared to the GARDASIL group for all Part A subjects will be estimated for all Tier-2 adverse experiences using the methodology proposed by Miettinen and Nurminen [28]. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

85 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Tier-3 adverse experiences will include summaries (counts and proportions) by vaccination group for any other adverse experiences, including all injection-site adverse experiences occurring from Day 1 to Day 5 following each vaccination visit and all systemic adverse experiences occurring within 14 days of each vaccination visit Multiplicity Considerations Part A Success in Part A will be declared if a formulation of 9-valent HPV VLP vaccine is selected for use in Part B. For the Month 7 hypothesis test of noninferiority of GMTs of the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B compared with GARDASIL, multiplicity with respect to the multiple hypotheses for HPV types 6, 11, 16 and 18 is addressed by requiring success for all 4 HPV types. The hypothesis for each of HPV types 6, 11, 16, and 18 will be tested at the nominal 1-sided α= level (with minor adjustment for the interim summary). As described in section 3.5.1, there will be an interim summary of immunogenicity in the study when ~100% of all serology data are available through Week 4 Postdose 2 for subjects in Part A. There will be no formal hypothesis testing done at the time of the interim analysis. Because it is recognized that informal reviews of unblinded data compromise the integrity of the final analysis of the data in Part A, a nominal 1-sided alpha level of will be used to test the hypothesis at Month 7 for each of HPV types 6, 11, 16, and 18 as if an extreme alpha-spending rule had been used for the interim summary, such as the Haybittle-Peto rule [29]. This would theoretically apply a nominal significance level of to any interim summary (even though no hypothesis testing will be conducted). The use of the slightly reduced nominal alpha level at the final analysis in Part A for hypothesis testing will protect the overall study type 1 error at not more than 5% 2-sided. Part B Efficacy Because no efficacy data will be summarized for the interim analysis in Part A, no multiplicity adjustment will be made to account for the subjects from Part A that will be included in the efficacy evaluation for Part B. Immunogenicity The primary immunogenicity analysis in Part B will be conducted independently from the analysis in Part A. Thus, no multiplicity adjustment is required. Multiplicity with respect to the multiple hypotheses for HPV types 6, 11, 16 and 18 is addressed by requiring success for all 4 HPV types thereby controlling the overall alpha level. Thus, no multiplicity adjustment is needed for the overall immunogenicity hypotheses. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

86 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Success in this study will be declared if the primary efficacy hypothesis and the primary immunogenicity hypothesis in Part B are achieved. Thus, no multiplicity adjustment is needed to control the overall type I error rate for the study. The supplementary analyses of efficacy to be carried out when full follow-up is complete will not require a multiplicity adjustment since no inference will be drawn at that time Sample Size and Power Calculations Efficacy It is expected that, for each of HPV 31, 33, 45, 52 and 58, no more than 23%, 22%, 16%, 22%, and 20%, respectively, of women aged years will be seropositive at Day 1 and/or PCR positive between Day 1 and Month 7 for any given type. This was based on previous studies in the GARDASIL program and the 8-valent HPV L1 VLP (V502) vaccine program. Assuming no more than 10% attrition between Day 1 and Month 7, at least 2,180 of the 3,100 women enrolled will be eligible for follow-up in the primary population for at least one of the 5 vaccine types not found in GARDASIL. Assuming that annual attrition post-month 7 is not greater than 5%, and that subjects who drop out after Month 7 will have an average follow-up that is at least half that of those subjects who complete, the total follow-up for the first 2 years of the study of eligible subjects will be >2,800 personyears (combined across vaccination groups). In planning the study, the phenomenon of cross-protection was considered. It is possible that the HPV types contained in GARDASIL may afford some protection against infection and disease related to other HPV types, and in particular HPV types 31, 33, 45, 52 and 58. This may reduce the incidence of the primary endpoint in subjects receiving GARDASIL, relative to completely untreated subjects. Based on existing crossprotection data from the GARDASIL program with respect to HPV infection and related disease, it has been assumed that this reduction may be up to 40% [30]. Based on existing data from the GARDASIL program, the assumed annual incidence rates for HPV 31-, 33-, 45-, 52-, and 58-related persistent infection or clinical disease are 0.78%, 0.48%, 0.42%, 1.19%, and 0.77%, respectively. To avoid uncertainty inherent in assumed event rates, this trial will employ a fixed-event design. The primary analysis will not be conducted until at least 43 subjects in the perprotocol efficacy population have developed a case of HPV 31//33/45/52/58-related persistent infection or disease. Assuming that the true efficacy of the vaccine is 75% (which is reduced relative to the estimated vaccine efficacy of GARDASIL because of the crossprotection phenomenon), 43 cases of the primary efficacy endpoint will provide at least 90% power to declare the vaccine efficacious (α=0.025, one-sided). With 43 total cases across both vaccination groups, 12 or fewer cases in the 9-valent HPV L1 VLP vaccine group will result in a conclusion that the vaccine is efficacious. If it appears that 43 cases of the primary efficacy endpoint will not be observed prior to the completion of V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

87 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Month 42 follow-up in all subjects, consideration will be given to extending the duration of follow-up in all subjects by protocol amendment. The sample size for the efficacy evaluation is driven by the secondary efficacy endpoint. Assuming that the true efficacy of the vaccine is 75% (which is reduced relative to the estimated vaccine efficacy of GARDASIL because of the crossprotection phenomenon), 30 cases of the primary efficacy endpoint will provide at least 90% power to declare the vaccine efficacious (α=0.025, one-sided). To ensure at least 30 cases persistent HPV 31/33/45/52/58-related infection for 12 months or clinical disease are observed by Month 24, approximately 3,100 subjects are needed for the efficacy evaluation (2,480 subjects enrolled for Part B combined with the ~620 subjects in the GARDASIL arm and selected dose formulation arm from Part A). This calculation assumes the same pre-positivity rates for each HPV type and attrition as in the primary efficacy analysis, and assumed annual event rates of 0.39%, 0.24%, 0.21%, 0.60%, and 0.38% for HPV 31-, 33-, 45-, 52-, and 58-related documented infection for 12 months and clinical disease, respectively. It is anticipated that the secondary efficacy analysis will be conducted at the time of the primary efficacy analysis. If it appears that fewer than 30 cases of the secondary endpoint will be observed at the time of the primary efficacy analysis, the vaccine efficacy against the secondary endpoint will be summarized, however, the secondary hypothesis will not be tested until at least 30 cases are observed. Immunogenicity Part A With the enrollment of 310 subjects per group (1,240 subjects enrolled total) to obtain around 209 to 230 evaluable subjects per arm and assuming a nominal α= level (1- sided), this study has an overall power of >99%. The calculations are based on the following assumptions: 1. 15%, 15%, 18% and 10% of the subjects are initially seropositive or PCR positive to HPV Types 6, 11, 16, and 18, respectively. 2. The expected attrition rate through the Month 7 time point is 10%. 3. The expected percentage of subjects excluded due to vaccinations or serology samples out of range is no more than 8% 4. The standard deviations of the natural logarithm of the Month 7 titers are no more than 1.2 for each HPV type when expressed as mmu/ml, the unitage used in previous GARDASIL studies. 5. HPV Types 6, 11, 16, and 18 responses are identical in the GARDASIL group and at least one 9-valent HPV L1 VLP vaccine formulation. The expected responses rates and noninferiority margins are as shown in Table 3-5 for each antigen. For GARDASIL, these are based on data available from previous studies. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

88 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Table 3-5 Power for Part A Immunogenicity Hypotheses of Protocol V Antigen Parameter Expected rates/sd Non-inferiority Margin N evaluable Power (1) Primary Hypothesis - Non-Inferiority of 9-valent HPV L1 VLP Vaccine to GARDASIL HPV 6 GMT σ = fold decrease 218 >0.99 HPV 11 GMT σ = fold decrease 218 >0.99 HPV 16 GMT σ = fold decrease 209 >0.99 HPV 18 GMT σ = fold decrease 230 >0.99 Overall Power for GMT hypothesis >0.99 Part B The overall sample size in the study is determined by the efficacy analyses, and the primary immunogenicity analysis requires fewer than the 2,480 subjects that will be enrolled in Part B. With approximately 750 subjects per group to obtain 508 to 558 evaluable subjects per arm, this study has >99% power for the primary immunogenicity hypothesis. In order to ensure that a representative sample is obtained throughout all of the study sites in Part B, the first ~60% of subjects randomized in each site in Part B will be included in the primary immunogenicity analysis. This study has 99.6% power for the secondary immunogenicity hypothesis of noninferiority with respect to seroconversion percentages for HPV types 6, 11, 16, and 18. The calculations are based on the same assumptions as in the Part A immunogenicity analysis. This study also has >99% power for the secondary immunogenicity hypothesis of acceptability with respect to seroconversion percentages for HPV types 31, 33, 45, 52, and 58. For the secondary seroconversion hypothesis, it is assumed that for each of HPV 31, 33, 45, 52 and 58, no more than 23%, 22%, 16%, 22%, and 20%, respectively, are initially seropositive or PCR positive for any given type. All statistical tests will be conducted at the α=0.025 level (1-sided). The expected standard deviations or assumed responses rates and power for the primary immunogenicity hypotheses are as shown in Table 3-6 for each antigen. For GARDASIL, these are based on data available from previous studies (Protocol V , 011, 012, 015 and 016, Merck & Co., Inc. data on file). The expected response rates and power for the secondary immunogenicity hypotheses are shown in Table 3-7. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

89 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Table 3-6 Power for Part B Immunogenicity Hypotheses of Protocol V Antigen Parameter Expected rates/sd Non-inferiority Margin N evaluable Power (1) Primary Hypothesis - Non-Inferiority of 9-valent HPV L1 VLP Vaccine to GARDASIL HPV 6 GMT σ = fold decrease 527 >0.99 HPV 11 GMT σ = fold decrease 527 >0.99 HPV 16 GMT σ = fold decrease 508 >0.99 HPV 18 GMT σ = fold decrease 558 >0.99 Overall Power for the primary immunogenicity hypotheses >0.99 Measured by Competitive Luminex immunoassay (clia). If there exists a modest interference on the Month 7 immune responses for HPV 6, 11, 16 and 18 in the 9-valent HPV L1 VLP vaccine dose formulation, i.e., GMT N /GMT G =0.9, then the study will have an overall power of ~93%. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

90 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Table 3-7 Power for Secondary Immunogenicity Hypothesis of Protocol V Antigen Parameter Expected rates/sd Non-inferiority Margin N evaluable Power (1) Secondary Immunogenicity Hypothesis - Non-Inferiority of 9-valent HPV L1 VLP Vaccine to GARDASIL HPV 6 Titer Seropositive Cutoff 98% 5% HPV 11 Titer Seropositive Cutoff 98% 5% HPV 16 Titer Seropositive Cutoff 98% 5% HPV 18 Titer Seropositive Cutoff 98% 5% 558 >0.99 Overall Power for the seroconversion hypothesis (2) Secondary Hypothesis - Acceptability of 9-valent HPV L1 VLP Vaccine HPV 31 Titer Seropositive Cutoff 98% Lower 477 >0.99 bound>90% HPV 33 Titer Seropositive Cutoff 98% Lower 483 >0.99 bound>90% HPV 45 Titer Seropositive Cutoff 98% Lower 521 >0.99 bound>90% HPV 52 Titer Seropositive Cutoff 98% Lower 483 >0.99 bound>90% HPV 58 Titer Seropositive Cutoff 98% Lower bound>90% 496 >0.99 Overall Power for the secondary immunogenicity hypothesis >0.99 Measured by Competitive Luminex immunoassay (clia). Seropositivity cutoff values for HPV 6, 11, 16, 18, 31, 33, 45, 52, 58 will be determined. Safety The probability of observing at least 1 SAE in this study depends on the number of subjects enrolled and the incidence rate of SAEs in the general population. If the incidence rate of an SAE is 1 of every 1240 (0.08%) subjects who received any formulation of 9-valent HPV L1 VLP vaccine in Part A, then there is a 53% chance of observing at least 1 such SAE among the 930 subjects who receive any of the 3 formulations of 9-valent HPV L1 VLP vaccine in Part A. If the incidence rate of an SAE is 1 of every 2880 (0.03%) subjects who received a formulation of 9-valent HPV L1 VLP vaccine, then there is a 28% chance of observing at least 1 such SAE among the 930 subjects who receive any of the 3 formulations of 9-valent HPV L1 VLP vaccine in Part A. If no SAEs are observed in the 930 subjects who received a 9-valent HPV L1 VLP V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

91 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: vaccine formulation in Part A, then this study will provide 95% confidence that the true incidence rate for SAEs is <0.4%. In Part B, there is a 71% chance of observing at least 1 SAE among the 1550 subjects who will receive the selected 9-valent HPV L1 VLP vaccine formulation if the incidence rate of an SAE is 1 of every 1280 subjects (0.08%). If the incidence rate is 1 of every 2880 recipients (0.03%), then there is a 42% chance of observing at least 1 SAE among the subjects who will receive the selected 9-valent HPV L1 VLP vaccine formulation. If no SAEs are observed in 1550 subjects, this study will provide 95% confidence that the true incidence rate for SAEs is <0.2% Interim Analyses Part A A preliminary summary of tolerability and immunogenicity data will be produced, by unblinded Merck & Co., Inc. personnel, when ~100% of subjects have Week 4 Postdose 2 serology data available. The interim immunogenicity analysis in Part A will be conducted in a subset of subjects who 1) are seronegative at Day 1 to the relevant HPV type, without regard to PCR status at Day 1; 2) receive the first 2 doses of 9-valent HPV L1 VLP vaccine or GARDASIL in acceptable day ranges; and 3) have postvaccination serum samples collected in acceptable day ranges. This will present summary statistics of immune responses by vaccination group (GMTs, seroconversion rates, confidence intervals, RCDF graphs). No serology or randomization data for individual study participants will be disclosed. No formal hypothesis testing will be done at the time of the interim analysis. It is important to note that the summary will focus on Postdose 2 data, while the study hypotheses concern Week 4 Postdose 3 (Month 7) responses in Part A subjects who receive the 9-valent HPV L1 VLP formulation that is selected for use in Part B or GARDASIL. Part B No interim analyses are planned for Part B of this study. The primary efficacy analysis will be conducted after 43 primary efficacy cases have been observed. Conclusions regarding the success of the study will be drawn from the results of both the primary efficacy analysis and the primary immunogenicity analysis in Part B. Estimates of vaccine efficacy and immunogenicity will be updated after Month 42 follow-up is complete in all subjects, if the primary efficacy analysis is conducted at an earlier time Definition of Compliance Measure Compliance is defined in this study as receipt of all scheduled study vaccinations. To summarize compliance, the numbers of subjects who receive each vaccination in Part A and Part B will be tabulated by vaccination group. Subjects who do not complete the full vaccination regimen will be excluded from the per-protocol primary analyses of efficacy and immunogenicity in Part B, but will be included in the supportive analyses. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

92 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Another important aspect of compliance is the degree of compliance with follow-up examinations during follow-up visits after the vaccination series has been completed. The numbers of subjects who complete each follow-up visit will be tabulated by vaccination group. The numbers of subjects with biopsies or excision procedures performed outside of the study and the numbers of these subjects for whom the tissue samples were not available to Merck & Co., Inc., will be tabulated by vaccination group. Differences in these numbers will be assessed observationally and the potential impact on the efficacy analyses noted. 3.6 LABELING, PACKAGING, STORAGE, DISPENSING, AND RETURN OF CLINICAL SUPPLIES Subject and Replacements Information Clinical supplies will be packaged to support enrollment of approximately 3,720 subjects. Study personnel will have access to an Interactive Voice Response System (IVRS) to allocate patients, to assign study vaccine to patients, and to manage the distribution of clinical supplies. Clinical supplies will be packaged according to a component schedule generated by the SPONSOR. Each person accessing the IVRS system must be assigned an individual unique PIN. They must use only their assigned PIN to access the system and they must not share their assigned PIN with anyone Product descriptions Investigational materials will be provided by the SPONSOR as summarized in Table 3-8. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

93 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Table 3-8 Product Descriptions Product Name and Potency (HPV Types 6/11/16/18/31/33/45/52/58) Dosage Form Comments (if applicable) Quadrivalent HPV VLP Vaccine 20/40/40/20/0/0/0/0/0 mcg/ 0.5mL dose Sterile solution for IM injection. Store at 2-8 C. DO NOT FREEZE. Protect from light. 9-valent HPV VLP Vaccine 20/40/40/20/20/20/20/20/20 mcg/ 0.5 ml dose 9-valent HPV VLP Vaccine 30/40/60/40/20/20/20/20/20 mcg/ 0.5 ml dose 9-valent HPV VLP Vaccine 30/40/80/55/30/30/30/30/30 mcg/ 0.5 ml dose Sterile solution for IM injection Sterile solution for IM injection Sterile solution for IM injection Store at 2-8 C. DO NOT FREEZE. Protect from light. Store at 2-8 C. DO NOT FREEZE. Protect from light. Store at 2-8 C. DO NOT FREEZE. Protect from light. GARDASIL = QUADRIVALENT HPV VLP VACCINE 9-VALENT HPV L1 VLP VACCINE = 9-VALENT HPV VLP VACCINE Primary Packaging and Labeling Information Quadrivalent / 9-valent HPV VLP Vaccine will be labeled as HPV VLP Vaccine and will be packaged in 3 ml glass vials containing approximately 0.75 ml of study vaccine for withdrawal of a single 0.5 ml dose. The appearance of all 9-valent HPV L1 VLP vaccine and GARDASIL vials will be indistinguishable. The tear-off portion of the clinical label will contain at least the packaging identification number and product/protocol. The tear-off portion must be affixed to the Vaccine Inventory and Label LV CRF. Container label text may include the following: V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

94 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: Packaging Control # / Lot Trace ID # Component ID # Dose Volume (0.5 ml Dose for IM injection) Re-evaluation date (If applicable) Product name AN: (If applicable) Dosing Instructions (Administer Per Protocol V ) Storage Conditions Country Regulatory requirements SPONSOR address (If applicable) Translation key (If applicable) Secondary Packaging and Labeling Information- Supplies Will Not be Kitted Clinical Supplies-Vaccine Disclosure IVRS should be used in order to unblind patients and to unmask drug identity. The SPONSOR will not provide disclosure envelopes with the clinical supplies. Vaccine group identification information is to be unmasked ONLY if necessary for the welfare of the patient. Every effort should be made not to unblind the patient unless necessary. Prior to unblinding, the investigator will attempt to contact the clinical monitor. Any unblinding that occurs at the site must be documented Storage Requirements All vaccine supplies will be shipped to the sites as a refrigerated solution to be stored at 2 C to 8 C (35.6 F to 46.4 F). Upon receipt at the investigational site, the vaccine should be removed from the outer secondary shipping box and placed immediately into the refrigerator. The TEMPTALE must be deactivated upon receipt of the shipment. Directions for inactivation are specified in the Instructions to Site, which are enclosed with each shipment. The TEMPTALE will indicate whether the shipment has remained within the specified temperatures. Return the TEMPTALE according to instructions accompanying the shipment. Notify the SPONSOR (Clinical Research Associate [CRA]) and IVRS technical support immediately if the TEMPTALE is in alarm. Store and hold product until instructed otherwise. The clinical supplies storage area at the site must be monitored by the site staff for temperature consistency with the acceptable storage temperature range specified in this protocol or in the product label attached to the protocol. Documentation of temperature monitoring should be maintained. Supplies should be stored in the original nested box with the lid closed to minimize exposure to light. If the refrigerator in which the study vaccine is stored deviates from the 2 C to 8 C (35.6 F to 46.4 F) range, study vaccinations should be suspended and the SPONSOR (CRA) and IVRS technical support should be contacted immediately. Vaccine must NOT be frozen. It is strongly recommended that a non-frost free laboratory grade refrigerator is used to store the study vaccine. This type of refrigerator is less likely to have wide temperature fluctuations, so it will be more likely to stay within the 2 C to 8 C (35.6 F to 46.4 F) temperature range. A daily refrigerator temperature log must be maintained at the site. The refrigerator must be equipped with an appropriately calibrated min/max thermometer and/or circular chart temperature recorder. The temperature log will be reviewed by the V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

95 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: CRA throughout the study. An appropriate back up system (i.e. alarm, generator) and study site personnel telephone numbers should be in place in the event of a refrigerator failure Standard Policies / Return of Clinical Supplies For studies using Controlled Substances, all Federal, State, Province, Country, etc., regulations must be adhered to in regard to the shipping, storage, handling, and dispensing of controlled substances. Additionally, the investigator should have the appropriate controlled drug license(s) as mandated by Federal, State, Province, Country, etc. laws in which the study is being conducted. Investigational clinical supplies must be received by a designated person at the study site, handled and stored safely and properly, and kept in a secured location to which only the investigator and designated assistants have access. Clinical supplies are to be administered only in accordance with the protocol. The investigator is responsible for keeping accurate records of the clinical supplies received from the SPONSOR, the amount administered to the subjects/patients, and the amount remaining at the conclusion of the study. The Clinical Monitor should be contacted with any questions concerning investigational products where special or protective handling is indicated. At the end of the study, all unused clinical supplies must be returned as indicated on the Contact Information page(s). U.S. sites should follow instructions for the Clinical Supplies Return Form (V464) and contact your SPONSOR representative for review of shipment and form before shipping. Sites outside of the United States should check with local country Merck & Co., Inc. personnel for appropriate documentation that needs to be completed for vaccine accountability. Used vials may be discarded per site s handling of biohazardous waste after documentation of vaccine administration/accountability guidelines have been met. All of the unused sealed vaccine that remain at the conclusion of the study should be retained at the site until the SPONSOR representative is able to account for all of the vials originally shipped to the sites. The unused vials should then be returned to the SPONSOR by the SPONSOR representative at the completion of the study Transport of the Vaccine In the event the vaccine needs to be transported, either between study sub-sites that are under the direction of the same principal investigator or because the location of the study site changes, the SPONSOR must be notified. If vaccine needs to be transported, a vaccine transport standard operating procedure that is acceptable to the study site and the SPONSOR will be followed and a copy of the procedure will be placed in the Administrative Binder. 3.7 DATA MANAGEMENT Information regarding Data Management procedures for this protocol will be provided by the SPONSOR. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

96 V503, Protocol Issue Date: 15-Jun Product: V Protocol/Amendment No.: BIOLOGICAL SPECIMENS The laboratory that is analyzing any clinical samples should be blinded to the subject s vaccination group. It is the responsibility of the primary investigator to ensure that all staff personnel who will be handling, packaging, and/or shipping clinical specimens act in conformance with International Air Transport Association (IATA) regulations relating to the handling and shipping of hazardous goods. All specimens will be labeled with preprinted computer-generated labels provided by the SPONSOR and/or by the SPONSOR s Central Laboratory. Labels can only be affixed to dry surfaces. However, the labels can be used on polypropylene or polyethylene and will survive freezing and thawing. Information regarding biological specimens and sample labeling for this protocol will be provided by the SPONSOR/Central Laboratory. V503_001-00_ProtDet VERSION 8.0 APPROVED 15-Jun-2007

97 V503, Protocol Issue Date: 15-Jun Product: V503 1 Protocol/Amendment No.: ADMINISTRATIVE AND REGULATORY DETAILS 4.1 CONFIDENTIALITY Confidentiality of Data For Studies Conducted Under the U.S. IND Particular attention is drawn to the regulations promulgated by the Food and Drug Administration under the Freedom of Information Act providing, in part, that information furnished to clinical investigators and Institutional Review Boards will be kept confidential by the Food and Drug Administration only if maintained in confidence by the clinical investigator and Institutional Review Board. For All Studies By signing this protocol, the investigator affirms to the SPONSOR that information furnished to the investigator by the SPONSOR will be maintained in confidence and such information will be divulged to the Institutional Review Board, Ethics Review Committee, or similar or expert committee; affiliated institution; and employees only under an appropriate understanding of confidentiality with such board or committee, affiliated institution and employees. Data generated by this study will be considered confidential by the investigator, except to the extent that it is included in a publication as provided in the Publications section of this protocol Confidentiality of Subject/Patient Records For All Studies By signing this protocol, the investigator agrees that the SPONSOR (or SPONSOR representative), Institutional Review Board/Independent Ethics Committee (IRB/IEC), or Regulatory Agency representatives may consult and/or copy study documents in order to verify worksheet/case report form data. By signing the consent form, the subject/patient agrees to this process. If study documents will be photocopied during the process of verifying worksheet/case report form information, the subject/patient will be identified by unique code only; full names/initials will be masked prior to transmission to the SPONSOR. For Studies Conducted Under the U.S. IND By signing this protocol, the investigator agrees to treat all patient data used and disclosed in connection with this study in accordance with all applicable privacy laws, rules and regulations, including all applicable provisions of the Health Insurance Portability and Accountability Act and its implementing regulations, as amended from time to time. ( HIPAA ) Confidentiality of Investigator Information For All Studies By signing this protocol, the investigator recognizes that certain personal identifying information with respect to the investigator, and all subinvestigators and study site V503_001-00_ProtA&R VERSION 2.0 APPROVED 15-Jun-2007

98 V503, Protocol Issue Date: 15-Jun Product: V503 2 Protocol/Amendment No.: personnel, may be used and disclosed for study management purposes, as part of a regulatory submissions, and as required by law. This information may include: name, address, telephone number, and address; hospital or clinic address and telephone number; curriculum vitae or other summary of qualifications and credentials; and other professional documentation. Consistent with the purposes described above, this information may be transmitted to the SPONSOR, and subsidiaries, affiliates and agents of the SPONSOR, in your country and other countries, including countries that do not have laws protecting such information. Additionally, the investigator s name and business contact information may be included when reporting certain serious adverse events to regulatory agencies or to other investigators. By signing this protocol, the investigator expressly consents to these uses and disclosures. For Multicenter Studies In order to facilitate contact between investigators, the SPONSOR may share an investigator s name and contact information with other participating investigators upon request. 4.2 COMPLIANCE WITH LAW, AUDIT, AND DEBARMENT By signing this protocol, the investigator agrees to conduct the study in an efficient and diligent manner and in conformance with this protocol; generally accepted standards of Good Clinical Practice; and all applicable federal, state, and local laws, rules and regulations relating to the conduct of the clinical study. The Code of Conduct, a collection of goals and considerations that govern the ethical and scientific conduct of clinical investigations sponsored by Merck & Co., Inc., is attached. The investigator also agrees to allow monitoring, audits, Institutional Review Board/Independent Ethics Committee review, and regulatory agency inspection of trialrelated documents and procedures and provide for direct access to all study-related source data and documents. The investigator agrees not to seek reimbursement from subjects/patients, their insurance providers, or from government programs for procedures included as part of the study reimbursed to the investigator by the SPONSOR. The Investigator shall prepare and maintain complete and accurate study documentation in compliance with Good Clinical Practice standards and applicable federal, state, and local laws, rules and regulations; and, for each subject/patient participating in the study, provide all data, and upon completion or termination of the clinical study submit any V503_001-00_ProtA&R VERSION 2.0 APPROVED 15-Jun-2007

99 V503, Protocol Issue Date: 15-Jun Product: V503 3 Protocol/Amendment No.: other reports to the SPONSOR as required by this protocol or as otherwise required pursuant to any agreement with the SPONSOR. Study documentation will be promptly and fully disclosed to the SPONSOR by the investigator upon request and also shall be made available at the investigator s site upon request for inspection, copying, review, and audit at reasonable times by representatives of the SPONSOR or any regulatory agencies. The investigator agrees to promptly take any reasonable steps that are requested by the SPONSOR as a result of an audit to cure deficiencies in the study documentation and worksheets/case report forms. International Conference of Harmonization Good Clinical Practice guidelines (Section 4.3.3) recommend that the investigator inform the subject s primary physician about the subject s participation in the trial if the subject has a primary physician and if the subject agrees to the primary physician being informed. According to European legislation, a SPONSOR must designate a principal or coordinating investigator (CI) to review the report (summarizing the study results) and confirm that to the best of his/her knowledge the report accurately describes conduct and results of the study. The SPONSOR may consider one or more factors in the selection of the individual to serve as the CI (e.g., thorough understanding of clinical trial methods, appropriate enrollment of subject/patient cohort, timely achievement of study milestones, availability of the CI during the anticipated review process). The investigator will promptly inform the SPONSOR of any regulatory agency inspection conducted for this study. Persons debarred from conducting or working on clinical studies by any court or regulatory agency will not be allowed to conduct or work on this SPONSOR s studies. The investigator will immediately disclose in writing to the SPONSOR if any person who is involved in conducting the study is debarred, or if any proceeding for debarment is pending or, to the best of the investigator s knowledge, threatened. In the event the SPONSOR prematurely terminates a particular trial site, the SPONSOR will promptly notify that site s IRB/IEC. 4.3 COMPLIANCE WITH FINANCIAL DISCLOSURE REQUIREMENTS By signing this protocol, the investigator agrees to provide to the SPONSOR accurate financial information to allow the SPONSOR to submit complete and accurate certification and disclosure statements as required by U.S. Food and Drug Administration regulations (21 CFR Part 54). The investigator further agrees to provide this information on a Financial Disclosure/Certification Form that is provided by Merck & Co., Inc. This requirement also extends to subinvestigators. The investigator also consents to the transmission of this information to Merck & Co., Inc. in the United States for these purposes. This may involve the transmission of information to countries that do not have laws protecting personal data. V503_001-00_ProtA&R VERSION 2.0 APPROVED 15-Jun-2007

100 V503, Protocol Issue Date: 15-Jun Product: V503 4 Protocol/Amendment No.: QUALITY CONTROL AND QUALITY ASSURANCE By signing this protocol, the SPONSOR agrees to be responsible for implementing and maintaining quality control and quality assurance systems with written SOPs to ensure that trials are conducted and data are generated, documented, and reported in compliance with the protocol, accepted standards of Good Clinical Practice, and all applicable federal, state, and local laws, rules and regulations relating to the conduct of the clinical study. 4.5 COMPLIANCE WITH INFORMATION PROGRAM ON CLINICAL TRIALS FOR SERIOUS OR LIFE THREATENING CONDITIONS Under the terms of The Food and Drug Administration Modernization Act (FDAMA), the SPONSOR of the study is solely responsible for determining whether the study is subject to the requirements for submission to the Clinical Trials Data Bank, Merck, as SPONSOR of this study, will review this protocol and submit the information necessary to fulfill this requirement. Merck entries are not limited to FDAMA mandated trials. Merck s voluntary listings, beyond those mandated by FDAMA, will be in the same format as for treatments for serious or life-threatening illnesses. Information posted will allow patients to identify potentially appropriate trials for their disease conditions and pursue participation by calling a central contact number for further information on appropriate study locations and site contact information. By signing this protocol, the investigator acknowledges that the statutory obligation under FDAMA is that of the SPONSOR and agrees not to submit any information about this study to the Clinical Trials Data Bank. 4.6 PUBLICATIONS As this study is part of a multicenter trial, publications derived from this study should include input from the investigator(s) and SPONSOR personnel. Such input should be reflected in publication authorship, and whenever possible, preliminary agreement regarding the strategy for order of authors names should be established before conducting the study. Subsequent to the multicenter publication, or 24 months after completion of the study, whichever comes first, an investigator and/or his/her colleagues may publish the results for their study site independently. However, the SPONSOR does not recommend separate publication of individual study site results due to scientific concerns. The SPONSOR must have the opportunity to review all proposed abstracts, manuscripts, or presentations regarding this study 60 days prior to submission for publication/presentation. Any information identified by the SPONSOR as confidential must be deleted prior to submission. SPONSOR review can be expedited to meet publication guidelines. V503_001-00_ProtA&R VERSION 2.0 APPROVED 15-Jun-2007

101 V503, Protocol Issue Date: 15-Jun Product: V503 1 Protocol/Amendment No.: LIST OF REFERENCES 1. Munoz, N, Bosch, XF, Catellsague, X, Diaz, M, De Sanjose, S, Hammouda, D, Shah, KV, and Meijer, C. Against which human papillomavirus types shall we vaccinate and screen? The international perspective. Int J Cancer 2004;111: Pagliusi SR, Aguado MT. Efficacy and other milestones for human papillomavirus vaccine introduction. Vaccine 2004;23: Jansen KU, Shaw AR. Human papillomavirus vaccines and prevention of cervical cancer. Annu Rev Med 2004;55: Chuang T-Y, Perry HO, Kurland LT, Ilstrup DM. Condyloma acuminatum in Rochester, Minn, :II anaplasias and unfavorable outcomes. Arch Dermatol 1984;120: Stone KM. Human papillomavirus infection and genital warts: update on epidemiology and treatment. Clin Infec Dis 1995;20(Suppl 1):S91-S Bosch FX, Munoz N. The viral etiology of cervical cancer. Virus Research 2002;89(2): Christensen ND. Emerging human papillomavirus vaccines. Expert Opin Emerg Drugs 2005;10(1): Glikman D, Baroody FM. Recurrent respiratory papillomatosis with lung involvement. N Engl J Med 2005;352(24):e Szeps M, Dahlgren L, Aaltonen LM, Ohd J, Kanter-Lewenshon L, Dahlstrand H, et al. Human papillomavirus, viral load and proliferation rate in recurrent respiratory papillomatosis in response to alpha interferon treatment. J Gen Virol 2005;86: Lin T-S, Lee H, Chen R-A, Ho M-L, Lin C-Y, Chen Y-H, et al. An association of DNMT3b protein expression with P16INK4a promoter hypermethylation in nonsmoking female lung cancer with human papillomavirus infection. Cancer Lett 2005;226: Lyronis ID, Baritaki S, Bizakis I, Tsardi M, Spandidos DA. Evaluation of the prevalence of human papillomavirus and Epstein-Barr virus in esophageal squamous cell carcinomas. Int J Biol Markers 2005;20(1): Yang H, Yang K, Khafagi A, Tang Y, Carey TE, Opipari AW, et al. Sensitive detection of human papillomavirus in cervical, head/neck, and schistosomiasisassociated bladder malignancies. Proceedings of the National Academy of Sciences 2005;102(21): Howley PM. Papillomavirinae: The viruses and their replication. In: Fields BN, Knipe DM, Howley PM, et al., eds. Virology. 3rd ed. Philadelphia: Lippincott-Raven Publishers, 1996: V503_001-00_ProtApp VERSION 2.1 APPROVED 15-Jun-2007

102 V503, Protocol Issue Date: 15-Jun Product: V503 2 Protocol/Amendment No.: Auvinen E, Crusius K, Steuer B, Alonso A. Human papillomavirus type 16 E5 protein (review). Int J Oncol 1997;11: Stern PL. Immune control of human papillomavirus (HPV) associated anogenital disease and potential for vaccination. J Clin Virol 2005;Suppl 32:S72-S Oriel JD. Natural history of genital warts. Brit J Vener Dis 1971;47: Xi LF, Koutsky LA. Epidemiology of genital human papillomavirus infections. Bull Inst Pasteur 1997;95: Koutsky L. Epidemiology of genital human papillomavirus infection. Am J Med 1997;102(5A): Walboomers JMM, Jacobs MV, Manos MM, Bosch FX, Kummer A, Shah KV, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189: Nobbenhuis MAE, Walboomers JMM, Helmerhorst TJM, Rozendaal L, Remmink AJ, Risse EKJ, et al. Relation of human papillomavirus status to cervical lesions and consequences for cervical-cancer screening: a prospective study. Lancet 1999;354: Wallin K-L, Wiklund F, Ångström T, Bergman F, Stendahl U, Wadell G, et al. Typespecific persistence of human papillomavirus DNA before the development of invasive cervical cancer. N Engl J Med 1999;341(22): Melbye M, Frisch M. The role of human papillomaviruses in anogenital cancers. Semin Cancer Biol 1998;8: Consensus Development Panel. Consensus statement: National Institutes of Health Consensus Development Conference statement on cervical cancer. Gynecol Oncol 1997;66: Anttila A, Pukkala E, Söderman B, Kallio M, Nieminen P, Hakama M. Effect of organised screening on cervical cancer incidence and mortality in Finland, : recent increase in cervical cancer incidence. Int J Cancer 1999;83: Östör AG. Natural history of cervical intraepithelial neoplasia: a critical review. Int J Gynecol Pathol 1993;12(2): Chan ISF, Bohidar NR. Exact power and sample size for vaccine efficacy studies. Commun Statist-Theory Meth 1998;27(6): Hasselblad V, Kong, DF. Statistical methods for comparison to placebo in activecontrol trials. Drug Inf J 2001; 35 (2): Miettinen O, Nurminen M. Comparative analysis of two rates. Stat Med 1985;4: Haybittle JL. Repeated assessment of results in clinical trials of cancer treatment. Brit J Rad 1971;44: Harper DM, Franco EL, Wheeler CM, Moscicki A-B, Romanowski B, Roteli-Martins CM, et al. Sustained efficacy up to 4 5 years of a bivalent L1 virus-like particle V503_001-00_ProtApp VERSION 2.1 APPROVED 15-Jun-2007

103 V503, Protocol Issue Date: 15-Jun Product: V503 3 Protocol/Amendment No.: vaccine against human papillomavirus types 16 and 18: follow-up from a randomised control trial. The Lancet 2006;367: V503_001-00_ProtApp VERSION 2.1 APPROVED 15-Jun-2007

104 Merck s policy on posting of redacted study protocols on journal websites is described in Merck Guidelines for Publication of Clinical Trials in the Scientific Literature on the website. For publicly posted protocols, Merck redacts the background and rationale sections because these sections may contain proprietary information. Merck also redacts the names of any individuals due to privacy issues. The appendices generally are not provided because they may be lengthy and contain non-essential information. The publicly posted protocol includes all the key sections that are relevant to evaluating the study, specifically those sections describing the study objectives and hypotheses, the patient inclusion and exclusion criteria, the study design and procedures, the efficacy and safety measures, the statistical analysis plan, and amendments relating to those sections. This report may include approved and non-approved uses, formulations, or treatment regimens. The results reported may not reflect the overall profile of a product. Before prescribing any product mentioned this report, healthcare professionals should consult local prescribing information for the product approved in their country. Copyright 2014 Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. All Rights Reserved. Not for regulatory or commercial use.

105 A Randomized, International, Double-Blinded (With In-House Blinding), Controlled With GARDASIL, Dose-Ranging, Tolerability, Immunogenicity, and Efficacy Study of a Multivalent Human Papillomavirus (HPV) L1 Virus-Like Particle (VLP) Vaccine Administered to 16- to 26-Year-Old Women

106 V503, Protocol Issue Date: 16-Aug Product: V503 Protocol/Amendment No.: THIS PROTOCOL AND ALL OF THE INFORMATION RELATING TO IT ARE CONFIDENTIAL AND PROPRIETARY PROPERTY OF MERCK SHARP & DOHME CORP., A SUBSIDIARY OF MERCK & CO., INC., WHITEHOUSE STATION, NJ, U.S.A. THIS PROTOCOL REPLACES THE ORIGINAL PROTOCOL AND ANY SUBSEQUENT AMENDMENTS AND SHOULD BE SIGNED BY ALL INVESTIGATORS SIGNING THE ORIGINAL PROTOCOL. SPONSOR: Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. (hereafter referred to as the SPONSOR or Merck) One Merck Drive P.O. Box 100 Whitehouse Station, NJ, , U.S.A. Protocol-specific Sponsor Contact information can be found in the Administrative Binder. TITLE: A Randomized, International, Double-Blinded (With In-House Blinding), Controlled With GARDASIL Dose-Ranging, Tolerability, Immunogenicity, and Efficacy Study of a Multivalent Human Papillomavirus (HPV) L1 Virus-Like Particle (VLP) Vaccine Administered to 16- to 26-Year-Old Women INVESTIGATOR: PRIMARY: CLINICAL PHASE: IIb/III US IND NUMBER: SITE: INSTITUTIONAL REVIEW BOARD/ETHICS REVIEW COMMITTEE: V503_001-02_ProtTitle APPROVED 16-Aug-2011

107 V503, Protocol Issue Date: 16-Aug Product: V503 1 Protocol/Amendment No.: SUMMARY OF CHANGES PRIMARY REASON FOR THIS AMENDMENT: The primary purpose of the Protocol amendment is to update the data analysis related to the first secondary objective: Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine will reduce the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 months or longer apart compared with GARDASIL - to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type. Hypothesis: Administration of 9-valent HPV L1 VLP vaccine to 16- to 26-yearold adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 months or longer apart compared with GARDASIL (The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the vaccine efficacy be greater than 0.) The statistical criterion for success relating to the lower bound of the 95% confidence interval was changed from 0 to 25%: Hypothesis: Administration of 9-valent HPV L1 VLP vaccine to 16- to 26-yearold adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 months or longer apart compared with GARDASIL (The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the vaccine efficacy be greater than 25%.) V503_001-02_ProtSoC APPROVED 16-Aug-2011

108 V503, Protocol Issue Date: 16-Aug Product: V503 2 Protocol/Amendment No.: OTHER CHANGES INCLUDED IN THE AMENDMENT: Protocol Section 1.7 Study Flow Chart Footnote Footnote Revision Updated anti-hpv measurements by removing reference to specific assay and MRL Secondary Objectives Changed the statistical criterion for success relating to the lower bound of the 95% confidence interval from 0 to 25% which is referenced in the hypothesis to Secondary Objective (1) Exploratory Replaced the term "type specific anti-hpv L1 VLP IgG ELISA" with "9-valent HPV total IgG Luminex immunoassay." 2.2 SUBJECT/PATIENT INCLUSION CRITERIA Revised Inclusion Criteria 6 to state: Subject has had 0 male and/or female sexual partners, is 18 years of age or older, and plans to become sexually active within the first 3-6 months of the study Summary of Study Design Replaced text in Part A of this section. to state: The primary immunogenicity analysis in Part A will be conducted based on the Week 4 Postdose 3 (Month 7) immunogenicity results of the 9- valent HPV L1 VLP vaccine formulation selected for use in Part B and the GARDASIL comparator. The formal statistical test of the Part A hypothesis based upon screened and cleaned Month 7 data will be provided in the final study report Immunogenicity Measurements Replaced the term "type specific anti-hpv L1 VLP IgG ELISA" with "9-valent HPV total IgG Luminex immunoassay." V503_001-02_ProtSoC APPROVED 16-Aug-2011

109 V503, Protocol Issue Date: 16-Aug Product: V503 3 Protocol/Amendment No.: Protocol Section 2.7 DATA ANALYSIS SUMMARY Revision - Updated Part A Month 7 Primary Immunogenicity Analysis text regarding Part A Month 7 data. - Changed the Part B statistical criterion for success relating to the lower bound of the 95% confidence interval from 0 to 25% which is referenced in the last sentence of paragraph 4 in the section Concomitant Medication(s)/Treatment(s) Serum for Anti-HPV Measurements at Scheduled Visits 3.3 EFFICACY/IMMUNOGENICITY MEASUREMENTS Competitive Luminex Immunoassay (clia) - Anti-HPV Levels in Serum Added the reporting of non-study HPV vaccinations as the 7th paragraph in the section. Updated serology testing in the third paragraph to include that serum may also be used, during or after the clinical trial, for further HPV immunologic testing in addition to tests specified in the protocol. Added Total IgG Luminex Immunoassay assay description as Section in the section. - Replaced Competitive Luminex Immunoassay (clia) assay description text with final Competitive Luminex Immunoassay (clia) V2 Assay Description. - Added Competitive Luminex Immunoassay (clia) V2 Assay Serostatus Cutoff Information Clinical and Laboratory Measurements for Safety Added text regarding serious adverse experiences and fetal loss and location of guidance for each to the last paragraph in the section. V503_001-02_ProtSoC APPROVED 16-Aug-2011

110 V503, Protocol Issue Date: 16-Aug Product: V503 4 Protocol/Amendment No.: Protocol Section Reporting of Pregnancy/Breastfeeding Events to the SPONSOR Revision - Revised the reporting cut off in the 3rd sentence of the section to specify pregnancies with LMP prior to or equal to Day 180 following the final vaccination. - Added reporting specifications regarding fetal loss for pregnancies in subjects with LMP prior to or equal to Day 180 following the final vaccination as the 4 th sentence in the section Responsibility for and Timing of Analyses Revised text for Part A Month 7 Primary Immunogenicity Analysis regarding Part A Month 7 data Immunogenicity Deleted first bullet point in PPI Population. 7. ATTACHMENTS - Replaced updated U.S. product circular to the attachments section. - Replaced updated U.S. patient product information document to the attachments section. Throughout the protocol amendment Replaced "Merck & Co., Inc." with "Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc." V503_001-02_ProtSoC APPROVED 16-Aug-2011

111 V503, Protocol Issue Date: 16-Aug PROTOCOL A Randomized, International, Double-Blinded (With In-House Blinding), Controlled With GARDASIL, Dose-Ranging, Tolerability, Immunogenicity, and Efficacy Study of a Multivalent Human Papillomavirus (HPV) L1 Virus-Like Particle (VLP) Vaccine Administered to 16- to 26-Year-Old Women TABLE OF CONTENTS Contents Application Starting Page SUMMARY OF CHANGES 3 Primary Reason For This Amendment: 3 Other Changes Included In The Amendment: 4 1. SUMMARY Title Indication Summary of Rationale Summary of Study Design Sample Dosage/Dosage Form, Route, and Dose Regimen Study Flow Chart CORE PROTOCOL Objectives and Hypotheses Part A Analysis Primary Part B Analysis (Tolerability and Efficacy Analyses Include Part A Subjects Who Received the Selected 9-Valent HPV L1 VLP Vaccine Dose or the Comparator GARDASIL ) Primary Secondary Exploratory Subject/Patient Inclusion Criteria 23 19

112 V503, Protocol Issue Date: 16-Aug TABLE OF CONTENTS (CONT.) Contents Application Starting Page 2.3 Subject/Patient Exclusion Criteria Study Design and Duration Summary of Study Design Vaccination Plan List of Immunogenicity and efficacy Measurements Immunogenicity Measurements Efficacy Measurements List of Safety Measurements Data Analysis Summary PROTOCOL DETAILS Rationale Introduction to Human Papillomavirus Disease Burden Biology of HPV Epidemiology of HPV Rationale for This Study Rationale for Part A Doses Summary of GARDASIL, 8-valent HPV L1 VLP Vaccine, and 9-valent HPV L1 VLP Vaccine Studies 3.2 Study Procedures Concomitant Medication(s)/Treatment(s) Prerequisites for Study Visits Summary of Scheduled Study Visit Procedures Calculation of Scheduled Visit Windows Informed Consent and Assignment of Baseline Numbers History & Medications Pregnancy Testing and Serum Collection 43 39

113 V503, Protocol Issue Date: 16-Aug TABLE OF CONTENTS (CONT.) Contents Application Starting Page Physical Examination & Vital Signs Gynecological Examination Genital/Cervical Swabs, Pap Test, and Sexually Transmitted Infection (STI) Testing External Genital Lesion Examination IVRS Assignment of Allocation Numbers and Vaccination Vials Study Vaccine Administration Clinical Follow-Up Summary of Unscheduled Study Visit Procedures Colposcopy, Cervical Biopsy, and Cervical Definitive Therapy External Genital Lesion and Vaginal Lesion Diagnosis and Follow-Up Processing of Tissue Specimens In the Context of the Study Pap Tests and Tissue Specimens Taken Outside the Context of the Study Procedures for Collection and Handling of Study Specimens Serum or Urine Specimen for Pregnancy Test Serum for Anti-HPV Measurements at Scheduled Visits Cord Serum and Maternal Serum for Anti-HPV Measurements (Applicable to Participating Study Sites) Labial/Vulvar/Perineal and Perianal (LVPP) Swabs for HPV PCR Endo/Ectocervical (EEC) Swab for HPV PCR Pap Test (ThinPrep ) Specimen Collection External Genital Lesion Biopsy and Vaginal Lesion Biopsy Colposcopy Guidelines

114 V503, Protocol Issue Date: 16-Aug TABLE OF CONTENTS (CONT.) Contents Application Starting Page Procedures for Cervical Biopsy Procedures for Endocervical Curettage (ECC) Procedures for Loop Electrosurgical Excision Procedure (LEEP) and Top Hat Method Procedure for Laser Conization Cold-Knife Conization Ablative Cervical Definitive Therapy Discontinuation/Withdrawal from Study Subject Relocation Conduct of the Clinical Trial Scientific Advisory Committee HPV Vaccine Program Pathology Panel Responsibility of the Data and Safety Monitoring Board (DSMB) Senior Management Committee for Dose Selection Efficacy/Immunogenicity Measurements Competitive Luminex Immunoassay (clia) - Anti-HPV Levels in Serum PCR Assays - Detection of HPV in Swabs and Tissue Specimens Multiplex PCR Assays Preparation and Disposition of Thinsections of Biopsy Tissue Total IgG Luminex Immunoassay Safety Measurements Clinical and Laboratory Measurements for Safety Recording Adverse Experiences Definition of an Overdose for This Protocol

115 V503, Protocol Issue Date: 16-Aug TABLE OF CONTENTS (CONT.) Contents Application Starting Page Reporting of Overdose to SPONSOR Reporting of Pregnancy/Breastfeeding Events to the SPONSOR Immediate Reporting of Adverse Experiences to the SPONSOR Serious Adverse Experiences Evaluating Adverse Experiences SPONSOR Responsibility for Reporting Adverse Experiences Data Analysis Responsibility for and Timing of Analyses Hypotheses Variables and Time Points of Interest Safety/Tolerability Efficacy Immunogenicity Analysis Populations Safety Efficacy Immunogenicity Statistical Methods Efficacy Immunogenicity Safety Multiplicity Considerations Sample Size and Power Calculations Interim Analyses Definition of Compliance Measure

116 V503, Protocol Issue Date: 16-Aug TABLE OF CONTENTS (CONT.) Contents Application Starting Page 3.6 Labeling, Packaging, Storage, Dispensing, and Return of Clinical Supplies Subject and Replacements Information Product Descriptions Primary Packaging and Labeling Information Secondary Packaging and Labeling Information- Supplies Will Not be Kitted Clinical Supplies-Vaccine Disclosure Storage Requirements Standard Policies / Return of Clinical Supplies Transport of the Vaccine Data Management Biological Specimens ADMINISTRATIVE AND REGULATORY DETAILS Confidentiality Confidentiality of Data Confidentiality of Subject/Patient Records Confidentiality of Investigator Information Compliance with Law, Audit, and Debarment Compliance with Financial Disclosure Requirements Quality Control and Quality Assurance Compliance with Information Program on Clinical Trials for Serious or Life Threatening Conditions 4.6 Publications LIST OF REFERENCES APPENDICES ATTACHMENTS SIGNATURES

117 V503, Protocol Issue Date: 16-Aug TABLE OF CONTENTS (CONT.) Contents Application Starting Page 8.1 SPONSOR S REPRESENTATIVE INVESTIGATOR 148

118 V503, Protocol Issue Date: 16-Aug Product: V503 1 Protocol/Amendment No.: SUMMARY 1.1 TITLE A Randomized, International, Double-Blinded (With In-House Blinding), Controlled With GARDASIL Dose-Ranging, Tolerability, Immunogenicity, and Efficacy Study of a Multivalent Human Papillomavirus (HPV) L1 Virus-Like Particle (VLP) Vaccine Administered to 16- to 26-Year-Old Women 1.2 INDICATION Prevention of cervical, vulvar, and vaginal cancers and related precancers, external genital lesions, Pap test abnormalities, and persistent infection caused by Human Papillomavirus (HPV) 6, 11, 16, 18, 31, 33, 45, 52, and SUMMARY OF RATIONALE Redacted V503_001-02_ProtCore APPROVED 16-Aug-2011

119 V503, Protocol Issue Date: 16-Aug Product: V503 2 Protocol/Amendment No.: Redacted 1.4 SUMMARY OF STUDY DESIGN Part A (dose-ranging and tolerability): Approximately 1,240 healthy 16- to 26-yearold women will be randomized in equal numbers to one of three 9-valent HPV L1 VLP vaccine formulations or the comparator GARDASIL in a three dose regimen (Day 1, Month 2, and Month 6). The best 9-valent HPV L1 VLP vaccine dose, among those doses tested, will be selected for Part B after approximately 100% of the Part A post-dose 2 immunogenicity and adverse experience data are available in the clinical database. The dose for Part B will be chosen by a senior management committee at the SPONSOR who is not directly involved with study conduct. This committee will choose the dose after reviewing immunogenicity summaries by vaccination group, and will relay the dose selection to the Data Safety Monitoring Board (DSMB), who will approve the decision based on unblinded tolerability data. If the DSMB expresses a concern about the tolerability profile of the selected dose, the senior management committee will select another dose and again request approval from the DSMB. In the unlikely event that the SPONSOR is unable to determine an acceptable 9-valent HPV L1 VLP vaccine formulation for use in Part B, the study may not continue to Part B. Part B (tolerability, immunogenicity, and efficacy): Approximately 13,380 additional healthy 16- to 26-year-old women will be randomized in equal numbers to the selected 9- valent HPV L1 VLP vaccine chosen from Part A or the comparator GARDASIL The primary immunogenicity analyses for Part B will be the formal comparison of the selected 9-valent HPV L1 VLP vaccine to the GARDASIL control and will include those subjects enrolled in Part B only. The overall tolerability and efficacy analyses, however, will also include subjects enrolled in Part A (those who received the selected 9- valent HPV L1 VLP vaccine or the comparator GARDASIL enrolled in Part B (14,620 total). The total duration of the study for each subject enrolled in either Part A or Part B who receive the selected 9-valent HPV L1 VLP vaccine or GARDASIL is 42 months. The primary efficacy analysis is case driven and will be conducted after at least 30 primary efficacy cases have been observed. In addition, in order to have a sufficient duration of follow-up of vaccinated subjects, the primary efficacy analysis will not be performed before the median follow-up of PART B subjects 4 The 9-valent HPV L1 VLP (30/40/60/40/20/20/20/20/20 mcg each of HPV Types 6/11/16/18/31/33/45/52/58) formulation was selected for study in Part B and will be referred to as the selected 9-valent HPV L1 VLP vaccine or with regards to Part B, simply as the 9-valent HPV L1 VLP vaccine in this protocol. V503_001-02_ProtCore APPROVED 16-Aug-2011

120 V503, Protocol Issue Date: 16-Aug Product: V503 3 Protocol/Amendment No.: is approximately 24 months after initial vaccination. Since the primary efficacy analysis will be conducted prior to completion of the study, the remainder of the study is considered an extension. The purpose of the extension is to collect data on the longerterm efficacy, tolerability, and immunogenicity of the vaccine. The clinical, statistical and data management study personnel at the SPONSOR who are involved with study conduct will remain blinded to subject vaccination group allocations until the required number of cases of the primary efficacy endpoint have been observed and the database is unblinded for the primary efficacy analysis. Laboratory personnel, the pathology panel, and the investigators, site personnel and subjects will remain blinded to whether subjects received the 9-valent HPV L1 VLP vaccine or GARDASIL period (~3.5 years). 1.5 SAMPLE This study will enroll healthy 16- to 26-year-old women. To minimize the number of enrolled subjects who are positive to study vaccine HPV types at baseline, subjects who have a history of 5 or more sexual partners, external genital or vaginal warts, a history of an abnormal Pap test, or a history of an abnormal cervical biopsy result will be excluded. 1.6 DOSAGE/DOSAGE FORM, ROUTE, AND DOSE REGIMEN In Part A, subjects will be given 9-valent HPV L1 VLP vaccine or GARDASIL as summarized in Table 1-1. The rationale for these formulations can be found in Section Table 1-1 Part A Vaccine Formulations (0.5-mL Dose) Arm 1 HPV 6/ 11/ 16/ 18 20/ 40/ 40/ 20 (GARDASIL HPV 31 (mcg) HPV 33 (mcg) HPV 45 (mcg) HPV 52 (mcg) HPV 58 (mcg) Total VLP (mcg) AAHS (mcg) Estimated Subjects / 40/ 40/ / 40/ 60/ / 40/ 80/ AAHS=Merck Aluminum Adjuvant (amorphous aluminum hydroxyl phosphate sulfate) In Part B, approximately 13,380 additional subjects will be randomized in equal numbers to receive the selected 9-valent HPV L1 VLP vaccine (in bold in table above) or GARDASIL Study vaccine will be administered as a 0.5-mL intramuscular injection at Day 1, Month 2, and Month 6. V503_001-02_ProtCore APPROVED 16-Aug-2011

121 Product: V503 4 Protocol/Amendment No.: STUDY FLOW CHART Key Scheduled Tests and Events Visit Windows : Obtain Informed Consent, Assign Baseline Number + Review Inclusion/Exclusion Criteria + Collect Medical History (past year)/lifetime Gynecologic History + Update Medical History and Gynecologic History (new conditions not already recorded as Medical History or Adverse Experiences) V503_001-02_ProtCore APPROVED 16-Aug-2011 Day Months 1 Day No visit window; 2 months 3 to 5 3 to 7 12, 18, 24, 30, 36, 42 months Only applies to the 6 months after Day weeks weeks after Day 1,±4 weeks; first 150 subjects after Day 1, 1, after after Mo. 7 is the final visit for 9-valent enrolled (for return ±4 weeks of VRC data) ±3 weeks Month 2 Month 6 doses that are not selected Review Medications and Non-Study Vaccines Pregnancy/Serum Specimen Collection: Pregnancy Test (serum or urine) Serum for Anti-HPV (including Retention Serum) Additional Serum for Anti-HPV Assay Development and + Reference Samples Perform Physical Examination Collect Vital Signs (sitting pulse, weight, blood pressure, + respirations, and oral temperature) Perform Gynecologic Physical Examination # Pelvic Specimen Collection (obtain in serial order): Labial/Vulvar/Perineal & Perianal (LVPP) Swabs for HPV PCR Endo/Ectocervical (EEC) Swab for HPV PCR Pap Test (ThinPrep at Day 1, Month 7, Month 18, Month 30, Month 42) STI Testing (local laboratory testing, if clinically indicated) Perform External Genital Wart Inspection Assign Allocation Number + Perform Vaccination (intramuscular [i.m.]; prefer deltoid muscle of nondominant arm; do not give in buttocks) Provide Vaccination Report Card (VRC) Review and Collect VRC data Review Adverse Experiences, Clinical Follow-up for Safety V503, Protocol Issue Date: 16-Aug

122 Product: V503 5 Protocol/Amendment No.: Study Flow Chart Footnotes To calculate visit windows, assume 1 month equals 30 days and 1 week equals 7 days. With regard to protocol study visit windows, the following situations require consultation between the investigator and the SPONSOR and written documentation of the collaborative decision: a subject needs to be scheduled earlier than the start of a visit window, the study site is considering skipping a visit, or a study site needs significant guidance on scheduling visit windows. For post Month 7 visits, if a subject is very late for a visit such that she is in the next visit window, schedule future visits a minimum of 4 months apart until the visits are within the protocol-specified window. See the Administrative Binder for a summary of deviations that require documentation in this study. The Day 16 visit only applies to the first 150 subjects enrolled in Part A. There is no visit window on the Day 16 visit, because the purpose of the visit is to obtain 40 VRCs in a timely manner for an early tolerability evaluation. Although data from only 40 VRCs are needed for the tolerability evaluation, the Day 16 visit will be scheduled for more than 40 subjects to allow for lost VRCs, cancelled visits, or other scenarios that may delay the return of the VRCs. See the protocol details for prerequisites for medications and non-study vaccines. The serum pregnancy test or urine pregnancy test (sensitive to 25 miu/ml β-hcg) will be performed per the manufacturer s instructions. All materials used for serum and/or urine pregnancy testing will provided by the study sites. Serum for anti-hpv measurements may be collected after the pelvic examination, but must be collected before vaccination.. Serum (including Reference Serum) must be shipped as specified by the SPONSOR/Central Laboratory. The Retention Serum vial must remain at the site until the SPONSOR notifies the study site to ship the samples.. If the subject has a fever (defined as an oral temperature of 100 F or 37.8 C) within the 24-hour period prior to receiving a study vaccination, the subject should not receive study vaccine, and the vaccination visit should be rescheduled until after the fever has resolved. # Collect pelvic study specimens prior to performing the bimanual pelvic examination. See the protocol details for prerequisites for pelvic sample collection. Type-specific HPV Polymerase Chain Reaction (PCR) testing to be performed at MRL. The 2 LVPP swabs are placed in one container of specimen transfer medium (STM) and 1 EEC swab is placed in a separate container of STM. Swabs must be shipped as specified by the SPONSOR/Central Laboratory. Pap test to be analyzed by the SPONSOR-designated Central Laboratory. The Pap test fluid will be evaluated for Chlamydia and gonorrhea at the specified visits, but additional local laboratory tests for Chlamydia and gonorrhea may be performed at other visits. Other sexually transmitted infection (STI) tests, including HSV, syphilis, and HIV, may be performed at any visit as needed. Observe subjects for 30 minutes after each vaccination for immediate untoward effects. The subject should use the Vaccination Report card (VRC) to document this safety information: oral temperature in the evening after each study vaccination and daily, at the same time of day, for 4 days after each study vaccination; injection-site or systemic adverse experience(s) starting after each study vaccination for a total of 15 days. The first 150 subjects enrolled will return the Day 1 VRC to the study site at Day 16, while the remaining subjects will return the Day 1 VRC at the Month 2 study visit. For all subjects, Month 2 VRCs will be returned at the Month 3 study visit and Month 6 VRCs will be returned at the Month 7 study visit. Serious adverse experiences (SAEs), pregnancy events (including non-randomized subjects with a positive Day 1 pregnancy test), and lactation events should be reported to the SPONSOR as instructed in the protocol and in the protocol Attachments. V503_001-02_ProtCore APPROVED 16-Aug-2011 V503, Protocol Issue Date: 16-Aug

123 V503, Protocol Issue Date: 16-Aug Product: V503 6 Protocol/Amendment No.: CORE PROTOCOL 2.1 OBJECTIVES AND HYPOTHESES Part A Analysis Primary (1) Objective: To evaluate the tolerability of the 9-valent HPV L1 VLP vaccine when administered to 16- to 26-year-old women. Hypothesis: 9-valent HPV L1 VLP vaccine administered to 16- to 26-year-old women is generally well-tolerated. (2) Objective: To evaluate a formulation of 9-valent HPV L1 VLP vaccine for use in the efficacy evaluation in Part B. Hypothesis: 9-valent HPV L1 VLP vaccine selected for use in Part B generates anti-hpv 6, 11, 16, and 18 GMTs 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The GMTs for each of HPV types 6, 11, 16, and 18 will be tested separately. Noninferiority for GMTs is defined as statistically less than a 2-fold decrease.) Part B Analysis (Tolerability and Efficacy Analyses Include Part A Subjects Who Received the Selected 9-Valent HPV L1 VLP Vaccine Dose or the Comparator GARDASIL Primary (1) Objective: To evaluate the tolerability of the 9-valent HPV L1 VLP vaccine when administered to 16- to 26-year-old women. Hypothesis: 9-valent HPV L1 VLP vaccine administered to 16- to 26-year-old women is generally well-tolerated. (2) Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine will reduce the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related highgrade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), highgrade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer, compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type. V503_001-02_ProtCore APPROVED 16-Aug-2011

124 V503, Protocol Issue Date: 16-Aug Product: V503 7 Protocol/Amendment No.: Hypothesis: Administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer, compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type. (The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the vaccine efficacy be greater than 25%) (3) Objective: To demonstrate that the 9-valent HPV L1 VLP vaccine induces noninferior GMTs for anti-hpv 6, 11, 16, and 18 compared to GARDASIL Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 6, 11, 16, and 18 GMTs 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26-year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The GMTs for each of HPV types 6, 11, 16, and 18 will be tested separately. Noninferiority for GMTs is defined as the lower bound of the twosided 95% confidence interval for the geometric mean ratio of 9-valent vaccine versus GARDASIL being greater than 0.67.) Secondary (1) Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine will reduce the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 months or longer apart compared with GARDASIL in 16- to 26-yearold adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type. Hypothesis: Administration of 9-valent HPV L1 VLP vaccine to 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 months or longer apart compared with GARDASIL. (The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the vaccine efficacy be greater than 25%.) (2) Objective: To demonstrate that 9-valent HPV L1 VLP vaccine is immunogenic with respect to HPV types 31, 33, 45, 52, and 58. Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 31, 33, 45, 52 and 58 seroconversion rates 4 weeks postdose 3 that are acceptable in 16- to 26- year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The V503_001-02_ProtCore APPROVED 16-Aug-2011

125 V503, Protocol Issue Date: 16-Aug Product: V503 8 Protocol/Amendment No.: seroconversion rate for each of HPV types 31, 33, 45, 52, and 58 will be tested separately. Acceptability is defined as the lower bound of the two-sided 95% confidence interval for the seroconversion rate is greater than 90%.) (3) Objective: To demonstrate that the 9-valent HPV L1 VLP vaccine induces noninferior immune responses with respect to seroconversion percentages for HPV 6, 11, 16, and 18 compared to GARDASIL Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 6, 11, 16, and 18 seroconversion percentages by 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The seroconversion rates for each of HPV types 6, 11, 16, and 18 will be tested separately. Noninferiority is defined as the lower bound of the two-sided 95% confidence interval for the difference (9-valent vaccine minus GARDASIL in seroconversion rate being greater than -5 percentage points.) (4) Objective: To quantify the amount by which the administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58- related cervical, vulvar and vaginal disease compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (5) Objective: To evaluate the persistence of anti-hpv 6, 11, 16, 18, 31, 33, 45, 52, and 58 immune responses generated by 9-valent HPV L1 VLP vaccine. (6) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of Pap test abnormalities (ASC-US [Positive for High Risk HPV] or worse) Exploratory (1) Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 12 months or longer apart compared with GARDASIL - to 26-yearold adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). Hypothesis: Administration of 9-valent HPV L1 VLP vaccine to 16- to 26-yearold adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 12 months or longer apart compared with GARDASIL. (The statistical criterion for V503_001-02_ProtCore APPROVED 16-Aug-2011

126 V503, Protocol Issue Date: 16-Aug Product: V503 9 Protocol/Amendment No.: success requires that the lower bound of the two-sided 95% confidence interval for the vaccine efficacy be greater than 0.) (2) Objective: To evaluate whether the administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of persistent HPV 16 and 18 infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 months or longer apart and HPV 16- and 18-related cervical, vulvar, and vaginal disease. Hypothesis: 9-valent HPV L1 VLP vaccine is non-inferior to GARDASIL in reducing the combined incidence of HPV 16- and/or 18-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 months or longer apart and HPV 16- and 18-related cervical, vulvar, and vaginal disease compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the difference (9-valent vaccine minus GARDASIL in vaccine efficacy relative to historical placebo, for which the incidence rate is conservatively assumed to be 4 per 100 person-years as observed from the GARDASIL program, is greater than -15 percentage points.) (3) Objective: To evaluate whether the administration of 9-valent HPV L1 VLP vaccine results in a combined incidence of HPV 6- and 11-related cervical, vulvar, and vaginal disease that is comparable to the combined incidence observed with GARDASIL (4) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the combined incidence of CIN, AIS, and cervical cancer caused by any HPV type. (5) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the combined incidence of vulvar and vaginal disease caused by any HPV type. (6) Objective: To characterize the titers of anti-hpv type 6/11 in both peripartum maternal blood and in cord blood of infants born to subjects for evaluating the potential impact of 9-valent HPV L1 VLP vaccine on recurrent respiratory papillomatosis. 5 (7) Objective: To evaluate the efficacy of the selected formulation of 9-valent HPV L1 vaccine against persistent HPV 35-, 39-, 51-, 56- and 59-related infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 months or longer apart and HPV 35-, 39-, 51-, 56-, and 59-related cervical, vulvar, and vaginal disease. 5 This objective is applicable to study sites that are able to collect and submit peripartum and cord serum samples for analysis. V503_001-02_ProtCore APPROVED 16-Aug-2011

127 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: (8) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of Pap test abnormalities (ASC-US [Positive for High Risk HPV] or worse) related to HPV Types 31, 33, 45, 52, & 58. (9) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of cervical biopsy and cervical definitive therapy treatments. (10)Objective: To assess humoral immune responses to HPV type 6, 11, 16, 18, 31, 33, 45, 52, and 58 using a 9-valent HPV total IgG Luminex immunoassay upon availability of a validated assay. 2.2 SUBJECT/PATIENT INCLUSION CRITERIA To be randomized and receive the first study vaccination, subjects should meet all inclusion criteria. For items with an asterisk (*), if the subject does not meet these inclusion criteria, the Day 1 visit may be rescheduled for a time when these criteria can be met. 1. Subject is female, between the ages of 16 years and 0 days and 26 years and 364 days on the day of randomization. 2. Subject has never had Pap testing or has only had normal Pap test results. 3. Subject (or, for minor subjects, parent/legal guardian and subject) fully understands study procedures, alternative treatments available, the risks involved with the study, and voluntarily agrees to participate by giving written informed consent. 4. Subject is able to read, understand, and complete the vaccination report card. 5. Subject is judged to be in good physical health on the basis of medical history, physical examination, and laboratory results. 6. The subject has the following lifetime sexual history at the time of enrollment: a) Subject has had 1 to 4 male and/or female sexual partners; or b) Subject has had 0 male and/or female sexual partners, is 18 years of age or older, and plans to become sexually active within the first 3-6 months of the study. Male partner is defined as someone with whom the subject has penile penetrative sexual intercourse. Female partner is defined as someone who has contacted, either by penetrative (with fingers or other objects) or non-penetrative means, the subject s genitalia during sexual activity. 7. *Subject has refrained from douching/vaginal cleansing and using vaginal medications or preparations for 2 calendar days prior to the Day 1 visit. Subject agrees to refrain from these activities for 2 calendar days prior to any future visit that V503_001-02_ProtCore APPROVED 16-Aug-2011

128 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: includes collection of study specimens (cervical/genital swabs, Pap test, or biopsy/definitive therapy tissue). 8. *Subject has refrained from sexual activity (including anal, vaginal, or genital/genital contact whether same sex or opposite sex) for 2 calendar days prior to the Day 1 visit. Subject agrees to refrain from these sexual activities for 2 calendar days prior to any future visit that includes collection of study specimens (cervical/genital swabs, Pap test, or biopsy/definitive therapy tissue). 9. *Since the first day of the subject s last menstrual period through Day 1, the subject has not had sex with males or has had sex with males and used effective contraception with no failures (an example of a failure is a male condom that ruptures during sexual intercourse). Effective contraception is defined as a marketed, approved contraceptive product that the subject has used per the manufacturer s instructions with every act of sexual intercourse. The subject understands and agrees that during the Day 1 through Month 7 period, she should not have sexual intercourse with males without effective contraception, and the uses of the rhythm method alone, withdrawal alone, and emergency contraception, are not acceptable methods per the protocol. 2.3 SUBJECT/PATIENT EXCLUSION CRITERIA To be randomized and receive the first study vaccination, subjects should not have any exclusion criteria. For items with an asterisk (*), if the subject meets these exclusion criteria, the Day 1 visit may be rescheduled for a time when these criteria are not met. 1. Subject has a history of an abnormal cervical biopsy result (showing cervical intraepithelial neoplasia [CIN] or worse). 2. Subject has a history of a positive test for HPV. 3. Subject is, at the time of signing informed consent, a user of recreational or illicit drugs or has had a recent history (within the last year) of drug or alcohol abuse or dependence. Alcohol abusers are defined as those who drink despite recurrent social, interpersonal, and/or legal problems as a result of alcohol use. 4. Subject has a history of severe allergic reaction (e.g., swelling of the mouth and throat, difficulty breathing, hypotension or shock) that required medical intervention. 5. Subject has known allergy to any vaccine component, including aluminum, yeast, or BENZONASE (nuclease, Nycomed [used to remove residual nucleic acids from this and other vaccines]). For the purpose of this exclusion criterion, an allergy to vaccine components is defined as an allergic reaction that met the criteria for serious adverse experiences defined in Section 3.4. V503_001-02_ProtCore APPROVED 16-Aug-2011

129 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Subject is currently immunocompromised or has been diagnosed as having a congenital or acquired immunodeficiency, HIV infection, lymphoma, leukemia, systemic lupus erythematosus (SLE), rheumatoid arthritis, juvenile rheumatoid arthritis (JRA), inflammatory bowel disease, or other autoimmune condition. 7. Subject has had a splenectomy. 8. Subject is receiving or has received in the year prior to enrollment the following immunosuppressive therapies: radiation therapy, cyclophosphamide, azathioprine, methotrexate, any chemotherapy, cyclosporin, leflunomide (Arava TNF-α antagonists, monocolonal antibody therapies (including rituximab [Rituxan ]), intravenous gamma globulin (IVIG), antilymphocyte sera, or other therapy known to interfere with the immune response. With regard to systemic corticosteroids, a subject will be excluded if she is currently receiving steroid therapy, has recently (defined as within 2 weeks of enrollment) received such therapy, or has received 2 or more courses of high dose corticosteroids (orally or parenterally) lasting at least 1 week in duration in the year prior to enrollment. Subjects using inhaled, nasal, or topical corticosteroids are considered eligible for the study. 9. Subject has received any immune globulin product (including RhoGAM [Ortho- Clinical Diagnostics]) or blood-derived product within the 3 months prior to the Day 1 vaccination, or plans to receive any such product during Day 1 through Month 7 of the study. 10. *Subject has received non-replicating (inactivated) vaccines within 14 days prior to the Day 1 vaccination or has received replicating (live) vaccines within 21 days prior to the Day 1 vaccination. 11. Subject has thrombocytopenia or other coagulation disorder that would contraindicate intramuscular injections. 12. *Subject has donated blood within 1 week prior to the Day 1 vaccination, or intends to donate during Day 1 through Month 7 of the study. 13. Subject is expecting to donate eggs during Day 1 through Month 7 of the study. 14. Subject is concurrently enrolled in clinical studies of investigational agents or studies involving collection of cervical specimens. 15. Subject has received a marketed HPV vaccine, or has participated in an HPV vaccine clinical trial and has received either active agent or placebo. 16. Subject has a history or current evidence of any condition, therapy, lab abnormality or other circumstance that might confound the results of the study, or interfere with the subject s participation for the full duration of the study, such that it is not in the best interest of the subject to participate. V503_001-02_ProtCore APPROVED 16-Aug-2011

130 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Subject is unlikely to adhere to the study procedures, keep appointments, or is planning to relocate during the study. 18. *Subject has had a fever (defined as an oral temperature of 100 F or 37.8 C) within the 24-hour period prior to the Day 1 vaccination. 19. Subject is pregnant (as determined by a serum pregnancy test or urine pregnancy test that is sensitive to 25 miu/ml β-hcg). 20. *Subject has clinical evidence of gross purulent cervicitis. 21. *Subject is having menses. 22. Subject has a history of or clinical evidence at the Day 1 pelvic examination of HPVrelated external genital lesions (e.g., condyloma acuminata or vulvar intraepithelial neoplasia [VIN]) or external genital cancer, HPV-related vaginal lesions (e.g., condyloma acuminata or vaginal intraepithelial neoplasia [VaIN]) or vaginal cancer. 23. Subject does not have an intact cervix uteri or has more than one cervix uteri. 2.4 STUDY DESIGN AND DURATION Summary of Study Design This is a randomized, international, multi-centered, double-blinded (with in-house blinding procedures), dose-ranging, tolerability, immunogenicity, and efficacy study of the 9-valent HPV L1 VLP vaccine. This study will be conducted in two parts (Part A and Part B). Enrollment in Part A of the study is expected to be complete approximately 3 months after the first subject has been enrolled. Enrollment in Part B of the study is expected to be complete approximately 15 months after the first subject has been enrolled. Part A (dose-ranging and tolerability): In Part A (dose-ranging and tolerability), approximately 1,240 healthy 16- to 26-year-old women will be randomized in equal numbers to one of three 9-valent HPV L1 VLP vaccine formulations or the comparator GARDASIL in a three dose regimen (Day 1, Month 2, and Month 6). The primary goal in Part A is to identify the best 9-valent HPV L1 VLP vaccine dose for the V503 program which maintains the immunogenicity of GARDASIL against HPV types 6, 11, 16 and 18, while extending prophylactic vaccine coverage to include new HPV types 31, 33, 45, 52 and 58. A selected 9-valent HPV L1 VLP vaccine formulation should be generally well-tolerated, have HPV 6, 11, 16, and 18 GMTs that are comparable to the GARDASIL control, and be highly immunogenic for the new HPV vaccine types. Based on results for GARDASIL and other previously studied 3-dose (Day 1, Month 2, and Month 6) HPV L1 VLP vaccines, the ratio of the HPV 6, 11, 16, or 18 GMTs for the 9-valent HPV L1 VLP vaccine versus GARDASIL 1 month after the V503_001-02_ProtCore APPROVED 16-Aug-2011

131 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: second dose of vaccine (post-dose 2) is highly indicative of the ratio after the completion of the 3-dose series. The 9-valent HPV L1 VLP vaccine is an important medical advance, offering the potential of significant prophylactic cancer coverage in addition to that already provided by GARDASIL Therefore, in order to be able to make a decision about the immunogenicity of 9-valent HPV L1 VLP vaccine formulations in Part A and to move efficiently to evaluate a chosen formulation for efficacy in Part B of the study, the decision about which 9-valent HPV L1 VLP vaccine formulation to proceed with will occur after approximately 100% of the Part A post-dose 2 immunogenicity and adverse experience data are available in the clinical database. The dose for Part B will be chosen by a senior management committee at the SPONSOR (referred to as sidmc) who is not directly involved with study conduct. This committee will choose the dose after reviewing immunogenicity summaries by vaccination group, and will relay the dose selection to the Data Safety Monitoring Board (DSMB). The DSMB will review the immunogenicity data and all tolerability data, and will communicate to the SPONSOR senior management committee if, in the opinion of the DSMB, there is concern with respect to the incidence of serious adverse experiences, severe injection-site adverse experiences, or severe systemic adverse experiences for the selected dose formulation as compared to the incidence observed in the GARDASIL group. If no tolerability concerns are raised, the senior management committee will inform the study team of the selected dose formulation and the study will proceed to Part B with the selected formulation and the control arm GARDASIL If the DSMB expresses a concern about the tolerability profile of the selected dose, the senior management committee will select another dose and again request approval from the DSMB. Once the senior management committee and the DSMB agree on a chosen dose, the senior management committee at the SPONSOR will inform the study team of the 9- valent HPV L1 VLP vaccine formulation selected for Part B. The selected dose will be communicated to the study sites by letter before Part B vaccine supplies are shipped to the sites. Following selection of the 9-valent HPV L1 VLP vaccine for study in Part B, the Merck study team will notify the Interactive Voice Response System (IVRS) of the selected 9- valent HPV L1 VLP vaccine formulation. In addition to activation of Part B enrollment, notifying IVRS will allow sites involved in Part A to find out the status of Part A subjects. For subjects enrolled in Part A, as each subject completes their Month 7 visit and all study data has been collected, study sites will call IVRS to receive notification of each subject s status. For the subjects in Part A who receive the selected 9-valent HPV L1 VLP vaccine or the comparator GARDASIL the total duration of follow-up will be approximately 42 months from enrollment. For the subjects in Part A who receive either of the 9-valent HPV L1 VLP vaccine doses not selected for study in Part B, the total duration of followup will be approximately 7 months from enrollment. Subjects who receive a 9-valent HPV L1 VLP dose with a lower total VLP amount than the formulation that is selected for use in Part B may be offered a single dose of GARDASIL Final determination of whether an extra dose of GARDASIL is recommended will be made when V503_001-02_ProtCore APPROVED 16-Aug-2011

132 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: approximately 100% of the Part A post-dose 3 immunogenicity data are available in the clinical database. Even though some subjects will stop their visits at Month 7 while others continue to Month 42, the integrity of the blinding will be maintained because no vaccination group information will be released to the study sites for subjects continuing beyond Month 7. Therefore, the subjects continuing beyond Month 7 will remain randomized between the two continuing vaccination groups, the selected 9-valent HPV L1 vaccine and GARDASIL. The primary immunogenicity analysis in Part A will be conducted based on the Week 4 Postdose 3 (Month 7) immunogenicity results of the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B and the GARDASIL comparator. The formal statistical test of the Part A hypothesis based upon screened and cleaned Month 7 data will be provided in the final study report. An important goal in Part A of the study is to evaluate the tolerability of the 9-valent formulations relative to that of GARDASIL HPV L1 VLP vaccines formulated with Merck s adjuvant AAHS have been well-studied clinically, have been found to be generally well-tolerated, and are currently approved for use as a vaccine in the form of GARDASIL Because, however, this is the first time these specific formulations have been studied clinically, Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. personnel will perform an early and expedited review of blinded Vaccine Report Card (VRC) data once data for the first 40 subjects (approximately 10 subjects per arm) in Part A are available in the clinical database, and will request an unblinded review by the chairman of the Data Safety Monitoring Board (DSMB) if the rates of severe and/or serious adverse experiences are not consistent with the known adverse experience profile in the GARDASIL program. Subsequently, the DSMB, which consists of members that are independent of Merck Research Laboratories and a non-voting Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. statistician that is not associated with the study, will review VRC data when approximately 25% of Part A Day 1 vaccination visits are available in the study database, and will perform additional reviews at milestones specified in this protocol (See Section ). Amendment 01: Part A Dose Selection Summary: As planned, summaries of immunogenicity data by vaccination group were reviewed by a senior management committee at the SPONSOR (sidmc) who is not directly involved with study conduct. The senior management committee at the SPONSOR selected the 9-valent HPV L1 VLP (30/40/60/40/20/20/20/20/20 mcg each of HPV Types 6/11/16/18/31/33/45/52/58) vaccine for use in Part B. The sidmc relayed the dose selection to the DSMB. The DSMB reviewed the unblinded tolerability profile of all three 9-valent HPV L1 VLP vaccine formulations and the GARDASIL control group, concluded there were no safety concerns with any of the formulations, and agreed with the sidmc s selected dose for study in Part B. Because immunogenicity was comparably high for each of the 9- valent vaccine formulations, the group level summaries were provided to the Merck study team who facilitated the selection of the preferred 9-valent formulation for pivotal V503_001-02_ProtCore APPROVED 16-Aug-2011

133 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: efficacy study in Part B. The Merck study team members reviewed grouped immunogenicity data only and were not unblinded to any individual treatments, or any safety or efficacy data. Part B (tolerability, immunogenicity, and efficacy): In Part B (tolerability, immunogenicity, and efficacy), approximately 13,380 additional healthy 16- to 26-year-old women will be randomized in equal numbers to the selected 9- valent HPV L1 VLP vaccine from Part A or the comparator GARDASIL The primary immunogenicity analyses for Part B will be the formal comparison of the selected 9- valent formulation to the GARDASIL control and will include those subjects enrolled in Part B only. Because the immunogenicity data for Part A will have been analyzed previously, the Part A immunogenicity data for the selected 9-valent HPV L1 VLP vaccine and the comparator GARDASIL nogenicity analysis. Since neither the unblinded safety or efficacy data for Part A subjects will have been viewed previously by Merck personnel involved in the conduct of the study, the overall tolerability and efficacy analyses for Part B will combine the data from subjects enrolled in Part A (those who received the selected 9-valent HPV L1 VLP vaccine or the comparator GARDASIL and subjects enrolled in Part B. The total duration of the study for each subject enrolled in either Part A or Part B who receive the selected 9- valent HPV L1 VLP vaccine or GARDASIL is 42 months. The primary efficacy analysis is case driven and will be conducted after at least 30 primary efficacy cases have been observed. In addition, in order to have a sufficient duration of follow-up of vaccinated subjects, the primary efficacy analysis will not be performed before the median follow-up of Part B subjects is approximately 24 months after initial vaccination. For the subjects in Part B, the total duration of follow-up will be approximately 42 months from enrollment. Since the primary efficacy analysis will be conducted prior to completion of the study, the remainder of the study is considered an extension. The purpose of the extension is to collect data on the longer-term efficacy, tolerability, and immunogenicity of the vaccine. Additional Study Design Considerations: In Part A or Part B, subjects will receive an allocation number from an allocation schedule generated by the Clinical Biostatistics department of the SPONSOR. There will be separate allocation schedules for Part A and Part B. The randomization will be balanced within sites. At the time of enrollment, subjects must meet all inclusion criteria and must not meet any exclusion criteria. Subjects should not be enrolled from settings in which the typical clinical population has a high risk for HPV infection at baseline, such as sexually-transmitted disease clinics. Eligible study sites include community health centers, academic health centers, and the offices of primary health care providers. The subjects, investigators (and his/her staff), laboratory staff, members of the Scientific Advisory Committee, and HPV Vaccine Program Pathology Panel will remain blinded to subject vaccination group allocations for the duration of the study (~3.5 years). The SPONSOR will remain blinded to subject vaccination allocations for Part A and Part B V503_001-02_ProtCore APPROVED 16-Aug-2011

134 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: subjects who receive the selected 9-valent HPV L1 VLP vaccine or the comparator GARDASIL until the required number of cases of the primary efficacy endpoint have been observed and the database is unblinded for the primary efficacy analysis, with the exception of unblinded personnel who will provide data summaries for dose selection and DSMB meetings, and those who will determine when the required number of cases of the primary efficacy endpoint have been observed. These unblinded personnel will not be associated with the conduct of the study or the design of any of the statistical analyses for the study (other than those requested by the DSMB). For subjects who are enrolled in countries that have appropriate centralized registry infrastructures, results for Pap tests and biopsies taken outside of the context of the study will be obtained from the registries and certain data from the V database may be sent to these registries. Also, subjects enrolled in countries with these registries may be asked to participate in a sub-study for long-term follow-up. If this sub-study is performed, it will occur after a subject has completed her Protocol 001 scheduled study visits and will use cervical cancer screening registries to capture Pap test and biopsy results to assess the long-term duration of vaccine efficacy Vaccination Plan In Part A, approximately 1,240 healthy 16- to 26-year-old women will be randomized in equal numbers to one of 3 dose formulations of 9-valent HPV L1 VLP vaccine or the comparator GARDASIL -1, Section 1.6). In Part B, approximately 13,380 additional subjects will randomized in equal numbers to receive the 9-valent HPV L1 VLP vaccine selected in Part A or GARDASIL At Day 1, Month 2, and Month 6, subjects will receive 9-valent HPV L1 VLP vaccine or GARDASIL as a series of three 0.5-mL intramuscular injections (see Section for injection procedures). Study vaccine must be stored in a refrigerator that has a temperature between 2 C and 8 C (35.6 F and 46.4 F). If the refrigerator in which the study vaccine is stored deviates from the 2 C and 8 C (35.6 F and 46.4 F) range, study vaccinations should be suspended and the SPONSOR should be contacted immediately so that an investigation can be conducted. For a sample of the label and for more information on packaging, labeling, and storage of clinical supplies, see Section LIST OF IMMUNOGENICITY AND EFFICACY MEASUREMENTS Immunogenicity Measurements Serum will be collected from all subjects for analysis of anti-hpv 6, 11, 16, 18, 31, 33, 45, 52, and 58 responses by competitive Luminex Immunoassay (clia). Serum will be analyzed to support study objectives and to support assay development work. As an exploratory objective, serum may also be analyzed using a 9-valent HPV total IgG Luminex immunoassay upon availability of a validated assay. V503_001-02_ProtCore APPROVED 16-Aug-2011

135 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: At Day 1, serum specimens will be collected before the first study vaccination to identify subjects who have been exposed to study vaccine-specific HPV types prior to enrollment. After Day 1, serum specimens will be collected at visits specified in the Study Flow Chart. All subjects in Part A will have clia measurements summarized for dose selection and the Part A primary immunogenicity analysis. All subjects enrolled in Part B will be included in the Part B immunogenicity hypothesis testing at Month 7. For the purpose of demonstrating vaccine induced anti-hpv antibody persistence at Months 12, 24, 36, and 42, a random sample of approximately 20% of Part B subjects will be tested and analyzed. Serology samples collected from subjects at visits after Month 7 for subjects not included in the random sampling will be tested as needed. In addition to collection of serum for anti-hpv measurements to support the study objectives, an extra vial of serum will be collected at Month 7 for use in anti-hpv assay development and for use as a standard reference for anti-hpv assays. The assay development may include, but is not limited to, identifying potential biomarkers (especially a surrogate marker for protection), improving laboratory methods for future HPV antibody assays evaluating the isotypes or total anti-hpv immunoglobulin levels, or assessing the ability of HPV study vaccines to offer cross-protection against types of HPV that are not included in the vaccine. An experimental objective for the study is to characterize the levels of anti-hpv type 6/11 in both peripartum blood and in cord blood of infants born to subjects for evaluating the potential impact of 9-valent HPV L1 VLP vaccine on recurrent respiratory papillomatosis. Only some sites will be able to provide serum samples for this exploratory objective, because an investigator or sub-investigator associated with this study must provide complete obstetrical care for the subject, must deliver the infant, and approval must be obtained for collection of the samples at the facility where the infant is delivered Efficacy Measurements Labial/vulvar/perineal and perianal (LVPP) and endo/ectocervical (EEC) swabs will be tested for detection of HPV types 6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59 by Polymerase Chain Reaction (PCR) assay. PCR analysis of the swabs will be used to identify subjects who have an active HPV infection at enrollment and to determine persistent HPV infection endpoints. A subject with abnormal Pap Tests will be referred to colposcopy using a protocolmandated triage algorithm. The colposcopist will obtain cervical biopsies of areas of the most severe abnormalities and may choose to perform an endocervical curettage (ECC) if no cervical dysplasia is observed. A subject may need subsequent cervical definitive therapy per the protocol-mandated cervical definitive therapy guidelines. Vaginal lesions will be biopsied if they are observed during a Pap test, colposcopy, or at any time. A thorough external genital wart/lesion inspection will be performed at routine visit intervals. External genital warts/lesions noted after Day 1 will be biopsied. In addition, subjects with histologically confirmed HPV-related external genital warts (e.g., condyloma acuminata, VIN, cancer) or vaginal warts (e.g., condyloma acuminata, VaIN, V503_001-02_ProtCore APPROVED 16-Aug-2011

136 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: cancer) will be referred to colposcopy if the biopsies were not obtained during a colposcopy. Tissue obtained from biopsy and cervical definitive therapy procedures will be analyzed by MRL HPV Thinsection PCR assay (detection of HPV types 6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59) and by a consensus diagnosis from the HPV Vaccine Program Pathology Panel to determine clinical disease efficacy endpoints. The Day 1 LVPP/EEC swabs and tissue specimens will include PCR analysis for nonvaccine HPV types (35, 39, 51, 56, 59, and potentially others). This data may be used for calculation of incidence rates for disease associated with non-vaccine HPV types and for cross-protection analyses. 2.6 LIST OF SAFETY MEASUREMENTS Each subject will receive a Vaccination Report Card (VRC) at the Day 1, Month 2, and Month 6 study vaccination visits. On the VRC, the subject will be asked to record her oral temperatures in the evening after each study vaccination and daily, at the same time of day, for 4 days after each study vaccination for the purpose of identifying febrile events. Also, beginning after each study vaccination and for a total of 15 days including the day of vaccination, the subject will be asked to record injection-site and systemic adverse experiences, concomitant medications, and concomitant vaccinations on the VRC. Serious adverse experiences, pregnancy information, and lactation information will be collected as described in Section DATA ANALYSIS SUMMARY Part A The dose formulation for Part B will be selected based on an interim analysis of the immunogenicity data from Part A that will be conducted when ~100% of the 310 enrolled subjects in each vaccination group have data available through Week 4 Postdose 2. Anti-HPV 6, 11, 16, 18, 31, 33, 45, 52, and 58 geometric mean titers (GMTs) and seropositivity percentages for each of the 9-valent HPV L1 VLP vaccine formulations and the GARDASIL comparator will be summarized. All available tolerability information will be summarized. The interim analysis will be conducted in a subset of subjects who 1) are seronegative at Day 1 to the relevant HPV type, without regard to PCR status at Day 1; 2) receive the first 2 doses of 9-valent HPV L1 VLP vaccine or GARDASIL in acceptable day ranges; and 3) have post-vaccination serum samples collected in acceptable day ranges. Group-level immunogenicity summaries will be provided by the unblinded statistician to a senior management committee at the SPONSOR who is unrelated to the trial. No serology or randomization data for individual study participants will be disclosed to the senior management committee. Unblinded summaries of all tolerability data will be provided to the DSMB for detailed safety review. The senior management committee and the DSMB will select the dose using the process outlined in Section V503_001-02_ProtCore APPROVED 16-Aug-2011

137 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: The primary immunogenicity analysis in Part A will be conducted in type-specific per protocol immunogenicity populations, as defined in section The formal statistical test of the Part A hypothesis based upon screened and cleaned Month 7 data will be provided in the final study. The primary hypothesis test will be conducted to compare the 9-valent HPV L1 VLP vaccine formulation that is selected for use in Part B and GARDASIL. For each of HPV types 6, 11, 16 and 18, the primary immunogenicity hypothesis of noninferiority of the selected 9-valent HPV L1 VLP vaccine to GARDASIL Week 4 Postdose 3 will be tested by constructing a confidence interval for the fold difference in GMTs between the groups (9- valent/gardasil In order to declare that the formulation of 9-valent HPV L1 VLP vaccine selected for use in Part B is noninferior to GARDASIL, noninferiority needs to be established for each of the HPV types 6, 11, 16, and 18. The primary immunogenicity hypothesis will be tested at the nominal 1-sided α level of to adjust for the informal Postdose 2 interim summary. Because Part A is considered a pilot study, the statistical criterion for noninferiority with respect to GMTs requires that the lower bound of the confidence interval for the fold difference in GMTs exclude 0.5 (that the confidence interval rules out a decrease of 2-fold or more). With approximately 310 subjects per group, Part A has an overall power of >99%. Part B - The primary efficacy analysis in Part B will be conducted in type-specific per protocol efficacy populations. Subjects enrolled in Part A in the selected 9-valent HPV L1 VLP vaccine arm or GARDASIL arm will also be included in the safety and efficacy analyses in Part B. There will be a single primary analysis of efficacy which will be conducted after at least 30 primary efficacy cases have been observed. In addition, in order to have a sufficient duration of follow-up of vaccinated subjects, the primary efficacy analysis will not be performed before the median follow-up of Part B subjects is approximately 24 months after initial vaccination. A primary efficacy case is defined as a subject with an incident HPV 31, 33, 45, 52, or 58 related high-grade cervical abnormality (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer after 4 weeks postdose 3. The primary efficacy hypothesis will be addressed by constructing a 2-sided exact confidence interval for vaccine efficacy. The statistical criterion for success requires that the lower bound of the 95% confidence interval for vaccine efficacy exclude 25%. The primary efficacy hypothesis will be tested at the α=0.025 level (1-sided). Assuming a true vaccine efficacy relative to GARDASIL of 83% (accounting for the potential for GARDASIL to provide a ~30% reduction in the incidence of the primary endpoint due to cross-protection, and assuming 88% vaccine efficacy relative to Placebo), the study will have at least 90% power to achieve success for the primary hypothesis. The primary immunogenicity analysis will be performed in type-specific per-protocol immunogenicity populations in Part B, as defined in section No subjects from Part A will be used in the immunogenicity analyses in Part B. The analysis will be performed at the time when the database is unblinded for the primary efficacy analysis. For each of HPV types 6, 11, 16 and 18, the primary immunogenicity hypothesis of V503_001-02_ProtCore APPROVED 16-Aug-2011

138 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: noninferiority of 9-valent HPV L1 VLP vaccine to GARDASIL with respect to GMTs at 4 weeks postdose 3 will be tested by constructing a confidence interval for the ratio of GMTs between the groups (9-valent/GARDASIL The statistical criterion for noninferiority with respect to GMTs requires that the lower bound of the confidence interval for the GMT ratio exclude 0.67 (that the confidence interval rules out a decrease of 1.5-fold or more). The first primary immunogenicity hypothesis will be tested at the α=0.025 level (1-sided). With the planned sample size, this study will have >99% power for the primary immunogenicity hypothesis. Success in this study will be declared if the primary efficacy hypothesis and the primary immunogenicity hypothesis in Part B are achieved. The secondary efficacy hypothesis of HPV 31-, 33-, 45-, 52-, and 58- related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 months or longer apart, will be addressed using the same methodology as for the primary hypothesis, and the statistical criterion for success requires that the lower bound of the 95% confidence interval for vaccine efficacy exclude 25%. The secondary immunogenicity hypothesis of noninferiority of 9-valent HPV L1 VLP vaccine to GARDASIL with respect to percentage of subjects who seroconvert to each of HPV 6, HPV 11, HPV 16, and HPV 18 by 4 weeks Postdose 3 will be tested at the α=0.025 level (1-sided) for each HPV type by constructing a 95% confidence interval for the difference in the percentage of subjects who seroconverted between the 9-valent HPV L1 VLP vaccine group and the GARDASIL group. The analysis will be conducted in type-specific per protocol immunogenicity populations, as defined in section Seroconversion is defined as changing serostatus from seronegative to seropositive. A subject with a clia titer at or above the serostatus cutoff for a given HPV type is considered seropositive for that type. The statistical criterion for noninferiority requires that the lower bound of the confidence interval for the difference in seroconversion percentages is greater than -5 percentage points. For HPV 31, 33, 45, 52 and 58, the secondary hypothesis of acceptability of the 9-valent HPV L1 VLP vaccine with respect to seroconversion rates at 4 weeks postdose 3 will be tested by constructing a confidence interval for the seroconversion rates for the 9-valent HPV L1 VLP vaccine group. The statistical criterion for acceptability requires that the lower bound of the confidence interval exclude 0.9 (90%). The secondary immunogenicity hypothesis will be tested at the α=0.025 level (1-sided). V503_001-02_ProtCore APPROVED 16-Aug-2011

139 V503, Protocol Issue Date: 16-Aug Product: V503 1 Protocol/Amendment No.: Redacted 3.1 RATIONALE 3. PROTOCOL DETAILS V503_001-02_ProtDet APPROVED 16-Aug-2011

140 V503, Protocol Issue Date: 16-Aug Product: V503 2 Protocol/Amendment No.: Redacted V503_001-02_ProtDet APPROVED 16-Aug-2011

141 Redacted V503, Protocol Issue Date: 16-Aug Product: V503 3 Protocol/Amendment No.: V503_001-02_ProtDet APPROVED 16-Aug-2011

142 Product: V503 4 Protocol/Amendment No.: Redacted V503, Protocol Issue Date: 16-Aug V503_001-02_ProtDet APPROVED 16-Aug-2011

143 Redacted V503, Protocol Issue Date: 16-Aug Product: V503 5 Protocol/Amendment No.: V503_001-02_ProtDet APPROVED 16-Aug-2011

144 V503, Protocol Issue Date: 16-Aug Product: V503 6 Protocol/Amendment No.: STUDY PROCEDURES Concomitant Medication(s)/Treatment(s) See the exclusion criteria for specific restrictions for prior and concomitant medications at Day 1. Use of medicines and non-study vaccines should be documented in the data collection system in the following manner: Medications (corticosteroids, immunosuppressives, immune globulins, and blood products) from 3 days prior to Day 1 through Month 7; Medications from 3 days prior to each study vaccination through 14 days after each study vaccination; -Study Non-Replicating (Inactive) Vaccines for 14 days prior to each study vaccination through 14 days after each study vaccination; and -Study Replicating (Live) Vaccines for 21 days prior to each study vaccination through 14 days after each study vaccination. For the specific case where a subject mistakenly receives any non-study HPV vaccines only, such as CERVARIX (GlaxoSmithKline), GARDASIL (Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc.) or SILGARD (Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc.), the non-study HPV vaccine must be reported in the study data collector, regardless of when the non-study HPV vaccine was received. If possible, the subject should not receive Special Medications or Non-Study Vaccines within the time periods given above. Subjects may receive allergen desensitization therapy and tuberculin skin testing while participating in the study Prerequisites for Study Visits This section summarizes prerequisites for visits with pelvic specimen collection and visits with study vaccinations. Deviations from these prerequisites require consultation between the investigator and the SPONSOR and written documentation of the collaborative decision. See the Administrative Binder for a summary of deviations that require documentation in this study Prerequisites for Study Visits With Pelvic Specimen Collection See the inclusion/exclusion criteria for specific restrictions at Day 1. For visits that include collection of study specimens, study personnel should verify by questioning the subject and/or by examination that: V503_001-02_ProtDet APPROVED 16-Aug-2011

145 V503, Protocol Issue Date: 16-Aug Product: V503 7 Protocol/Amendment No.: The subject has refrained from douching/vaginal cleansing and using vaginal medications or preparations for 2 calendar days prior to any visit that includes collection of labial/vulvar/perineal and perianal (LVPP) and endo/ectocervical (EEC) swabs, a Pap test and/or biopsies/definitive therapy specimens. 2. The subject has refrained from sexual activity (including anal, vaginal, or genital/genital contact whether same sex or opposite sex) for 2 calendar days prior to any visit that includes collection of LVPP/EEC swabs, a Pap test and/or biopsies/definitive therapy specimens. 3. The subject does not have visible blood in the vagina. As a guide, a study visit that includes collection of LVPP/EEC swabs, a Pap test and/or biopsies/definitive therapy specimens should not be scheduled between the time period of 2 days prior to menses (if date of menses can be predicted) to 2 days after menses. If, despite using these collection timeframe guidelines, visible blood is noted in the vagina, then the specimens may be collected. 4. The subject does not have gross purulent cervicitis at a study visit that includes collection of LVPP/EEC swabs, a Pap test and/or cervical biopsies/definitive therapy specimens. If the subject does not meet the requirements listed above, the study visit (including specimen collection and/or study vaccination) should be re-scheduled. If the subject has gross purulent cervicitis, the re-scheduled study visit should occur after the condition is diagnosed and treated according to the study site s standards and practices. Prerequisites for Study Vaccination Visits See the inclusion/exclusion criteria for specific restrictions at Day 1. For visits with study vaccinations, study personnel should verify by questioning the subject and/or by examination that: 1. The subject has not had a fever (defined as an oral temperature of 100 F or 37.8 C) within the 24-hour period prior to any study vaccination visit. 2. The subject has not received any systemic (oral or parenternal) corticosteroids in the 2 weeks prior to the Month 2 and Month 6 study vaccination visits. 3. The subject has not received a Non-Study Non-Replicating (Inactive) Vaccine within 14 days prior to any study vaccination visit or a Non-Study Replicating (Live) Vaccine within 21 days prior to any study vaccination visit. If the subject does not meet the requirements listed above, the study visit (including specimen collection and study vaccination) should be re-scheduled. V503_001-02_ProtDet APPROVED 16-Aug-2011

146 V503, Protocol Issue Date: 16-Aug Product: V503 8 Protocol/Amendment No.: Summary of Scheduled Study Visit Procedures The Study Flow Chart summarizes procedures for scheduled study visits and items appear in the order they should be performed. This section provides clarifications to the scheduled study visit procedures (presented in order of the Study Flow Chart) Calculation of Scheduled Visit Windows The visit windows provided in this protocol are for site scheduling purposes. The visits windows used for exclusion from analyses will be provided in the Statistical Analysis Plan (SAP). The Day 1 visit is defined as the day that the first study vaccination is given. The Month 2 visit is based on Day 1 and has a 3 week visit window. The Month 3 visit has a visit window from 3 weeks after Month 2 through 5 weeks after Month 2. The Month 6 visit is based on Day 1 and has a 4 week visit window. The Month 7 visit has a visit window from 3 weeks after Month 6 through 7 weeks after Month 6. Post Month 7 visits are each based on Day 1 and have a 4 week visit window. To calculate visit windows, assume 1 month equals 30 days and 1 week equals 7 days Informed Consent and Assignment of Baseline Numbers Written consent will be obtained for each subject (or, for minors, from the parent/legal guardian and the subject) prior to enrollment, study data collection, or study procedures. A copy of the signed informed consent form will be given to each subject for her records. Verification of the subject s identity and age is to be determined prior to obtaining written consent. Any government-issued photo identification will suffice for verification purposes and should be documented in the subject s file. If the date of the Day 1 visit is later than the consent date, the interval between the date of the consent and the date of the Day 1 visit should be no more that 14 days apart. If this interval is 15 days or longer, then the subject must be re-consented. Baseline numbers will be supplied by the SPONSOR via the electronic data capture (EDC) system. Each subject will be assigned a unique baseline number immediately after the informed consent is signed. Baseline numbers will be assigned sequentially from the lowest number to the highest number at each site. Baseline numbers will not be skipped or reassigned for any reason. A single subject cannot be assigned more than 1 baseline number. During the Day 1 visit, one may discover a condition that will make the subject ineligible for the study. For example, exclusionary medical conditions mentioned in the inclusion/exclusion questions may be found during medical history review or during the pelvic examination. Thus, all pre-vaccination data must be collected and all study procedures must be completed before a subject is randomized (i.e. assigned an allocation number using Interactive Voice Response System [IVRS]) and vaccinated. V503_001-02_ProtDet APPROVED 16-Aug-2011

147 V503, Protocol Issue Date: 16-Aug Product: V503 9 Protocol/Amendment No.: History & Medications At the Day 1 visit, the subject s lifetime gynecologic history and medical history for the year prior to Day 1 will be collected. After the Day 1 visit, any new gynecologic history and new medical history that have not been previously documented (either as adverse experiences [AEs] or as medical history conditions) will be collected. In addition to documentation of new medical history and AEs, concomitant and prior medications will be handled as described in Section Other documentation, such as demographics, sexual history, pregnancy history, and contraceptive use, will be collected in the data collection system, as discussed in the ecrf (Electronic Case Report Form) Entry Guidelines Pregnancy Testing and Serum Collection A serum pregnancy test or urine pregnancy test (sensitive to 25 miu/ml beta human chorionic gonadotropin [β-hcg]) will be performed at each Day 1 through Month 7 visit. The pregnancy test results will be obtained prior to each study vaccination on the day the subject is vaccinated. Any subject found to be pregnant at the Day 1 visit will not be randomized and will not participate in the study. For randomized subjects who become pregnant after receiving one or two study vaccinations, study visits and study vaccinations will be paused until resolution of the pregnancy (e.g., term, elective termination, spontaneous abortion). Study visits and study vaccinations in pregnant subjects will be handled as described in Table 3-1. V503_001-02_ProtDet APPROVED 16-Aug-2011

148 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Table 3-1 Guidelines for Pregnant Subjects: Managing Study Visits and Study Vaccinations Visit Where Pregnancy is Action Detected Day 1 Do not randomize subject Between Day 1 and Month 2 No scheduled visits until resolution of the pregnancy (e.g., term, elective termination, spontaneous abortion). The Month 2 study vaccination should be administered 4 weeks following resolution of pregnancy and after normalization of β- hcg levels. The Month 6 study vaccination should be administered 4 months after the Month 2 study vaccination. The Month 7 visit should be conducted 1 month after the Month 6 study vaccination. Between Month 2 and Month 6 No scheduled visits until resolution of the pregnancy (e.g., term, elective termination, spontaneous abortion). The Month 6 study vaccination should be administered 4 weeks following resolution of pregnancy and after normalization of β- hcg levels. The Month 7 visit should be conducted 1 month after the Month 6 study vaccination. After Month 6 Continue with scheduled study visits during the pregnancy. Completion of protocol-specified study procedures in pregnant subjects is at the investigator s discretion. The Month 7 visit should be conducted 1 month after the Month 6 study vaccination. After resolution of the pregnancy (e.g., term, elective termination, spontaneous abortion), the subject will undergo all protocolspecified procedures. The use of a cytobrush (Cytobrush for ThinPrep Pap test collection is contraindicated in women who are 10 weeks pregnant or more, according to the current manufacturer s label. The use of a broom (Wallach Papette for ThinPrep Pap test collection is not contraindicated in pregnancy, according to the current manufacturer s label. The use of a Wallach Papette instead of Cytobrush will be allowed if the investigator decides to perform a ThinPrep Breastfeeding is not a contraindication to enrollment or to receiving study vaccinations. Pregnancy and breast-feeding in study subjects and infant serious adverse experiences (SAEs) must be reported as described in Section 3.4 and the Attachments. If allowed at the study site, an umbilical cord blood sample must be collected within 15 minutes of birth and a blood sample from the subject must be collected shortly before or at the time of delivery for anti-hpv testing (see Section ). Serum for anti-hpv measurements must be collected at Day 1, and at Months 3, 7, 12, 24, 36, and 42. For collection of serum samples, the study site must adhere to the procedures described in this protocol, must follow instructions provided by the SPONSOR/Central Laboratory, and must use the materials provided by the SPONSOR/Central Laboratory (see Section 3.2.5). V503_001-02_ProtDet APPROVED 16-Aug-2011

149 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Physical Examination & Vital Signs A complete physical examination will be conducted at Day 1 and Months 7, 18, 30, & 42. Physical examination details will be documented in the subject s chart and any medical conditions will be documented in the data collection system. Vital signs (sitting pulse, weight, blood pressure, respirations, and oral temperature) will be taken before each study vaccination. The pre-vaccination oral temperature will be documented in the data collection system and the other vital signs will be documented in the subject s chart. If the subject has a fever (defined as an oral temperature of 100 F or 37.8 C) within the 24-hour period prior to receiving a study vaccination, the subject should not receive study vaccine, and the vaccination visit should be rescheduled Gynecological Examination A complete gynecological examination will be performed at Day 1 and Months 7, 18, 30, & 42. A sterile, non-lubricated, single use, individually wrapped, plastic speculum should be used. If lubrication of the speculum is needed, only use sterile saline or sterile water. All study samples (See Section ) must be taken from the pelvic area before the bimanual pelvic examination is performed to prevent contamination of study samples. If the subject has gross purulent cervicitis at a visit with specimen collection (LVPP/EEC swabs, a Pap test, and/or cervical biopsy/definitive therapy specimens), the visit should be re-scheduled for a time after the condition is diagnosed and treated according to the study site s standards and practices. Gynecologic conditions other than gross purulent cervicitis, such as vaginitis, bacterial vaginosis, and vaginal yeast infections, do not require a visit to be re-scheduled. These conditions should be diagnosed and treated according to the study site s standards and practices and treatment should be given after study sample collection and/or after study vaccination Genital/Cervical Swabs, Pap Test, and Sexually Transmitted Infection (STI) Testing Two (2) labial/vulvar/perineal and perianal (LVPP) swabs for HPV Polymerase Chain Reaction (PCR) assay, 1 endo/ectocervical (EEC) swab for PCR assay, and a ThinPrep Pap test for liquid-based cytology Papanicolaou testing will be collected at Day 1 and Months 7, 12, 18, 24, 30, 36 & 42. At Day 1 and Months 7, 18, 30, & 42, tests for Chlamydia and gonorrhea are performed on the Pap test fluid. For collection of swabs and the ThinPrep Pap test, the study site must adhere to the procedures described in this protocol, must follow instructions provided by the SPONSOR/Central Laboratory, and must use the materials provided by the SPONSOR/Central Laboratory (see Section 3.2.5). A non-lubricated, single use, individually wrapped, plastic speculum should be used for pelvic specimen collection. If lubrication of the speculum is needed, only use sterile saline or sterile water. During the Pap test, an inspection for vaginal lesions should be V503_001-02_ProtDet APPROVED 16-Aug-2011

150 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: performed. Per the exclusion criteria, a clinical impression of HPV-related vaginal lesions (e.g., condyloma acuminata or VaIN) will exclude a subject at Day 1. However, if these types of vaginal lesions are identified after Day 1, the lesions should be classified and managed as described in Section ThinPrep Pap tests will be submitted to the SPONSOR-designated Central Laboratory. Cytology specimens will be evaluated using The Bethesda System For a diagnosis of atypical squamous cells of undetermined significance (ASC-US), the Central Laboratory will perform reflex testing for high-risk/low-risk HPV probes (Digene Hybrid Capture II Assay) on residual ThinPrep material. Site staff, subjects, and MRL personnel will remain blinded to the identity of the probe. The Pap test diagnoses generated by the Central Laboratory will be reported to the study sites for subject management, and subjects will be referred to colposcopy according to a protocolmandated triage algorithm (see Table 3-2). All Pap test reports must be reviewed and signed by a physician (M.D./D.O.) investigator/sub-investigator. At Day 1 and Months 7, 18, 30, & 42, tests for Chlamydia and gonorrhea are performed on the Pap test fluid and processed through the SPONSOR-designated Central Laboratory, but additional local laboratory tests for Chlamydia and gonorrhea may be performed at other visits at the discretion of the investigator if clinically indicated. For these additional local laboratory tests for Chlamydia and gonorrhea, the study site will collect samples using their own materials. Tests for other sexually transmitted infections (STIs), including HSV, syphilis, Hepatitis B, and HIV, may be performed at any visit at the discretion of the investigator if clinically indicated. However, per the exclusion criteria for this study, known HIV-positive subjects should not be enrolled. Any STI sampling of the cervix or external genital region should occur after the swabs and Pap test are obtained. Subjects who are enrolled into the study and test positive for STIs may remain in the study. All subjects who test positive for STIs should be referred for counseling and treatment. It is important to document all tests for STIs on the appropriate ecrf as discussed in the ecrf Entry Guidelines External Genital Lesion Examination After obtaining all study specimens (swabs and Pap test), and after a change of gloves, an external genital examination should be performed using a good light source and a handheld magnifying glass (4x to 5x power) or, optionally, a colposcope with low-power magnification. The following guidance should be used for all external genital lesion examinations performed within the study: 1. The examiner should ask the subject if she has noticed any bumps, lesions, or unusual symptoms (e.g., itching, discomfort, dyspareunia, dysuria). 2. The labia should be spread, and the condition of the hymen and vulvovaginal skin and clitoris are to be examined and noted. Note abnormalities, such as abnormal hair growth distributions, abnormalities of the skin, rashes, lacerations, or bruises. V503_001-02_ProtDet APPROVED 16-Aug-2011

151 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: The examination should include a systematic inspection of the entire external genitalia (periurethral, perineal, perianal, and vulvar regions) for apparent cutaneous and subcutaneous lesions or wart-like growths. 4. Acetic acid is not to be used routinely. It may be used for confirmation of a suspected lesion. 5. All observations should be recorded into the subject s chart. Medical conditions and lesions should be recorded on the appropriate ecrfs. Per the exclusion criteria, a clinical impression of any HPV-related external genital lesions (e.g., condyloma acuminata or VIN) will exclude a subject at Day 1. However, if these types of external genital lesions are identified after Day 1, the lesions should be classified and managed as described in Section IVRS Assignment of Allocation Numbers and Vaccination Vials The Interactive Voice Response System (IVRS) will assign the subject an allocation number and then subsequently assign a unique vial identification number for the vial of clinical material the subject should receive at that visit. The IVRS will automatically assign the appropriate clinical material based on the subject s treatment allocation. Randomization will be balanced within each study site. The study personnel will access IVRS at each subsequent visit when administration of vaccine is to occur for assignment of a unique vial identification number for the clinical material to be administered to the subject. In addition, for Part A subjects, sites must contact IVRS at the completion of each subject s Month 7 visit to determine those subjects from Part A who will continue in the study to Month 42 and those who will complete the study at their Month 7 visit. A single patient/subject cannot be assigned more than 1 allocation number. Randomization deviations (e.g., a subject is assigned two allocation numbers by mistake through IVRS) require written documentation. See the Administrative Binder for a summary of deviations that require documentation in this study Study Vaccine Administration Preparation for Administration The study vaccine must be used as supplied (no dilution before administration). Prior to administration, mix the contents of the vial thoroughly by rolling the vial between the palms of both hands for 30 seconds. Withdraw a 0.5-mL dose from the vial, which contains approximately 0.75 ml of study vaccine. The study vaccine should be a whitish, semi-translucent suspension when thoroughly mixed. If the appearance is otherwise, do not administer, and contact the SPONSOR immediately V503_001-02_ProtDet APPROVED 16-Aug-2011

152 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Study Vaccine Administration Study vaccine will be administered at Day 1, Month 2, and Month 6. At each visit, subjects will receive 9-valent HPV L1 VLP vaccine or GARDASIL as a 0.5-mL intramuscular injection. The deltoid muscle of the nondominant arm is the preferred site of vaccination. Study vaccinations should not be administered into the buttocks area. If the subject has received injectable contraceptives (e.g., DEPO-PROVERA TM ) or implanted contraceptives (e.g., NORPLANT in both arms in the past 9 months, the injection may be given in the thigh rather than in an arm. If the subject has an implanted contraceptive (e.g., NORPLANT he time of study vaccination, avoid administering the study vaccination in the arm with the implanted contraceptive. Injections should not be given within 2 cm of a tattoo, scar, or skin deformity. Study vaccine should be administered using a 1.0-mL syringe. Injections should be administered at a 90 angle into the muscle tissue using a needle long enough to ensure intramuscular deposition of vaccine. For injection into the arm muscle, use the following needle length and gauge specifications: 1-inch needle, 22 to 23 gauge, for women weighing <200 pounds (<90.9 kg.) or 1 ½-inch needle, 22 to 23 gauge for women weighing 200 pounds ( 90.9 kg). For injection into the thigh muscle, a 1 ½-inch needle, 22 to 23 gauge, should be used Clinical Follow-Up Please refer to Section 3.4 for Vaccine Report Card (VRC) information and information on AE reporting Summary of Unscheduled Study Visit Procedures Colposcopy, Cervical Biopsy, and Cervical Definitive Therapy Subjects with abnormal ThinPrep Pap tests will be referred to colposcopy and biopsy according to the protocol-mandated triage algorithm. In addition, subjects with histologically confirmed HPV-related external genital warts (e.g., condyloma acuminata, VIN, cancer) or vaginal warts (e.g., condyloma acuminata, VaIN, cancer) will be referred to colposcopy if the biopsies were not obtained during a colposcopy. If a colposcopy is required per the protocol-mandated triage algorithm, the colposcopy must be performed within 2 months of receipt of the abnormal result by the study investigator. If abnormalities of the cervix are found during colposcopy, cervical biopsies will be taken of the areas with the most severe abnormalities. If no cervical dysplasia is observed, the colposcopist may choose to perform an endocervical curettage (ECC). A vaginal exam is required during all study colposcopies. Additional Pap testing during the unscheduled colposcopy study visit is not allowed and is considered a deviation from the protocolmandated triage algorithm. A summary of colposcopy referral is given in Table 3-2. V503_001-02_ProtDet APPROVED 16-Aug-2011

153 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Table 3-2 Protocol-Mandated Triage Algorithm for Referral to Colposcopy ThinPrep Pap Result Action Negative for intraepithelial lesion or malignancy (includes reactive, reparative, inflammatory, etc.) Routine visit interval as specified by the protocol. Atypical Squamous Cells Undetermined Central laboratory performs reflex HPV testing Significance (ASC-US) on residual ThinPrep ial (High Risk and Low Risk Probe, Hybrid Capture II, DIGENE no result is obtained, the subject will be referred for colposcopy. If both probes are negative, the subject will return for Pap screening at the routine visit interval. Atypical Squamous Cells cannot exclude HSIL Referral to colposcopy. (ASC-H) Low-grade Squamous Intraepithelial Lesion (LSIL) Referral to colposcopy. High-grade Squamous Intraepithelial Lesion (HSIL) Referral to colposcopy. Atypical Glandular Cells (to include atypical Referral to colposcopy. endocervical, endometrial, NOS; Adenocarcinoma in Situ, adenocarcinoma, etc.) Unsatisfactory Repeat ThinPrep but not less than 4 weeks from the Pap test that had the unsatisfactory finding. External Genital Biopsy Result Action HPV-related (e.g., condyloma acuminata, VIN, cancer) Vaginal Biopsy Result HPV-related (e.g., condyloma acuminata, VaIN, cancer) Referral to colposcopy if the biopsy was not obtained at a colposcopy. Action Referral to colposcopy if the biopsy was not obtained at a colposcopy. Notes: Additional guidance for the protocol-mandated triage algorithm, including follow-up for discordant cases, is given in the Mandatory Regimen for Triage Attachment. Deviations to the protocol-mandated triage algorithm for colposcopy, including repeat Pap tests at unscheduled colposcopy study visits, require consultation between the investigator and the SPONSOR and written documentation of the collaborative decision. Deviations that require this documentation include a colposcopy that is performed for another medical reason, such as unresolved post-coital bleeding that persists after completion of the appropriate work-up. In addition, if a deviation is discovered by the SPONSOR or study investigator after an error is made, retrospective written documentation is required. See the Administrative Binder for a summary of deviations that require documentation in this study. If the Pap test diagnosis is AGC (to include AIS, atypical endocervical cells, adenocarcinoma, etc.), ASC-H, or HSIL and the subsequent biopsy diagnosis is CIN 1 or less, then this is defined as a discordant case. A discordant diagnosis will require a review of the results by the Sponsor-designated Central Laboratory. Upon completion of the review, a consultation report will be sent to the investigator. If the results are amended upon completion of this review, the protocol-mandated colposcopy triage V503_001-02_ProtDet APPROVED 16-Aug-2011

154 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: algorithm (Table 3-2) and guidelines for cervical definitive therapy (Table 3-3) should be followed. If the discordant results are reviewed but not amended, see Table 3-3 and the Mandatory Regimen for Triage Attachment (Figure 4) for guidelines. Subjects will be referred for cervical definitive therapy according to Table 3-3. If cervical definitive therapy is performed, on the day of the procedure, pre-definitive cervical biopsies must be taken of the areas with the most severe abnormalities prior to performing the cervical definitive therapy. Loop Electrosurgical Excision Procedure (LEEP) is the preferred method for cervical definitive therapy. LEEP has contraindications for subjects with the following conditions: allergy to all local anesthetics; pregnancy; severe acute cervicitis (severe acute cervicitis should be diagnosed and treated, and excision conducted after the infection has resolved); obvious invasive cancer; significant glandular neoplasia (AGUS favor neoplasia, AIS, Adenocarcinoma); and microinvasive cancer. In addition to LEEP, laser conization is also acceptable if it is the standard of care at the study site. Cold-knife conization and ablative cervical definitive therapy (e.g., cervical cryotherapy or cervical laser vaporization) should be reserved for rare instances where cervical definitive therapy is required and LEEP or laser conization is not indicated. LEEP, laser conization, and coldknife conization are study procedures, but ablative cervical definitive therapy is not a study procedure because this procedure is not tissue preserving. Table 3-3 Protocol-Mandated Guidance for Referral to Cervical Definitive Therapy Biopsy Results that Require Cervical Definitive Therapy Cervical biopsy result or ECC result of CIN 2, CIN 3, Adenocarcinoma in Situ (AIS), or cervical cancer. Cervical biopsy (including ECC, if taken) result of CIN 1 on at least 2 consecutive biopsies for a duration of 18 months or longer. Discordant Case Guidance for Cervical Definitive Therapy (After Discordant Case Review Confirms Discrepancy) High Grade Pap Result (AGC [to include AIS, atypical endocervical cells, adenocarcinoma, etc.)], ASC-H, or HSIL), negative vaginal examination, cervical biopsy/ecc is CIN 1 (or less) or if no lesion is seen and cervical biopsy/ecc is not taken: If colposcopic examination was unsatisfactory- Cervical Definitive Therapy required. If colposcopic examination was satisfactory- See Attachment (Mandatory Regimen for Triage, Figure 4) for guidance. Management of women 20 years of age should be consistent with local standards of care and may include observation in these instances. Two repeat High Grade Pap diagnoses (regardless of whether they are consecutive or not) without a confirmed cervical biopsy or ECC diagnosis of CIN 2, CIN 3, or cervical cancer will be referred to definitive therapy. General Notes: Pre-definitive cervical biopsies must be obtained on the day of the cervical definitive therapy procedure and must be taken before the cervical definitive therapy procedure is performed. Ablative cervical definitive therapy is not a study procedure. Deviations to the protocol-mandated guidance for referral to cervical definitive therapy and the footnotes in this table require consultation between the investigator and the SPONSOR and written documentation of the collaborative decision. In addition, if a deviation is discovered by the SPONSOR or study investigator after an error is made, retrospective written documentation is required. See the Administrative Binder for a summary of deviations that require documentation in this study. If cervical biopsies, pre-definitive cervical biopsies, endocervical curettage (ECC) specimens, or cervical definitive therapy specimens are obtained, they must be sent to the SPONSOR-designated Central Laboratory for analysis. Separate biopsy forceps must be used for each of the discrete lesions that are biopsied. See Section for detailed guidance on colposcopy, biopsy, ECC, and cervical definitive therapy procedures. V503_001-02_ProtDet APPROVED 16-Aug-2011

155 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Subjects who undergo cervical biopsy and/or cervical definitive therapy will continue to be followed through the completion of scheduled study visits. The interval between a cervical biopsy and/or cervical definitive therapy and the next scheduled visit that includes a pelvic exam must be at least 2 months. Follow-up care after cervical definitive therapy will occur according to the study site s standards and practices. Colposcopy, biopsy, and ECC procedures must be performed by an experienced health care professional (>50 colposcopies per year for at least 2 years). The cervical definitive therapy provider should be an experienced physician (defined as performing 20 LEEP procedures per year for at least 2 years) and must have documented formal instruction [i.e., residency or postgraduate course or American Society for Colposcopy and Cervical Pathology (ASCCP) training course]. To ensure that colposcopy practices and skill levels are standardized across study sites, and that colposcopists at all sites develop a uniform approach to all study colposcopies, study colposcopists will be required to participate in colposcopic standardization training. The purpose of this training is to evaluate the colposcopic practices at all study sites, to standardize colposcopy and biopsy practices, and to provide a guide for all aspects of the protocol related to colposcopy and histological sampling of the cervix. An alternative to the SPONSOR-organized colposcopy training, the Colposcopy Course given by the American Society for Colposcopy and Cervical Pathology (ASCCP), will be accepted as equivalent for the purpose of standardization to participate in this study, provided it has been taken within the year prior to study start External Genital Lesion and Vaginal Lesion Diagnosis and Follow-Up External genital lesions or vaginal lesions may be identified during a scheduled visit exam, Pap test, colposcopic examination, or at any other time during the study. For the purposes of this study, lesions that require access via the introitus (with the exception of the cervix, which is considered a separate location) are to be defined as vaginal lesions. Lesions that are external to the introitus are to be defined as external genital lesions. To classify the location of external genital lesions, the external genital area will be further divided into the following 4 regions: periurethral (includes the clitoris) perineal perianal vulvar (includes all areas other than the periurethral, perineal or perianal regions) It is important to make an accurate evaluation of any identified external genital lesions. The practitioner must classify a lesion as condyloma acuminata, other HPV-related lesion (e.g., a flat wart, a papular wart, a keratotic wart, VIN, Bowenoid papulosis, Bowen s disease) or a non HPV-related lesion. If lesions are identified, they must be documented on the appropriate ecrf. V503_001-02_ProtDet APPROVED 16-Aug-2011

156 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: External genital lesions suspected to be possibly, probably, or definitely HPV-related (i.e., clinical impression of condyloma acuminata or other HPV-related lesion) or of unknown etiology are to be biopsied. Separate biopsies should be performed if external genital lesions are at different locations (periurethral, perineal, perianal, or vulvar region) or if more than one external genital lesion is at the same location and has a different morphology. An additional external genital lesion biopsy should be obtained if a new HPV-related lesion (or lesion of unknown etiology) appears and is of differing morphology and/or differing location from the initial lesion(s). A recurrent lesion should not be biopsied. A recurrent external genital lesion is defined as a lesion that has both of the following characteristics: (1) appears within 2 months of the complete resolution of an initial lesion(s) and (2) has the same location and a similar morphology as the initial lesion(s). If an HPV-related lesion or a lesion of unknown etiology appears after 2 months of complete resolution of the initial lesion(s), a biopsy of the lesion should be taken. Vaginal lesions suspected to be possibly, probably, or definitely HPV-related (i.e., condyloma acuminata or other HPV-related lesion) or of unknown etiology are to be biopsied. Separate biopsies should be performed for each vaginal lesion with a different morphology. An additional vaginal biopsy should be obtained if a new potentially or definitely HPV-related lesion (or lesion of unknown etiology) appears and is of differing morphology from the initial lesion(s). A recurrent lesion should not be biopsied. A recurrent vaginal lesion is defined as a lesion that has both of the following characteristics: (1) appears within 2 months of the complete resolution of an initial lesion(s) and (2) has a similar morphology as the initial lesion(s). If an HPV-related lesion or a lesion of unknown etiology appears after 2 months of complete resolution of the initial lesion(s), a biopsy of the lesion should be taken. External genital lesion biopsies and vaginal lesion biopsies must be sent to the SPONSOR-designated Central Laboratory for analysis. Separate biopsy forceps must be used for each of the discrete lesions that are biopsied. Biopsy of an external genital lesion or vaginal lesion should ideally be performed on the same day that the lesion is first observed, because the lesion may resolve before the subject returns for a biopsy procedure if it is performed at a later time. See Section for detailed guidance on these procedures. Subjects with histologically confirmed HPV-related external genital warts (e.g., condyloma acuminata, VIN, cancer) or vaginal warts (e.g., condyloma acuminata, VaIN, cancer) will be referred to colposcopy if the external genital or vaginal biopsies were not obtained during a colposcopy (see Section 3.2.5). Subjects who undergo external genital lesion biopsy, vaginal lesion biopsy, and any subsequent treatment for the lesions will continue to be followed through the completion of scheduled study visits. After a biopsy is obtained, management and treatment of external genital lesions and vaginal lesions are at the investigator s discretion and according to the study site s standards and practices. If excision is chosen as the method of treatment, all excised tissue is to be submitted to the SPONSOR-designated Central Laboratory for analysis. The interval between the external genital lesion biopsy/excision or vaginal lesion biopsy/excision and the next V503_001-02_ProtDet APPROVED 16-Aug-2011

157 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: scheduled visit involving a pelvic exam must be at least 2 months. External genital lesion biopsy/excision and vaginal lesion biopsy/excision must be performed by an experienced health care professional Processing of Tissue Specimens In the Context of the Study Extra visits that are associated with tissue excision (cervical, vaginal, external genital, and cervical definitive therapy samples) will be considered part of the study. Any specimens collected will be obtained as described in the protocol and collected and shipped as described in the Laboratory Manual provided by the SPONSOR/Central Laboratory. Slides of tissue specimens will be prepared at the SPONSOR-designated Central Laboratory and reviewed by a pathologist, and reports will be provided to the study sites for subject management. All tissue slides will be reviewed by the HPV Vaccine Program s expert Pathology Panel, which will consist of up to 5 pathologists. The consensus diagnosis of HPV Vaccine Program Pathology Panel will represent the final diagnosis for study purposes and not for subject management. Also, all tissue will be tested at MRL for detection of HPV types by PCR assay (See Section 3.3.2). The SPONSOR and the HPV Vaccine Program Pathology Panel will follow established guidelines for review of the slides. The investigator will be notified of any histological diagnosis made by the HPV Vaccine Program Pathology Panel that is more severe than the one made by the SPONSOR-designated Central Laboratory Pap Tests and Tissue Specimens Taken Outside the Context of the Study Cervical, vaginal, and external genital tissue samples and Pap tests collected outside the context of the study are strongly discouraged. For procedures with sample collection, rather than through the SPONSOR-designated Central Laboratory. For procedures without sample collection, such as a colposcopy without biopsy, the context of the study a non-study clinician. If a subject undergoes a study procedure outside the context of the study, all efforts will be made to obtain and send the following to the SPONSOR: (1) Pap or colposcopy/operative report; (2) diagnostic slides (for tissue biopsies only); (3) local pathology report; and (4) tissue block for diagnostic slide preparation and HPV analysis Procedures for Collection and Handling of Study Specimens For scheduled study visits, consult the Study Flow Chart for the specific samples needed for each visit and the order of sample collection. The following Sections describe the step-by-step procedures for collection of study specimens, a description of supplies needed, and the guidelines for handling specimens. Samples should be shipped, labeled, and handled as instructed by the SPONSOR/Central Laboratory. Specimen collection supplies provided by the SPONSOR/Central Laboratory must be used by the site without substitution. V503_001-02_ProtDet APPROVED 16-Aug-2011

158 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Within 30 minutes of collection, place serum and LVPP/EEC swabs in a freezer at -20 C (or lower) until the samples are shipped on dry ice. If the samples thaw, contact the SPONSOR. Thawed serum/swab samples require written documentation, including details such as allocation number, visit interval, date of collection, and sample accession numbers (see the Administrative Binder for a summary of deviations that require documentation in this study). For collection of pelvic specimens (EEC swab, Pap test, cervical biopsy/definitive therapy), a non-lubricated, single use, individually wrapped, plastic speculum should be used. If lubrication of the speculum is needed, only use sterile saline or sterile water may be used Serum or Urine Specimen for Pregnancy Test The serum pregnancy test or urine pregnancy test (sensitive to 25 miu/ml β-hcg) will be performed per the manufacturer s instructions. All materials used for serum and urine pregnancy testing will provided by the study sites Serum for Anti-HPV Measurements at Scheduled Visits For each visit that requires a serum specimen for anti-hpv measurements, a 10-mL (nonheparinized, non-serum separator, red-top tube provided by the SPONSOR-designated Central Laboratory) blood specimen will be collected and should be separated to avoid hemolysis. A minimum of 3.0 ml of serum should be aliquoted to a vial provided by the SPONSOR-designated Central Laboratory. An additional 1.5 ml of serum ( Serum ) should be aliquoted to a vial provided by the SPONSOR-designated Central Laboratory. At Month 7, an additional 10-mL (non-heparinized, non-serum separator, red-top tube provided by the SPONSOR-designated Central Laboratory) blood specimen will be collected and should be separated to avoid hemolysis. The serum ( eference Serum, approximately 4.5 ml) should be transferred to a vial provided by the SPONSORdesignated Central Laboratory. Serum and Serum vials will be stored at the site at -20 C (or lower) until shipped on dry ice. Serum serum vials will remain stored at the site at -20 C (or lower) until the SPONSOR notifies the study site to ship the vials. All available serum should be used for conducting assays specified in the clinical protocol. Serum and retention serum may also be used, during or after the clinical trial, for further HPV immunologic testing in addition to tests specified in the protocol Cord Serum and Maternal Serum for Anti-HPV Measurements (Applicable to Participating Study Sites) The collection of cord blood should not affect the care of the mother or child, and there should be no significant deviation from normal procedures. Blood from the mother should be collected by venipuncture shortly before or at the time of delivery. Cord blood should be collected according to local standard procedures. Collections of cord blood are V503_001-02_ProtDet APPROVED 16-Aug-2011

159 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: generally more successful if initiated within 15 minutes of the birth of the infant. The placenta need not be delivered in order to collect cord blood. Clean the site of the cord that will be punctured with antiseptic swabs. The needle of a syringe can be inserted into the cleansed area of the umbilical vein. For the cord blood or peripartum maternal blood, a 10-mL (non-heparinized, non-serum separator, red-top tube provided by the SPONSOR-designated Central Laboratory) blood specimen will be collected and serum should be separated to avoid hemolysis. A minimum of 3.0 ml of cord serum or peripartum maternal serum should be aliquoted to a vial provided by the SPONSOR-designated Central Laboratory. An additional 1.5 ml ("Retention Serum" vials) of cord serum or peripartum maternal serum should be aliquoted to a vial provided by the SPONSOR-designated Central Laboratory. Serum vials will be stored at the site at -20 C (or lower) until shipped on dry ice. The etention Serum serum vials will remain stored at the site at -20 C (or lower) until the SPONSOR notifies the study site to ship the vials Labial/Vulvar/Perineal and Perianal (LVPP) Swabs for HPV PCR Use DACRON swabs and Specimen Transport Medium (STM) vials supplied by the SPONSOR-designated Central Laboratory. When collecting swab specimens, remember to keep all swab tips and swab shafts (i.e. the portion that will be placed in the STM) untouched. 1. Remove DACRON 2. Using the first swab, swab back and forth in a tight zigzag motion from the clitoral prepuce down to the posterior fourchette on first one side and then the other so two parallel zigzag paths down the perineum will allow collection between the folds of the labia minora and majora. 3. With the second swab, swab the perianal area. 4. Place both swabs in the appropriately labeled collection/transport tube containing STM. Break off the end of the shafts protruding from the tube by bending them sharply against the rim of the tube. The shafts are pre-scored to facilitate breakage. 5. Securely cap the collection/transport tube containing the specimen. 6. Ensure proper labeling, store the sample at -20 C (or lower), and ship the sample as specified by the SPONSOR/Central Laboratory Endo/Ectocervical (EEC) Swab for HPV PCR Use DACRON swabs and Specimen Transport Medium (STM) vials supplied by the SPONSOR-designated Central Laboratory. When collecting swab specimens, remember to keep all swab tips and swab shafts (i.e. the portion that will be placed in the STM) untouched. V503_001-02_ProtDet APPROVED 16-Aug-2011

160 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Remove one DACRON 2. Introduce the swab into the cervical os with enough pressure to maintain contact with the epithelium but not to induce bleeding. Twirl the swab only 1 to 2 times, and then use a back-and-forth swiping motion across the ectocervix from anterior to posterior cervical lip (top to bottom). 3. Place the swab into the collection/transport tube containing the STM. Break off the end of the shaft protruding from the tube by bending it sharply against the rim of the tube. The shaft is pre-scored to facilitate breakage. 4. Securely cap the collection/transport tube containing the specimen. 5. Ensure proper labeling, store the sample at -20 C (or lower), and ship the sample as specified by the SPONSOR/Central Laboratory Pap Test (ThinPrep Use PreservCyt the SPONSOR-designated Central Laboratory. 1. Using a plastic spatula, scrape the ectocervix in a 360 arc for 2 full rotations, being sure to include the squamocolumnar junction in the portion sampled. Place the spatula into the open PreservCyt vial, using vigorous shaking and swishing of the spatula to rinse all of the cellular debris from the spatula. It is acceptable to leave the spatula in the open PreservCyt tobrush collection is completed. 2. Insert a cytobrush 1 into the cervical os, and slowly rotate 1/2 turn one direction. DO NOT OVER-ROTATE. Do not insert deeper than the length of the brush head. Place the cytobrush immediately into the PreservCyt 3. Use the spatula to rub against the cytobrush in order to dislodge as much cellular debris from the cytobrush and spatula as possible. Swish and rinse the cytobrush, and knock it against the side of the PreservCyt vial at least 10 times to release enough epithelial cells for an adequate Pap test sample. In order to minimize adherence of cellular debris to the collection devices, the rinsing and swishing procedures should be performed within 2 minutes of collection. 4. Remove the spatula and the cytobrush from the PreservCyt the vial securely. 5. Ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. Store PreservCyt vials containing 1 The use of a cytobrush (Cytobrush for ThinPrep Pap test collection is contraindicated in women who are 10 weeks pregnant or more, according to the current manufacturer s label. The use of a broom (Wallach Papette for ThinPrep Pap test collection is not contraindicated in pregnancy, according to the current manufacturer s label. The use of a Wallach Papette instead of Cytobrush will be allowed if the investigator decides to perform a ThinPrep V503_001-02_ProtDet APPROVED 16-Aug-2011

161 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: the specimens at room temperature until they are shipped to the Central Laboratory. According the manufacturer s instructions, ThinPrep specimens must be processed within 21 days of collection to preserve sample integrity. To meet this requirement, specimens must be shipped according to instructions and time intervals specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. 6. Prior to removal of the speculum, inspect the lateral walls of the vagina for vaginal lesions. Then collapse the speculum and rotate 90 degrees and re-open the blades to inspect the anterior and posterior walls of the vagina for vaginal lesions. If vaginal lesions are present, use the guidance given in the protocol to classify vaginal lesions and to obtain required biopsies (see Sections and ). Collapse and remove the speculum. 7. Change gloves and perform a systematic inspection of the entire external genitalia using a good light source and a hand-held magnifying glass (4x to 5x power) or, optionally, a colposcope with low-power magnification. If external genital lesions are present, use the guidance given in the protocol to classify external genital lesions and to obtain required biopsies (see Sections and ) External Genital Lesion Biopsy and Vaginal Lesion Biopsy Use the appropriate ecrfs to note locations, clinical impressions, and biopsy numbers during the procedure. Use the specimen collection kits (labels/requisition) and formalin fixative container provided by the SPONSOR-designated Central Laboratory. The external genital biopsy kit should be used for external genital biopsies and the vaginal biopsy kit should be used for vaginal biopsies. 1. Cleanse the biopsy area with antiseptic solution. Using a 30-gauge needle and a syringe of 1 to 2 ml of 1% lidocaine or bacteriostatic saline, infiltrate below the epidermis of the lesion. 2. Elevate the lesion and remove tangentially with fine (iris) scissors or a scalpel blade to obtain the specimen. 3. Place the biopsy piece in the fixative container and ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. 4. If multiple biopsies are obtained, each specimen must be placed in a separate container of fixative. To prevent cross-contamination, a different set of instruments is to be used for each biopsy taken. A maximum of 6 external genital lesion biopsies and 3 vaginal biopsies should be obtained. 5. If the biopsy procedure does not remove the entire lesion(s) and surgical excision is chosen as a treatment method, the excised tissue is to be submitted to the Central Laboratory in a separate container of fixative and labeled as an additional biopsy. V503_001-02_ProtDet APPROVED 16-Aug-2011

162 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: To promote hemostasis after the procedure, apply gentle pressure. Monsel s solution may be used. For larger areas, a single interrupted suture may be used. Electrocauterization is to be avoided, but the decision is left to the discretion of the practitioner. 7. Advise the patient to keep the area clean and to expect spotting for 3 days with healing by 1 week. 8. On the day of collection, ship the sample at room temperature as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory Colposcopy Guidelines 1. Assist patient to dorsal lithotomy position. 2. Insert speculum and visualize the entire cervix through the colposcope at low power. 3. Use optional visualization adjuncts (i.e., condom or rubber glove finger over speculum, lateral side-wall retractor) if necessary to enhance colposcopic visualization. 4. Remove mucus and debris by liberally applying 5% acetic acid to the cervix using cotton balls and ring forceps, large cotton swabs, or by spray technique. Avoid use of 4 4 gauze pads. 5. Reapply acetic acid (or Lugol s solution) to the cervix for a minimum of 60 seconds. Thereafter, reapply acetic acid every 3 to 5 minutes or when the columnar epithelium is no longer blanched white. 6. Identify the entire squamocolumnar junction (360 ), if able. 7. Identify acetowhite cervical lesions if present. 8. Assess the severity of each cervical lesion using green filter as needed to examine the vascular patterns. 9. As a guide, use the Reid Colposcopic Index (RCI), which is provided in the Administrative Binder, to classify any identified cervical lesions. 10. Obtain cervical biopsies and, if medically indicated, performed an endocervical curettage (ECC) sample collection (see below for cervical biopsy and ECC procedures). 11. Prior to removal of the speculum, inspect the lateral walls of the vagina with the colposcope. If clinically indicated, apply 5% acetic acid or Lugol s solution to the entire epithelium and then view by low power colposcopic magnification, noting acetowhite (or brownish if Lugol s solution used) vaginal lesions. Collapse the speculum and rotate 90 degrees and re-open the blades to inspect the anterior and V503_001-02_ProtDet APPROVED 16-Aug-2011

163 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: posterior walls of the vagina for vaginal lesions (again using the 5% acetic acid or Lugol s solution if clinically indicated). If vaginal lesions are present, use the guidance given in the protocol to classify vaginal lesions and to obtain required biopsies (see Sections and ). Collapse and remove the speculum. 12. Change gloves and perform a systematic inspection of the entire external genitalia using a good light source and a hand-held magnifying glass (4x to 5x power) or, optionally, a colposcope with low-power magnification. If external genital lesions are present, use the guidance given in the protocol to classify external genital lesions and to obtain required biopsies (see Sections and ) Procedures for Cervical Biopsy Use the appropriate ecrfs to note locations, clinical impressions, and biopsy numbers during the procedure. Use the cervical biopsy kit (labels/requisition) and formalin fixative provided by the SPONSOR-designated Central Laboratory. 1. Each discrete abnormal area that is observed on colposcopy should be biopsied. The most severe area of abnormality of a lesion observed on colposcopy should be biopsied. 2. Apply local anesthesia, if this is the local standard of care. 3. If taking multiple biopsies, start with the most posterior (i.e. lower) lesion first. This will prevent contamination of subsequent biopsy sites by the flow of blood from the previous biopsy sites. 4. Place biopsy forceps over the abnormal lesion (usually near the squamocolumnar junction). 5. Open forceps jaws to sufficient extent. 6. Check to make sure the forceps are properly aligned (fixed end of forceps jaw towards cervical os). 7. Rotate biopsy handles to place lesion in the center of biopsy jaw angle if necessary. 8. Exert moderate pressure with biopsy forceps to push cervix backwards until cervix is fixed in position. 9. Squeeze handles together quickly and firmly to close forceps jaws and excise lesion. 10. Lock biopsy jaws to secure tissue (if locking mechanism available) or just hold tightly, then pass the forceps to the assistant. 11. Confirm by colposcopic visualization that an appropriate and adequate biopsy was collected. The biopsy should be perpendicular to the epithelium and deep enough to V503_001-02_ProtDet APPROVED 16-Aug-2011

164 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: sample the entire epithelium along with a small amount of stroma (at least 2 mm) for histology. 12. Tamponade biopsy site with cotton swab. 13. Place the biopsy piece in the fixative container and ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. 14. Repeat the biopsy procedure for each additional biopsy performed using a different pair of forceps. If multiple biopsies are obtained, each specimen must be placed in a separate fixative container. A maximum of 4 cervical biopsies should be obtained. Collect all biopsy specimens prior to establishing hemostasis, if possible. 15. Obtain complete hemostasis with directed silver nitrate stick or Monsel s (ferric subsulfate) paste application. 16. Instruct subject to: (1) take nonsteroidal anti-inflammatory drugs for uterine cramping (provided no allergy), (2) report significant bleeding immediately, and (3) refrain from use of vaginal products, douching, tampons, and sexual intercourse for 3 to 7 days. 17. On the day of collection, ship the sample at room temperature as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. If no lesion is noted on colposcopy, and the subject was referred based on a cytological abnormality, endocervical curettage (ECC) may be performed Procedures for Endocervical Curettage (ECC) Use the cervical biopsy kit (labels/requisition) and formalin fixative provided by the SPONSOR-designated Central Laboratory. A curet must be used (not a cytobrush) when obtaining the ECC specimen. The decision to perform an ECC will be made by the study investigator according to local standards and practices. 1. Obtain an endocervical curettage specimen either before or after endocervical biopsies (based on local standard of care). 2. Insert open basket endocervical curet into cervical os. 3. Apply gentle pressure at the distal tip of curet against the endocervical canal and pull the curet from inside to outside with pressure, while simultaneously rotating the curet in a circular direction. The curet should advance no more that 20 mm up the canal and should be rotated completely 2 to 3 times. Avoid sampling ectocervical lesions that extend proximally into the endocervical canal, if possible. 4. Spin the curet rapidly to trap epithelium in the basket and remove the curet from the canal. Scraped material remaining in the canal can be retrieved with small forceps. V503_001-02_ProtDet APPROVED 16-Aug-2011

165 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Place the ECC specimen into the fixative vial (or place the ECC specimen on paper or Telfa and transfer it to the fixative vial) and ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. 6. On the day of collection, ship the sample at room temperature as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory Procedures for Loop Electrosurgical Excision Procedure (LEEP) and Top Hat Method The LEEP is a cervical definitive therapy method that uses a shallow pass to remove the cervical transformation zone. It may be a single or multiple pass and is ~8 mm deep. A hat cone implies a deeper wedge resection, ~15 to 25 mm deep. It is performed by the y more internal and smaller loops. Use the appropriate ecrfs to note clinical impressions and biopsy numbers during the procedure. Use the definitive therapy kit (labels/requisitions) and formalin fixative provided by the SPONSOR-designated Central Laboratory. The LEEP should be done under colposcopic control. All instruments used for the procedure must be nonconductive. 1. Assist subject into dorsal lithotomy position. 2. Place dispersive pad near operative site (thigh) and plug into electrosurgical unit (ESU). 3. Place vaginal speculum and secure smoke evacuation tubing. 4. Set electrosurgical unit parameters (power, cut/blend) and test safety systems. 5. Identify cervical pathology by colposcopic examination. 6. If appropriate, apply half-strength Lugol s solution to cervix and identify transformation zone limits. 7. Give local anesthesia if no allergy 4 quadrant (locations 3, 6, 9, 12 o clock) intracervical injection of lidocaine with epinephrine. 8. Obtain pre-definitive cervical biopsies of the areas with the most severe abnormalities (2- to 3-mm size per location) and use the cervical biopsy labels/requisition provided within the definitive therapy kit (see Section for cervical biopsy guidelines). 9. Select appropriate loop electrode size to remove transformation zone and lesion, insert into hand piece and plug hand piece into electrosurgical unit. 10. Activate smoke evacuator. V503_001-02_ProtDet APPROVED 16-Aug-2011

166 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Initiate excision by depressing switch if using handpiece, cut down (perpendicular to tissue), across transformation zone and straight out at opposite side of transformation zone (depth of 8 mm) and avoid vaginal sidewall contact. 12. Repeat cut procedure (reduced power) along endocervical canal with smaller 10 x 10 loop electrode if top hat conization is necessary. 13. Hand excised tissue to assistant and dictate its orientation. Orient the tissue by placing a suture at the 12 o clock location. Place tissue in the fixative container and ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. Note: If the top hat procedure is performed, the tissue from each loop or should be placed in separate labeled fixative containers. 14. Inspect endocervical canal opening and perform ECC, if medically indicated. Use the ECC label/requisition provided within the definitive therapy kit (see Section for ECC guidelines). 15. Fulgurate base of the excision with coagulation using the ball or paddle electrode until adequate hemostasis and fulgurate 5 mm of the ectocervical margin. If lesion extends beyond the excision, ablate the area and record the ablation on the appropriate ecrf and source document. 16. If hemostasis is inadequate, pack base of excision with Monsel s (ferric subsulfate) paste or in rare event suture may be required. 17. Remove blood from posterior fornix. 18. Deactivate ESU and smoke evacuator. 19. Remove dispersive pad and vaginal speculum. 20. Assist subject up from table. 21. Instruct subject to: (1) take nonsteroidal anti-inflammatory drugs as needed (provided no allergy), (2) refrain from use of vaginal products, douching, tampons, and sexual intercourse for 2 to 4 weeks, and (3) report significant bleeding and signs of infection (fever, pain) immediately. 22. On the day of collection, ship the sample at room temperature as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory Procedure for Laser Conization Laser Conization is a cervical definitive therapy procedure that uses a laser beam to excise tissue, using small depressor instruments or hooks to position the cone for optimal excision. V503_001-02_ProtDet APPROVED 16-Aug-2011

167 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Use the appropriate ecrfs to note clinical impressions and biopsy numbers during the procedure. Use the definitive therapy kit (labels/requisitions) and formalin fixative provided by the SPONSOR-designated Central Laboratory. 1. Assist subject into dorsal lithotomy position. 2. Place vaginal speculum and secure smoke evacuation tubing. 3. Set laser unit parameters (power, optical focusing of the laser beam) and test safety systems. 4. Identify cervical pathology by colposcopic examination. 5. If appropriate, apply half-strength Lugol s solution to cervix and identify transformation zone limits. 6. Give local anesthesia if no allergy 4 quadrant (3, 6, 9, 12) intracervical injection of lidocaine with epinephrine. If needed paracervical injection of the anesthetic may also be used. General anesthesia may be provided if so preferred. If general anesthesia is preferred, intracervical injection of lidocaine with epinephrine may still be used if there is no allergy to the components of the injection. 7. Obtain pre-definitive cervical biopsies of the areas with the most severe abnormalities (2- to 3-mm size per location) and use the cervical biopsy labels/requisition provided within the definitive therapy kit (see for cervical biopsy guidelines). 8. Activate smoke evacuator. 9. Initiate excision by marking the border of the area to be excised with the laser beam. 10. Cut the cone by the laser beam, using small depressor instruments or hooks to position the cone for optimal excision with the laser beam. 11. Hand excised tissue to assistant and dictate its orientation. Orient the tissue by placing a suture at the 12 o clock location. Place tissue in the fixative container and ensure proper labeling of the sample as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory. 12. Inspect endocervical canal opening and perform ECC, if medically indicated. Use the ECC label/requisition provided within the definitive therapy kit (see Section for ECC guidelines). 13. Use the defocused laser beam in the wound area until adequate hemostasis, and vaporize 5 mm of the ectocervical margin. If lesion extends beyond the excision, ablate the area and record the ablation on the appropriate ecrf and source document. V503_001-02_ProtDet APPROVED 16-Aug-2011

168 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: If hemostasis is inadequate, inject more lidocaine with adrenaline or in rare event suture may be required. 15. Remove blood from posterior fornix. 16. Remove dispersive pad and vaginal speculum and deactivate smoke evacuator. 17. Assist subject up from table. 18. Instruct subject to: (1) take nonsteroidal anti-inflammatory drugs as needed (provided no allergy), (2) refrain from use of vaginal products, douching, tampons, and sexual intercourse for 2 to 4 weeks, and (3) report significant bleeding and signs of infection (fever, pain) immediately. 19. On the day of collection, ship the sample at room temperature as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory Cold-Knife Conization Cervical cold-knife conization should be performed only when the LEEP or laser conization procedures are not indicated. A cervical cold-knife conization specimen represents a conically shaped section of the cervix that will vary in size according to the lesion. A broad, shallow conization would be performed for a predominantly exocervical lesion and a narrow, deep conization is appropriate for a predominantly endocervical lesion. Pre-definitive biopsies should be obtained prior to the cervical cold-knife conization. Use the appropriate ecrfs to note clinical impressions and biopsy numbers during the procedure. Use the definitive therapy kit (labels/requisitions) for pre-definitive biopsy/ecc (if indicated)/definitive therapy tissue and use the formalin fixative provided by the SPONSOR-designated Central Laboratory. The samples should be labeled/shipped as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory (ship at room temperature, ship on day of collection) Ablative Cervical Definitive Therapy Ablative cervical definitive therapy is not a study procedure, because this procedure is not tissue preserving. If a subject undergoes ablative cervical definitive therapy (e.g., cervical cryotherapy or cervical laser vaporization) or a hysterectomy during the study, obtain pre-definitive biopsy specimens. Use the definitive therapy kit (labels/requisitions) for pre-definitive biopsy/ecc (if indicated) and use the formalin fixative provided by the SPONSOR-designated Central Laboratory. The samples should be labeled/shipped as specified in the Laboratory Manual that is provided by the SPONSOR/Central Laboratory (ship at room temperature, ship on day of collection) Discontinuation/Withdrawal from Study Subjects/patients may withdraw at any time or be dropped from the study at the discretion of the investigator should any untoward effects occur. In addition, a subject/patient may be withdrawn by the investigator or the SPONSOR if he/she violates the study plan or for V503_001-02_ProtDet APPROVED 16-Aug-2011

169 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: administrative and/or other safety reasons. The investigator or study coordinator must notify the SPONSOR immediately when a subject/patient has been discontinued/withdrawn due to an adverse experience (telephone or FAX). When a subject/patient discontinues/withdraws prior to study completion, all applicable activities scheduled for the final study visit should be performed at the time of discontinuation. Any adverse experiences which are present at the time of discontinuation/withdrawal should be followed in accordance with the safety requirements outlined in section 3.4 SAFETY MEASUREMENTS - DETAILS. If a subject must discontinue/withdraw prior to the Month 12 visit, the site must attempt to contact the subject approximately 6 month s following the subject s last vaccination to ascertain whether the subject has experienced any serious adverse experiences. Serious adverse experiences must be reported as described in Section If a subject must discontinue/withdraw after the Month 7 study visit, the subject should be asked to return for a final visit if it has been at least 4 months since the last study visit. This visit would consist of the same specimen collections and tests conducted at the last study visit and the subject would be formally discontinued from the study at the end of this visit. The discontinuation visit should not be done if it is medically contraindicated or if the subject refuses. If no discontinuation visit is performed, the subject should be formally discontinued from the study on the day the decision to discontinue is made. All attempts must be made to contact a subject who is lost to follow-up (a certified letter must be sent at the final attempt). Subjects who are lost to follow-up should be formally discontinued from the study on the day of the last unsuccessful attempt at contact. At a minimum, the cover sheet and subject status ecrfs must be completed and submitted to the SPONSOR for all discontinuations Subject Relocation Given the duration of the study and the age of the study population, it can be expected that subjects may relocate during the study. The SPONSOR must be contacted for each temporary and permanent relocation as soon as the situation is known. Every effort should be made to adjust study visits around a subject s temporary absence (e.g., college breaks, summer vacation) so that the visits will be within the visit windows. Every effort should be made to have a relocated subject seen at another site participating in this study in order to keep the visits within the visit windows and to allow the subject to complete the study Conduct of the Clinical Trial The Scientific Advisory Committee will provide scientific input and direction. Clinical efficacy endpoints will be adjudicated by the HPV Vaccine Program Pathology Panel, who is responsible for the definitive pathologic diagnoses in all clinical conditions which may be considered as possible endpoints in subjects in this trial. The Data and Safety V503_001-02_ProtDet APPROVED 16-Aug-2011

170 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Monitoring Board will be responsible for monitoring tolerability and, in conjunction with the Senior Management Committee, will participate in the selection of the best Part A 9- valent HPV L1 VLP vaccine dose for use in Part B of the trial. The operation of the Scientific Advisory Committee, the HPV Vaccine Program Pathology Panel, the Data and Safety Monitoring Board, and the Senior Management Committee are described below Scientific Advisory Committee The SAC (Scientific Advisory Committee) is a body of internal representatives and external scientific leaders in the field of gynecology, oncology, and/or HPV research. The SAC will convene to provide advice on the protocol direction, Statistical Analysis Plan, and to interpret and publish the study results of confirmatory studies. The SAC will be comprised of external leaders and internal representatives HPV Vaccine Program Pathology Panel The HPV Vaccine Program Pathology Panel will be responsible for providing the definitive pathologic diagnoses of cervical biopsies, vaginal biopsies, external genital lesion biopsies, endocervical curettage, and cervical definitive therapy specimens for study purposes (not for medical management). Cervical histology slides and slides from external genital lesion biopsies and vaginal lesion biopsies will be evaluated by HPV Vaccine Program Pathology Panel. The HPV Vaccine Program Pathology Panel will prepare reports on each tissue specimen without knowing the vaccination groups of the subjects. A separate guideline that details the HPV Vaccine Program Pathology Panel process has been approved by the HPV Vaccine Program Pathology Panel Responsibility of the Data and Safety Monitoring Board (DSMB) The Data and Safety Monitoring Board (DSMB) assesses the effects of the study vaccine during the trial. The DSMB will review all available tolerability data at the following milestones during the Part A and Part B vaccination periods: 1. Milestone 1: VRC data from approximately 25% of Part A Day 1 vaccination visits are available in the study database, 2. Milestone 2: VRC data from approximately 100% of Part A Day 1 vaccination visits are available in the study database, 3. Milestone 3: VRC data from approximately 25% of Part A Month 2 vaccination visits are available in the study database, 4. Milestone 4: VRC data from approximately 100% of Part A Month 2 vaccination visits are available in the study database (this milestone will involve review of the dose selected by the senior management committee, as discussed in Section ), V503_001-02_ProtDet APPROVED 16-Aug-2011

171 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Milestone 5: VRC data from approximately 100% of Part A Month 6 vaccination visits are available in the study database (review will include all available VRC data from Part B), 6. Milestone 6: VRC data from approximately 25% of Part B Month 2 vaccination visits are available in the study database, 7. Milestone 7: VRC data from approximately 50% of Part B Month 2 vaccination visits are available in the study database, 8. Milestone 8: VRC data from approximately 100% of Part B Month 2 vaccination visits are available in the study database. In addition to above milestones, the DSMB may also meet on an ad-hoc basis at the request of the DSMB chair, the protocol team, the site, or a study investigator. With the exception of a non-voting Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. statistician, the members of the DSMB are independent of Merck Research Laboratories and clinical investigators participating in this trial, and will not have any other involvement in the study, nor will they have any relation to the study participants. The DSMB monitors the trial for evidence of adverse effects of the study vaccine using the guidelines proposed by the protocol. The DSMB may recommend any steps to ensure the safety and integrity of the trial. Furthermore, the DSMB may recommend that the trial be terminated or that specific high-risk patient groups be withdrawn from the study, if any subgroup manifests serious or widespread side effects. To guarantee the unrestricted performance of its task, the DSMB may receive the individual study morbidity and mortality data from a designated MRL statistician. Unblinded reports of all serious adverse experiences in this study from the MRL Worldwide Adverse Experience database and the clinical database will be presented to the DSMB. The monitoring guidelines that the DSMB will follow will be described in the Statistical Analysis Plan. The DSMB minutes will summarize the actions and deliberations of the DSMB and will be made available to Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. at the conclusion of the trial. A full manual of operations for the DSMB will be written and approved by the DSMB Senior Management Committee for Dose Selection As outlined in section 2.4.1, when ~100% postdose 2 Part A data are available in the clinical database, a senior management committee will choose the dose for Part B after reviewing immunogenicity summaries by vaccination group, and will relay the dose selection to the DSMB, who will approve the decision based on unblinded tolerability data. Details of the process for dose selection and the procedures used to maintain blinding will be provided in a guidance document, which will be approved before the senior management committee or the DSMB convenes to review the dose-selection data. V503_001-02_ProtDet APPROVED 16-Aug-2011

172 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: The senior management committee for dose selection will include a small number of Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. personnel who are experts in medical, regulatory, statistical and epidemiological aspects of clinical trials. Members will not be involved in study conduct. The senior management committee for dose selection will follow a protocol-specific charter of operations, which the committee will approve before convening to review dose-selection data. 3.3 EFFICACY/IMMUNOGENICITY MEASUREMENTS Competitive Luminex Immunoassay (clia) - Anti-HPV Levels in Serum The purpose of the 9-valent human papillomavirus competitive Luminex immunoassay (HPV9 clia) assay is to measure antibodies to HPV virus-like particles (VLPs), types 6, 11, 16, 18, 31, 33, 45, 52 and 58 before and after vaccination with the HPV 9-valent vaccine. This assay is used by PPD Vaccines and Biologics Lab to evaluate the serological response following vaccination and to measure HPV infection induced antibodies for seroepidemiology studies. Yeast-derived VLPs are coupled to a set of nine distinct fluorescent Luminex microspheres. Antibody titers are determined in a multiplexed, competitive format in which known, type-specific phycoerythrin (PE)-labeled, neutralizing monoclonal antibodies (mabs) compete with the subject s serum antibodies for binding to typespecific, conformationally sensitive, neutralizing epitopes on the VLPs. The fluorescent signals from the bound HPV-specific detection mabs are inversely proportional to the subject s neutralizing antibody titers. Results for the assay are reported as concentration of antibody in arbitrary milli-merck Units per milliliter (mmu/ml). The HPV 6, 11, 16, 18, 31, 33, 45, 52 and 58 clia is performed in a 96-well microtiter plate. A 12-point standard reference serum pool from adult females immunized with a 9- valent vaccine, 4 controls and 16 samples are added to the plate in duplicate. Samples are tested at a 1:4 and a 1:40 dilution. To each well is added the detection antibodies followed by the VLP-microspheres for types 6, 11, 16, 18, 31, 33, 45, 52 and 58. The plates are sealed with foil covers and incubated for 15 to 25 hours. Following incubation, the plates are washed 3 times and the samples are analyzed on a BioPlex (Luminex) instrument. The high, medium, low and negative controls used for this assay were collected from humans that were either HPV sero-negative, had low Ab titers from natural infection or had medium to high Ab titers to the nine HPV types following vaccination. Serostatus Cutoffs The serostatus cutoff is the antibody titer level within the assay s quantifiable range that reliably distinguishes from samples. To determine the serostatus cutoff, the percentage of positive samples for 352 samples in 10 sample panels were assessed at 11 different cutoffs. Prior to testing, sera were classified into panels according to their potential for being a true positive or negative based on V503_001-02_ProtDet APPROVED 16-Aug-2011

173 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: clinical history (age 9-12 years old) and prior HPV-8 clia or HPV-9 clia ver 1.0 seropositivity. The serostatus cutoffs for HPV types 6, 11, 16, 18, 31, 33, 45, 52 and 58 are 30, 16, 20, 24, 10, 8, 8, 8 and 8 mmu/ml, respectively PCR Assays - Detection of HPV in Swabs and Tissue Specimens Cervicovaginal swabs and all Thinsection microtomy biopsy specimens will be tested for detection of HPV types 6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59. In addition to this testing, cervicovaginal swabs and Thinsection microtomy biopsy specimens may tested for other HPV types. HPV types 6, 11, 16, 18, 31, 33, 45, 52, and 58 will be analyzed by type-specific multiplex (L1, E6, E7 gene detection) PCR assay (described in Section ). HPV types other than 6, 11, 16, 18, 31, 33, 45, 52, and 58 will be analyzed by the duplex (E6, E7 gene detection) PCR assay (using the preparation method described in Section ) Multiplex PCR Assays The following procedures will be done for the detection of HPV Types 6, 11, 16, 18, 31, 33, 45, 52, and 58 in frozen swabs and Thinsection microtomy biopsy samples. Specimens are received and then prepared for multiplex PCR using a DNA purification method (Qiagen Technology Kit). Multiplex PCR (based on real-time fluorescent PCR) allows the simultaneous detection of 3 gene products (L1, E6, and E7) for a given HPV type in 1 reaction. The HPV type-specific primer pairs based on the published HPV L1, E6, and E7 sequences, are used to specifically amplify a portion of each gene simultaneously. The specific amplicons are detected in real-time by fluorescently-labeled oligonucleotide probes. The gene-specific oligonucleotide probes are each labeled with a different fluorescent label, and the fluorescent emission is captured during PCR cycling. After analysis of the raw fluorescent data by the real time PCR instrument software, a threshold cycle (Ct), which represents the PCR cycle at which an increase in reporter fluorescence above a baseline signal can first be detected, is determined. Each genespecific assay (i.e., gene-specific dye layer) is considered positive if the Ct is <45 cycles. A gene-specific assay is considered negative if the Ct = "No Ct". A sample is called positive when 2 or 3 genes are positive or when the same single gene scores positive on consecutive tests Preparation and Disposition of Thinsections of Biopsy Tissue The following procedures will be performed at the SPONSOR-designated Central Laboratory. The procedures will be performed by an experienced, qualified histotechnologist according to the Central Laboratory s (Standard Operating Procedure (SOP). The histotechnologist will assure that the microtome and work areas are clean and free of contaminants. All Thinsection microtomy for PCR will be performed at a time when all other routine work has been completed, so that potential contaminations can be minimized. Prior to sectioning each block, a new blade will be installed in the microtome. The blade will only be positioned so that it is at the left margin of the blade surface. Technicians sectioning study blocks will utilize clean gloves V503_001-02_ProtDet APPROVED 16-Aug-2011

174 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: while handling the blocks (new gloves for each block). First, the histotechnologist will face the block by removing two 4-micron sections from the face of the block. These sections will be discarded. Using sterile plastic forceps, the next two 4-micron paraffin sections are collected and floated in a water bath for the preparation of 1 H&E slide (Slide 1, with 2 sections). Nine additional, consecutive sections will then be cut to be used for Thinsection PCR. There will be 9 individual tubes (Tube 1, 2, 3, 4, 5, 6, 7, 8, 9), and one 4-micron section will be placed in each tube using a sterile disposable plastic forceps. The pair of sterile plastic forceps used is then discarded after placing the cut section in each tube. Each tube is then placed inside a plastic sleeve and sealed. Two additional, consecutive 4-micron sections will then be cut and the 2 sections floated in the water bath for preparation of the second H&E slide both sections to be placed on one slide (Slide 2 with 2 sections each). All H&E slides (Slides 1 and 2) will have a histopathologic review by the central laboratory s pathologist. Slides and tubes should be labeled with subject s allocation number. The specimen tubes are collated with the appropriate specimen requisition and prepared for shipping to the SPONSOR-designated Central Laboratory and then in turn, shipped on to MRL. The microtome is cleaned in preparation for the next block and the process above is repeated. The microtome blade is replaced with a new blade and adjusted for each new biopsy block and the same procedure is to be followed. A new pair of clean gloves and a new pair of clean, disposable forceps will be used for each block being sectioned. The blade may be retained for cutting non-pcr blocks. The total number of sections to be cut from each block is 13. A total of 2 slides and 9 tubes: 1. Slide 1 (H&E), with 2 sections each, stained. 2. Tubes 1, 2, 3, 4, 5, 6, 7, 8, 9 (HPV PCR Analysis), 1 section per tube. 3. Slide 2 (H&E), with 2 sections each, stained Total IgG Luminex Immunoassay The purpose of the nine-valent human papillomavirus (HPV) total IgG Luminex immunoassay (HPV9 IgG) assay is to measure antibody (Ab) concentrations to HPV virus-like particles (VLPs) types 6, 11, 16, 18, 31, 33, 45, 52 and 58 before and after vaccination with the HPV 9-valent vaccine. This assay is used by PPD Vaccines and Biologics Lab to evaluate the serological response following vaccination and to measure HPV infection induced antibodies for seroepidemiology studies. Yeast-derived VLPs are coupled to a set of nine distinct fluorescent Luminex microspheres. Antibody concentrations are determined in a multiplexed, direct binding format by measuring the amount of VLP-specific IgG bound to VLP-microspheres. Following incubation with human serum, fluorescent signal from an anti-human IgG detection antibody that binds directly to serum IgG and equally to each IgG subclass V503_001-02_ProtDet APPROVED 16-Aug-2011

175 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: (1-4), is directly measured on the Luminex or BioPlex instrument. The fluorescent signal from the IgG bound fluorescent detection antibody is proportional to the individual s anti-vlp IgG antibody levels. Results for the assay are reported as concentration of antibody in arbitrary milli-merck Units per milliliter (mmu/ml). The HPV 6, 11, 16, 18, 31, 33, 45, 52 and 58 IgG assay is performed in a 96-well microtiter filter plate. A 12-point standard reference serum pool from adult females immunized with a 9-valent vaccine, 4 controls and 16 samples are added to the plate in duplicate. Samples are tested at a 1:100 and a 1:10,000 dilution. To each well is added the VLP-microspheres for types 6, 11, 16, 18, 31, 33, 45, 52 and 58. The plates are sealed with foil covers and incubated for minutes. The contents of the filter plate are washed and incubated with the mouse, anti-human IgG 1-4 monoclonal antibody conjugated to phycoerythrin (PE). The plates are covered with foil and incubated for an additional minutes. Following the second incubation period, the plates are washed 3 times and the samples are analyzed on a BioPlex (Luminex) instrument. The high, medium, low and negative controls used for this assay were collected from humans that were either HPV sero-negative, had low Ab concentrations from natural infection or had medium to high Ab concentrations to the nine HPV types following vaccination. 3.4 SAFETY MEASUREMENTS Clinical and Laboratory Measurements for Safety All subjects will be observed for at least 30 minutes after each study vaccination for any untoward effects, including allergic reactions. This observation period will be documented in the subject s study chart. Each subject will receive a Vaccination Report Card (VRC) at the Day 1, Month 2, and Month 6 study vaccination visits. On the VRC, the subject will be asked to record her oral temperatures in the evening after each study vaccination and daily for 4 days after each study vaccination for the purpose of identifying febrile events. The subject should be instructed to collect temperatures at the same time of day whenever possible. Also, beginning after each study vaccination and for a total of 15 days including the day of vaccination, the subject will be asked to record injection-site and systemic adverse experiences, concomitant medications, and concomitant vaccinations on the VRC. The information on the VRC should be generated only by the subject and must be signed and dated by the subject to confirm the accuracy of the recorded information. The subject will be asked to bring the VRC to the study site at the next scheduled visit (with the exception of the first 150 subjects enrolled in Part A, who will submit the VRC at a Day 16 visit). When the VRC is returned, the VRC should be reviewed for completeness by study site personnel. If clarification is needed, the study site personnel will discuss the VRC with the subject. Original information on the VRC should never be altered by study personnel, although comments can be written in the designated area for study site personnel on the front of the VRC. Only the subject can make corrections to her information on the VRC. Any corrections to the VRC should be made by using an ink pen to add the omitted data V503_001-02_ProtDet APPROVED 16-Aug-2011

176 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: and/or to draw a single line through the error and add the correct information. The subject should initial/date all VRC corrections. All VRC information will be recorded in the ecrfs. The physician investigator/subinvestigator will determine causality of systemic and injection-site adverse experiences recorded on the VRC using the reporting guidelines given in the protocol and will classify each event as a serious adverse experience (SAE) or non-serious adverse experience (NSAE). If an oral temperature indicates a fever (defined as an oral temperature of 100 F or 37.8 C), the adverse experience of must be documented in the ecrfs. At the time of VRC review at the next scheduled visit, subjects will be questioned regarding any new medical conditions that occurred beyond Postdose Day 15. The physician investigator/sub-investigator will determine if the medical condition is to be reported as an SAE using the reporting guidelines. Serious adverse experiences must be reported as described in section Any event of fetal loss must be reported as a serious adverse experience as described in Section Recording Adverse Experiences An adverse experience is defined as any unfavorable and unintended change in the structure, function, or chemistry of the body temporally associated with the use of the SPONSOR s product, whether or not considered related to the use of the product. Any worsening (i.e., any clinically significant adverse change in frequency and/or intensity) of a preexisting condition which is temporally associated with the use of the SPONSOR s product, is also an adverse experience. Changes resulting from normal growth and development which do not vary significantly in frequency or severity from expected levels are not to be considered adverse experiences. Examples of this may include, but are not limited to, teething, typical crying in infants and children, and onset of menses or menopause occurring at a physiologically appropriate time. All adverse experiences will be collected from the time the consent form is signed through 14 days following the first vaccination(s) and from the time of any subsequent vaccination(s) through 14 days thereafter, and such events will be recorded at each examination on the Adverse Experience Case Report Forms/Worksheets Definition of an Overdose for This Protocol In this study, an overdose is defined as a subject receiving >3 doses (0.5 ml each dose) or receiving 0.75 ml of vaccine in any one dose Reporting of Overdose to SPONSOR If an adverse experience(s) is associated with ( vaccine, the adverse experience(s) is reported as a serious adverse experience, even if no other criteria for serious are met. V503_001-02_ProtDet APPROVED 16-Aug-2011

177 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: If a dose of test drug or vaccine meeting the protocol definition of overdose is taken without any associated clinical symptoms or abnormal laboratory results, the overdose is reported as a non-serious Event of Clinical Interest (ECI), using the terminology verse effect. All reports of overdose with and without an adverse experience must be reported within 24 hours to one of the individuals listed on the sponsor contact information page found in the Administrative Binder Reporting of Pregnancy/Breastfeeding Events to the SPONSOR Although not considered an adverse experience, it is the responsibility of the investigators or their designees to report any pregnancy in a subject (spontaneously reported to them or detected by urine or serum pregnancy test per protocol). If a subject becomes pregnant during the study, including subjects who were never randomized and had a positive pregnancy test at Day 1, all related information must be reported in the Electronic Data Capture (EDC) system. For subjects who become pregnant with LMP prior or equal to Day 180 following the final vaccination, the pregnancy will be reported by the Safety Clinical Research Specialist to the NWAES database. Furthermore, all fetal loss pregnancy outcomes (ectopic pregnancy, elective termination, spontaneous abortion, late fetal death) for pregnancies in subjects with LMP prior to or equal to Day 180 following the final vaccination will be reported as a Serious Adverse Experience (Other Medical Event). For randomized subjects who become pregnant after receiving one or two study vaccinations, study visits and study vaccinations will be paused until resolution of the pregnancy (e.g., term, elective termination, spontaneous abortion). Study visits and study vaccinations in pregnant subjects will be handled as described in Table 3-1. All randomized subjects who receive study vaccine, including discontinued subjects who agree to provide further information, will be followed to the completion/termination of the pregnancy. In addition, if the pregnancy continues to term, the outcome (health of the infant) must be reported. If a subject receives study vaccine while breastfeeding during the Day 1 through Month 7 period, all related information must be reported in the Electronic Data Capture (EDC) system and its outcome must be reported. Infant serious adverse experiences (SAEs) for all infants born to subjects who received study vaccine must be reported to the SPONSOR. The reporting of pregnancy, lactation, and infant SAE events involves entering the information directly into the Electronic Data Capture System within 24 hours. Refer to the Data Entry Guidelines (DEGs) for what information must be reported Immediate Reporting of Adverse Experiences to the SPONSOR Serious Adverse Experiences Any serious adverse experience, including death due to any cause, which occurs to any subject from the time the consent is signed through 6 months following the last vaccination, whether or not related to the investigational product, must be reported within V503_001-02_ProtDet APPROVED 16-Aug-2011

178 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: hours to one of the individual(s) listed on the sponsor contact information page found in the Administrative Binder. Additionally, any serious adverse experience brought to the attention of an investigator who is a qualified physician at any time outside of the time period specified in the previous paragraph also must be reported immediately to one of the individuals listed on the sponsor contact information page (found in the administrative binder) if the event is either: 1. A death which resulted in the subject/patient discontinuing the study or 2. A serious adverse experience that is considered by an investigator who is a qualified physician to be possibly, probably, or definitely vaccine related. or 3. A serious adverse experience that is considered by an investigator who is a qualified physician to be possibly, probably, or definitely related to a study procedure. All subjects/patients with serious adverse experiences must be followed up for outcome Evaluating Adverse Experiences Refer to Table 3-4 for instructions in evaluating adverse experiences. V503_001-02_ProtDet APPROVED 16-Aug-2011

179 Product: V Protocol/Amendment No.: Table 3-4 An investigator who is a qualified physician, will evaluate all adverse experiences as to: Maximum Mild awareness of sign or symptom, but easily tolerated (for pediatric studies, awareness of symptom, but easily tolerated) Intensity Moderate discomfort enough to cause interference with usual activity (for pediatric studies, definitely acting like something is wrong) Severe incapacitating with inability to work or do usual activity (for pediatric studies, extremely distressed or unable to do usual activities) Injection site redness or swelling from the day of vaccination through Day 4 post-vacc will be evaluated by maximum size. Seriousness A serious adverse experience is any adverse experience occurring at any dose that: Results in death; or Is life threatening; or places the subject/patient, in the view of the investigator, at immediate risk of death from the experience as it occurred [Note: This does not include an adverse experience that, had it occurred in a more severe form, might have caused death.]; or Results in a persistent or significant disability/incapacity (substantial disruption of one s ability to conduct normal life functions); or Results in or prolongs an existing inpatient hospitalization (hospitalization is defined as an inpatient admission, regardless of length of stay, even if the hospitalization is a precautionary measure for continued observation. (Note: Hospitalization [including hospitalization for an elective procedure] for a preexisting condition which has not worsened does not constitute a serious adverse experience.); or Is a congenital anomaly/birth defect (in offspring of subject/patient taking the product regardless of time to diagnosis);or Is a cancer; or Is an overdose (Whether accidental or intentional.) Any overdose whether or not associated with an adverse experience must be reported within 24 hours to one of the individuals on the Contact Information Page found in the Administrative Binder. Other important medical events that may not result in death, not be life threatening, or not require hospitalization may be considered a serious adverse experience when, based upon appropriate medical judgment, the event may jeopardize the subject/patient and may require medical or surgical intervention to prevent one of the outcomes listed previously (designated above by a ). Duration Action taken Relationship to test vaccine Record the start and stop dates of the adverse experience. If less than 1 day, indicate the appropriate length of time and units Did the adverse experience cause the test vaccine to be discontinued? Did the test vaccine cause the adverse experience? The determination of the likelihood that the test vaccine caused the adverse experience will be provided by an investigator who is a qualified physician. The investigator s signed/dated initials on the source document or worksheet, that supports the causality noted on the AE form, ensures that a medically qualified assessment of causality was done. This initialed document must be retained for the required regulatory time frame. The criteria below are intended as reference guidelines to assist the investigator in assessing the likelihood of a relationship between the test vaccine and the adverse experience based upon the available information. The following components are to be used to assess the relationship between the test vaccine and the AE; the greater the correlation with the components and their respective elements (in number and/or intensity), the more likely the test vaccine caused the adverse experience (AE): Exposure Time Course Likely Cause Is there evidence that the subject/patient was actually exposed to the test vaccine such as: reliable history, acceptable compliance assessment (e.g. diary), seroconversion or identification of vaccine virus in bodily specimen? Did the AE follow in a reasonable temporal sequence from administration of the test vaccine? Is the time of onset of the AE compatible with a vaccine-induced effect? Is the AE not reasonably explained by another etiology such as underlying disease, other drug(s)/vaccine(s), or other host or environmental factors V503_001-02_ProtDet APPROVED 16-Aug-2011 V503, Protocol Issue Date: 16-Aug

180 Product: V Protocol/Amendment No.: Relationship to test vaccine (continued) The following components are to be used to assess the relationship between the test vaccine and the AE: (continued) Dechallenge (not applicable for vaccines) Rechallenge Consistency with Study Vaccine Profile Was the subject/patient reexposed to the test vaccine in this study? If yes, did the AE recur or worsen? If yes, this is a positive rechallenge. If no, this is a negative rechallenge. (Note: This criterion is not applicable if: (1) the initial AE resulted in death or permanent disability, or (2) the study is a single-dose vaccine study.) NOTE: IF A RECHALLENGE IS PLANNED FOR AN ADVERSE EVENT WHICH WAS SERIOUS AND WHICH MAY HAVE BEEN CAUSED BY THE TEST VACCINE, OR IF REEXPOSURE TO THE TEST VACCINE POSES ADDITIONAL POTENTIAL SIGNIFICANT RISK TO THE SUBJECT/PATIENT, THEN THE RECHALLENGE MUST BE APPROVED IN ADVANCE BY THE U.S. CLINICAL MONITOR AND THE INSTITUTIONAL REVIEW BOARD/INDEPENDENT ETHICS COMMITTEE. Is the clinical/pathological presentation of the AE consistent with previous knowledge regarding the test vaccine or vaccine class pharmacology or toxicology? The assessment of relationship will be reported on the case report forms /worksheets by an investigator who is a qualified physician according to his/her best clinical judgment, including consideration of the above elements. Use the following scale of criteria as guidance (not all criteria must be present to be indicative of a vaccine relationship). Definitely related Probably related Possibly related Probably not related Definitely not related There is evidence of exposure to the test vaccine. The temporal sequence of the AE onset relative to administration of the test vaccine is reasonable. The AE is more likely explained by the test vaccine than by another cause. Dechallenge is positive. Rechallenge (if feasible) is positive. The AE shows a pattern consistent with previous knowledge of the test vaccine or test vaccine class. There is evidence of exposure to the test vaccine. The temporal sequence of the AE onset relative to administration of the test vaccine is reasonable. The AE is more likely explained by the test vaccine than by another cause. Dechallenge (if performed) is positive. There is evidence of exposure to the test vaccine. The temporal sequence of the AE onset relative to administration of the test vaccine is reasonable. The AE could have been due to another equally likely cause. Dechallenge (if performed) is positive. There is evidence of exposure to the test vaccine. There is another more likely cause of the AE. Dechallenge (if performed) is negative or ambiguous. Rechallenge (if performed) is negative or ambiguous. The subject/patient did not receive the test vaccine. OR Temporal sequence of the AE onset relative to administration of the test vaccine is not reasonable. OR There is another obvious cause of the AE. V503_001-02_ProtDet APPROVED 16-Aug-2011 V503, Protocol Issue Date: 16-Aug

181 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: SPONSOR Responsibility for Reporting Adverse Experiences All adverse experiences will be reported to regulatory agencies, IRB/IECs, and investigators in accordance with all applicable global laws and regulations. 3.5 DATA ANALYSIS Responsibility for and Timing of Analyses The data collected under this protocol will be analyzed by the Vaccine Clinical Biostatistics Department of the SPONSOR. A comprehensive Statistical Analysis Plan will be provided prior to the interim analysis of Part A. This study will be conducted using in-house blinding procedures. This study will have a data and safety monitoring board (DSMB). An unblinded statistician at the SPONSOR who is otherwise unrelated to this protocol will serve as a non-voting member of the DSMB and will be responsible for providing summaries of safety results to the rest of the DSMB. The unblinded statistician will also be responsible for generating grouped summaries of immunogenicity data for the Week 4 Postdose 2 (Month 3) interim analysis for distribution to a senior management committee at the SPONSOR who is not directly involved with study conduct for the purpose of selecting a dose for the safety, immunogenicity, and efficacy evaluations in Part B. Part A The interim analysis of immunogenicity will be conducted when ~100% of all serology data are available through Week 4 Postdose 2 (Month 3) for subjects in Part A. At that time, the database will be made available only to the unblinded statistician in order to carry out the immunogenicity analysis. All available data related to the immunogenicity summaries will be screened and cleaned in a blinded manner before the database is made available to the unblinded statistician. Group-level summaries of immune responses (GMTs, seroconversion rates) will be provided to a senior management committee at the SPONSOR. No serology or randomization data for individual study participants will be disclosed. Unblinded summaries of all available tolerability data in Part A will be provided to the DSMB for review. No efficacy results will be summarized. The senior management committee with the assistance provided by the DSMB will select the dose using the process outlined in Section All subjects in Part A will be followed through at least Month 7, and subjects who are not in the vaccination groups selected for Part B will be discontinued from the study after the Month 7 visit. Sites will access the Interactive Voice Response System (IVRS) for notification of which Part A subjects will continue past Month 7 and which will complete the study at Month 7. No vaccination group information for those subjects who received the selected 9-valent HPV L1 VLP vaccine or GARDASIL will be released to the study sites. The primary immunogenicity analysis in Part A will be conducted based on the Week 4 Postdose 3 (Month 7) immunogenicity results of the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B and the GARDASIL comparator. The formal V503_001-02_ProtDet APPROVED 16-Aug-2011

182 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: statistical test of the Part A hypothesis based upon screened and cleaned Month 7 data will be provided in the final study report. As indicated previously in this section, an interim analysis based on partially screened/cleaned Month 2 immunogenicity data was performed by the unblinded statistician to guide dose selection for Part B. To protect blinding of the on-going study, only group-level summaries of the immune responses (GMTs, seroconversion rates) were provided to the senior management committee. Details on the blinding/unblinding process and list of Merck personnel with access to group level summaries are described in a study file. Subsequently, group-level summaries of Month 7 immune responses (GMTs, seroconversion rates) were generated by the unblinded statistician to provide confirmation of the dose chosen for continuation into Part B of the study. This followed the same blinding/unblinding procedure as used for the Month 2 interim immunogenicity analysis. Subjects who were enrolled in a 9-valent HPV L1 VLP vaccine formulation with a lower total VLP concentration than the formulation that was selected for use in Part B may be offered a single dose of GARDASIL Although identifying subjects who will be offered a single dose of GARDASIL could potentially unblind the clinical team to their vaccination group assignment, these subjects will be discontinued after the Month 7 visit in Part A and thus will not participate in the planned efficacy component of the study. Thus, the integrity of the blind will be preserved for the subjects in the selected 9-valent HPV L1 VLP vaccine formulation and GARDASIL arms who will proceed to Part B. The Interactive Voice Response System (IVRS) will provide notification of those subjects who may be offered a single dose of GARDASIL to the study. In the unlikely event that the SPONSOR is unable to determine an acceptable 9-valent HPV L1 VLP vaccine formulation for use in Part B, the study may not continue to Part B. Part B If the formulation selected by the senior management committee based on the post-dose 2 immunogenicity data is confirmed to have a favorable tolerability profile by the DSMB, then an additional 13,380 subjects will be randomized in a 1:1 ratio to receive the selected 9-valent HPV L1 VLP vaccine formulation or GARDASIL Subjects from Part A enrolled in the GARDASIL arm and the 9-valent HPV L1 VLP formulation selected for Part B will remain in the trial for the evaluation of efficacy and safety. Because Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. personnel (senior management committee) will view summaries of immunogenicity data from Part A, no immunogenicity data from Part A will be used in the primary immunogenicity analyses in Part B. Part B employs a fixed event design. Although the study will continue for 42 months, the study conclusions regarding vaccine efficacy with respect to the primary endpoint will be based on the analyses conducted when at least 30 cases of the primary efficacy endpoint (HPV 31/33/45/52/58-related high-grade cervical abnormalities (CIN 2/3), V503_001-02_ProtDet APPROVED 16-Aug-2011

183 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer) have been observed. In addition, in order to have a sufficient duration of follow-up of vaccinated subjects, the primary efficacy analysis will not be performed before the median follow-up of PART B subjects is approximately 24 months after initial vaccination. Inference with respect to the primary efficacy hypothesis will be based on this analysis. The counting of efficacy cases will be done by the same unblinded statistician at the SPONSOR who will provide summaries of safety to the DSMB. If the primary efficacy analysis is conducted prior to the end of the study (i.e. before the Month 42 follow-up visit of the last subject enrolled), the remainder of study follow-up still to be accrued will be considered an extension, the purpose of which is to collect additional information in order to assess vaccine efficacy over the entire follow-up period. Although the immunogenicity results in Part B may be available prior to the primary efficacy analysis, the primary immunogenicity analysis will be conducted at the time the database is unblinded for the primary efficacy analysis. The official clinical database for the primary efficacy analysis and primary immunogenicity analysis in part B will be unblinded only after medical/scientific review has been completed, and the database has been declared complete. After the primary efficacy analysis and the primary immunogenicity analysis are performed, data cleaning, data screening, and medical/scientific review of the database will continue unblinded until follow-up is complete. However, investigators, laboratory staff (including SPONSOR s laboratory staff), the members of the HPV Vaccine Program Pathology Panel, and study subjects will remain blinded until the conclusion of the study. All investigators and technicians who are responsible for the ascertainment and confirmation of efficacy endpoints will remain blinded for the duration of the study. If changes are made after the start of the study to the statistical analysis plan presented in this Data Analysis section, the changes made, along with an explanation as to why they occurred, will be listed in the Statistical Analysis Plan and/or the Clinical Study Report of this study, as appropriate Hypotheses The study hypotheses are listed in Section 2.1 of the protocol. The study will be considered a success if the primary efficacy hypothesis and the primary immunogenicity hypothesis in Part B are demonstrated Variables and Time Points of Interest Safety/Tolerability The important variables for safety/tolerability are the incidences of severe injection-site reactions and any vaccine-related serious adverse experiences. V503_001-02_ProtDet APPROVED 16-Aug-2011

184 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Efficacy Endpoint Associated with Primary Efficacy Hypothesis The primary efficacy endpoint is the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer. Subjects with Pap test abnormalities at scheduled visits will be referred for colposcopy based on a mandatory triage strategy. Lesions observed during the colposcopy will be biopsied. Sections of each biopsy will be sent to members of the Pathology Panel for pathological diagnosis. Additional sections will be sent to Merck Research Laboratories (MRL) for PCR analysis. Should a subject be referred for definitive therapy, sections of the tissue excised during the definitive therapy procedure will be sent to the Pathology Panel for diagnosis. Additional sections will be sent to MRL for PCR analysis. To be classified as an endpoint case for the primary analysis, a subject must develop at least one of the following after the completion of the Month 7 visit: 1. High-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma related to HPV 31, 33, 45, 52 or 58. This endpoint is defined to have occurred if on a single cervical biopsy, ECC, LEEP or Conization (cold knife/laser) specimen, there is: (a) a HPV Vaccine Program Pathology Panel consensus diagnosis of CIN (grade 2 or 3), AIS, or cervical cancer; AND (b) detection of at least 1 of HPV types 31, 33, 45, 52 or 58 by Thinsection PCR in an adjacent section from the same tissue block. 2. High-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer related to HPV 31, 33, 45, 52 or 58. This endpoint is defined to have occurred if on a single biopsy or excised tissue, there is: (a) a HPV Vaccine Program Pathology Panel consensus diagnosis of VIN (grade 2 or 3), VaIN (grade 2 or 3), vulvar cancer, or vaginal cancer; AND (b) detection of at least 1 of HPV types 31, 33, 45, 52 or 58 by thin section PCR in an adjacent section from the same tissue block. An additional efficacy endpoint will be evaluated in a sensitivity analysis. In the sensitivity analysis, a case of disease will require a consensus diagnosis of HPV disease by the pathology panel, the appropriate HPV type detected in an adjacent section of the same tissue block AND at least 1 specimen immediately prior to or subsequent to the biopsy or definitive therapy sample that is positive to the same HPV type that is found in the biopsy. V503_001-02_ProtDet APPROVED 16-Aug-2011

185 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Endpoints Associated with Secondary Efficacy Objectives S1: Incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits ±1 month visit windows) 6 months or longer apart. This endpoint is defined to have occurred if a subject who, after completion of the Month 7 visit, is positive for the same HPV type by the HPV 31/33/45/52/58 PCR assay to at least 1 common gene in 2 or more consecutive cervicovaginal/external genital swab, biopsy, or definitive therapy samples obtained at two or more consecutive visits 6 months apart (± 1 month) or longer. Persistent infection is defined based on the protocol visit intervals. Scheduled visits for the collection of LVPP/EEC swabs are at 6-month intervals, so most cases of persistent infection will be based on consecutive specimens obtained 6 months apart; however, due to protocol-allowable deviations for scheduled visit intervals, the minimum length of time between samples for a subject to be counted as a case of persistent infection will be 4 months. If a subject has 2 consecutive samples that are PCR positive to at least one common gene for the same HPV type and at least one of the samples is a biopsy or definitive therapy sample showing pathologic evidence of HPV disease, then the subject will be considered a case of persistent infection without regard to the length of time between the samples. S4: Incidence of HPV 31-, 33-, 45-, 52-, and 58-related cervical, vulvar and vaginal disease. This endpoint is defined as a subject who, after completion of the Month 7 visit, is found to have a biopsy or excised tissue specimen with (a) a HPV Vaccine Program Pathology Panel consensus diagnosis of CIN(any grade), AIS, cervical cancer, genital wart, VIN(any grade), VaIN(any grade), vulvar cancer, or vaginal cancer, AND (b)detection of at least 1 of HPV types 31, 33, 45, 52, or 58 by Thinsection PCR in an adjacent section from the same tissue block. S6: Incidence of Pap test abnormalities. A case is defined as a subject who, after completion of the Day 1 visit, is found to have a Pap test result of atypical squamous cells of undetermined significance (ASC-US) with positive HPV probe or worse (ASC-US [positive for High Risk HPV], LSIL, atypical squamous cells suggestive of a high-grade intraepithelial lesion [ASC-H], HSIL, atypical glandular cells, or adenocarcinoma in situ [AIS]). Endpoints Associated with Exploratory Efficacy Objectives E1: Incidence of HPV 31-, 33-, 45-, 52-, and 58-related documented infection detected in samples from two or more consecutive visits (+/- 1 month visit windows) 12 months or longer. This endpoint is defined to have occurred if a subject who, after completion of the Month 7 visit, is positive for the same HPV type by the HPV 31/33/45/52/58 PCR assay to at least 1 common gene in 2 or more consecutive cervicovaginal/external genital swab, biopsy, or definitive therapy samples obtained for over a period of at least 12 months (within ± 1 month windows) or longer. V503_001-02_ProtDet APPROVED 16-Aug-2011

186 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: E2: Combined incidence of HPV 16- and 18-related persistent infection detected in samples from two or more consecutive visits (+/- 1 month visit windows) 6 month apart or HPV 16- and 18-related cervical, vulvar, and vaginal disease. To be classified as an case for this endpoint, a subject must develop at least one of the following after the completion of the Month 7 visit: 1. Persistent HPV 16 or 18 Infection. The definition of this endpoint is the same as those given for the incidence of HPV 31-, 33-, 45-, 52- and 58-related persistent infection in the secondary efficacy hypothesis (S1), except in relation to HPV types 16 and 18 instead of types 31, 33, 45, 52 and HPV 16- and 18-related cervical, vulvar, and vaginal disease. The definition of this endpoint is the same as those given for the incidence of HPV 31-, 33-, 45-, 52- and 58-related cervical, vulvar, and vaginal disease in the secondary efficacy hypothesis (S4), except in relation to HPV types 16 and 18 instead of types 31, 33, 45, 52 and 58. E3: Incidence of HPV 6- and 11-related cervical, vulvar and vaginal disease. The definition of this endpoint is the same as those given for the incidence of HPV 31-, 33-, 45-, 52- and 58-related cervical, vulvar, and vaginal disease in the secondary efficacy hypothesis (S4), except in relation to HPV types 6 and 11 instead of types 31, 33, 45, 52 and 58. E4: Incidence of CIN, AIS, or cervical cancer caused by any HPV type (vaccine or nonvaccine). This endpoint is defined as a subject who, after completion of the Day 1 visit, has a consensus diagnosis of CIN (any grade), AIS, or cervical cancer on a single cervical biopsy, ECC, LEEP or Conization (cold knife/laser) specimen. E5: Incidence of vulvar or vaginal disease caused by any HPV type (vaccine or nonvaccine). This endpoint is defined as a subject who, after completion of the Day 1 visit, has a consensus diagnosis of genital wart, VIN (any grade), VaIN (any grade), vulvar cancer, or vaginal cancer on a single biopsy or excised tissue. E7: Combined incidence of HPV 35-, 39-, 51-, 56- and 59-related persistent infection detected in samples from two or more consecutive visits (+/- 1 month visit windows) 6 month or longer apart and HPV 35-, 39-, 51-, 56- and 59-related cervical, vulvar, and vaginal disease This endpoint is defined as a subject who, after completion of the Day 1 visit, is found to have a HPV 35-, 39-, 51-, 56-, or 59-related persistent infection or HPV 35-, 39-, 51-, 56-, or 59- related cervical, vulvar and vaginal disease. The definition for persistent infection and disease is the same as those given for the combined incidence of HPV 16 and 18-related persistent infection and disease in the exploratory efficacy hypothesis (E2), except in relation to HPV types 35, 39, 51, 56 and 59 instead of types 16 and 18. V503_001-02_ProtDet APPROVED 16-Aug-2011

187 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: E8: Incidence of HPV 31-, 33-, 45-, 52-, and 58-related Pap test abnormalities. A case is defined as a subject who, after completion of the Month 7 visit, is found to have a Pap test diagnosis of ASC-US [with positive for High Risk HPV ], atypical squamous cells suggestive of a high-grade intraepithelial lesion [ASC-H], HSIL, LSIL, atypical glandular cells, or adenocarcinoma in situ [AIS], and at least one cervicovaginal/external genital swab that is PCR positive for HPV 31, 33, 45, 52, or 58 at the same study visit. E9: Incidence of cervical biopsy and cervical definitive therapy treatments. A subject is defined to have an incident cervical biopsy or definitive therapy procedure if she has a cervical biopsy or definitive therapy after completion of the Day 1 visit. For all efficacy analyses, if a subject has experienced one or more of the components of the respective composite endpoint, she will be classified as a single for the analysis at the time of detection of the first endpoint. She will be classified as a - case only if she has experienced none of the components of the respective composite endpoint. A subject will be counted as a at most once in each efficacy analysis. For analyses in which cases of the composite endpoint are further classified by each individual infection or disease component, a subject will be counted as a at most once in each applicable sub-category but may appear in multiple sub-categories (e.g., persistent HPV 16 infection and HPV 16-related CIN 1) Immunogenicity Part A The primary immunogenicity endpoints in Part A are geometric mean titers (GMTs) to HPV 6, 11, 16, and 18 at Week 4 Postdose 3. Other important endpoints are geometric mean titers (GMTs) to HPV 31, 33, 45, 52 and 58 at Week 4 Postdose 3. Part B The primary immunogenicity endpoints in Part B are geometric mean titers (GMTs) to HPV 6, 11, 16, and 18 at Week 4 Postdose 3. The secondary immunogenicity endpoints are a) the seroconversion percentages to each of HPV 6, 11, 16, and 18 by Week 4 Postdose 3; and b) the percentages of subjects who seroconvert for each HPV type (31, 33, 45, 52 and 58) by Week 4 Postdose 3. Seroconversion is defined as changing serostatus from seronegative to seropositive, by Week 4 Postdose 3. A subject with a clia titer at or above the serostatus cutoff for a given HPV type is considered seropositive for that type. Additional immunogenicity endpoints include the geometric mean titers (GMTs) for HPV 31, 33, 45, 52 and 58 at Week 4 Postdose 3 and the immune responses for HPV 6, 11, 16, 18, 31, 33, 45, 52, and 58 at the persistence time points (Months 12, 24, 36, and 42). The exploratory immunogenicity endpoints are the geometric mean titers (GMTs) to HPV 6 and HPV 11 measured in peripartum maternal blood and cord blood of infants born to women vaccinated at sites participating in cord blood data collection. V503_001-02_ProtDet APPROVED 16-Aug-2011

188 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Analysis Populations Safety All subjects who received at least 1 study vaccination and have follow-up data will be included in the primary analysis of safety Efficacy All efficacy evaluations will include subjects enrolled in Part B of the study combined with subjects from Part A who were randomized to receive the selected 9-valent HPV L1 VLP vaccine or GARDASIL Per Protocol Efficacy (PPE) Populations The approach for the primary efficacy hypothesis will be per-protocol. To be included in the primary efficacy evaluation, subjects must have received all 3 vaccinations with the correct clinical material within 1 year, have 1 or more follow-up visits following Month 7, and have Month 7 PCR swab samples collected within 14 to 72 days postdose 3. Subjects must also be seronegative to the appropriate HPV type(s) at Day 1 and PCR negative to the appropriate HPV type(s) on all cervicovaginal swabs and biopsies from Day 1 through Month 7. For example, for evaluations of efficacy with respect to either HPV type 6 or HPV type 11, subjects must be negative at the aforementioned time points for both HPV 6 and 11. For evaluations of efficacy with respect to any other vaccine HPV type, subjects need only be negative at the aforementioned time points for the HPV type under evaluation. Modified Intent-to-Treat Analyses Three modified intent-to-treat (MITT) populations will be analyzed to support the analyses in the per protocol population, to explore the robustness of the vaccine efficacy and to evaluate the secondary and exploratory efficacy objectives. Subjects who received at least 1 vaccination and have any follow-up visit following the first vaccination will be included in all 3 populations. In all 3 populations, events occurring anytime after Day 1 are eligible to be counted as endpoint cases if they meet the appropriate case definitions. The populations are defined as follows: 1. HPV Type-Specific Naïve Population (HN-TS) - This population will include only subjects who are seronegative and PCR-negative at enrollment to the appropriate HPV types, as defined in the Primary Per Protocol Analysis. 2. Full Analysis Set (FAS) - This population will not be restricted to subjects who are seronegative and PCR-negative at enrollment to the appropriate HPV types, as defined in the Primary Per Protocol Analysis. This analysis in essence includes all subjects randomized, except those who did not receive any vaccination or those without any follow-up. V503_001-02_ProtDet APPROVED 16-Aug-2011

189 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: All HPV Naïve Population (All-HN) This population includes only subjects who (a) are seronegative and PCR-negative at enrollment to all 9 vaccine HPV types; (b) are PCR-negative at enrollment to all other non-vaccine HPV types for which PCR assays will be available; and (c) have a normal Pap test result at enrollment. In all the MITT populations, subjects will be counted in the vaccination group to which they were originally randomized, regardless of the actual clinical material received at each of the three vaccination visits Immunogenicity Part A immunogenicity evaluations will include all subjects enrolled in Part A. Part B immunogenicity evaluations will include those subjects enrolled in Part B only. Per Protocol Immunogenicity (PPI) Populations The primary approach to the analyses of immunogenicity will also be per protocol (immunogenicity). The per-protocol immunogenicity population is defined in the same way as the per protocol efficacy population with the following exceptions: The subjects vaccination visits must each occur within acceptable day ranges relative to Day 1. The subjects must have at least 1 Month 7 serology result within 21 to 49 days Postdose 3. All Type-Specific Naïve Subjects with Serology Populations (ANSS) A supportive immunogenicity analysis will be carried out on the all type-specific naïve subjects with serology population. To be included in this analysis subjects must be seronegative and PCR negative to the relevant HPV type(s) as described above, and must have provided serology data Statistical Methods Efficacy Vaccine Efficacy of the 9-Valent HPV L1 VLP Vaccine Relative to GARDASIL To address the primary efficacy hypothesis, a one-sided test of the null hypothesis that the vaccine efficacy is 25% will be conducted. The alternative hypothesis states that the vaccine efficacy is > 25%. The statistical criterion for success with respect to the primary efficacy hypothesis requires that the lower bound of the confidence interval for vaccine efficacy exclude 25% for the primary efficacy endpoint. The primary efficacy hypothesis will be tested at the α=0.025 (1-sided) level. Vaccine efficacy is defined as: VE = 100%*{1-(r N /r G )} V503_001-02_ProtDet APPROVED 16-Aug-2011

190 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: where r N, the incidence rate among 9-valent HPV L1 VLP vaccine recipients, is defined as r N = C N /τ N. C N = number of primary efficacy cases among 9-valent HPV L1 VLP vaccine recipients and τ N = total person-years of follow-up among 9-valent HPV L1 VLP vaccine recipients. Similarly, r G = C G /τ G is the incidence rate among GARDASIL recipients, where C G is the number of primary efficacy cases among GARDASIL recipients and τ G is the total person-years of follow-up among GARDASIL recipients The null hypothesis that vaccine is not efficacious (i.e., VE 25%) will be tested by constructing a two-sided exact confidence interval for VE. Under the assumption that r N and r G are the means of independent Poisson processes, and given that there are a total of n = C N + C G primary efficacy cases observed on all subjects, the number of primary efficacy cases C N among 9-valent HPV L1 VLP vaccine recipients is distributed as Binomial(n,p), where the binomial probability p is defined as p= τ N r N /(τ N r N +τ G r G ). The probability p is a person-years-adjusted estimate of the probability that a given primary efficacy case is a 9-valent HPV L1 VLP vaccine recipient. The lower bound of the 100*(1 α)% exact confidence interval for the probability p is obtained by searching for the proportion p L such that the probability of observing C N or more primary efficacy cases out of n total primary efficacy cases is α/2. Similarly, the upper bound of the 100*(1 α)% exact confidence interval for the probability p is obtained by searching for the proportion p U such that the probability of observing C N or fewer primary efficacy cases out of n total primary efficacy cases is α/2 [26]. The upper and lower bounds of the 100*(1 α)% confidence interval for the vaccine efficacy can be computed from the upper and lower bounds of the confidence interval for p. To address the secondary efficacy hypothesis concerning the combined incidence of HPV 31/33/45/52/58-related persistent infection detected in samples from two or more consecutive visits (+/- 1 month visit windows) 6 months or longer apart and exploratory efficacy hypothesis concerning the composite endpoint of HPV 31/33/45/52/58-related persistent infection detected in samples from two or more consecutive visits (+/- 1 month visit windows) 12 months or longer apart, a one-sided test of the null hypothesis that the vaccine efficacy is 0% will be conducted using the same methodology as for the primary efficacy hypothesis. The statistical criterion for success with respect to the secondary efficacy hypothesis requires that the lower bound of the 95% confidence interval for vaccine efficacy exclude 0% for the secondary efficacy endpoint. Vaccine Efficacy of the 9-Valent HPV L1 VLP Vaccine Relative to Historical Placebo Given the expected high efficacy of both the 9-valent HPV L1 VLP vaccine and GARDASIL against the original HPV types (6, 11, 16, and 18), a small number of cases may be observed in both groups, making it difficult to assess vaccine efficacy of the 9-valent HPV L1 VLP vaccine relative to GARDASIL Therefore, additional analysis to estimate vaccine efficacy of the 9-valent HPV L1 VLP vaccine relative to placebo is needed. V503_001-02_ProtDet APPROVED 16-Aug-2011

191 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Because there is no placebo group in the current study, a historic cohort of subjects in Protocols 007 and 012 from the Phase II/III placebo-controlled efficacy studies for GARDASIL will be used. Specifically, the historic cohort will consist of subjects who meet the same eligibility criteria for inclusion in the Per-Protocol Efficacy (PPE) population for the relevant HPV type as in the current study. To provide an estimate of the vaccine efficacy of 9-valent HPV L1 VLP vaccine relative to placebo, an indirect method will be used [27]: VE N/P = 100%* (1-RR N/P ) = 100% {1-(r N /r G * r G /r P )} where r N /r G = 1-VE[current study], is the observed relative risk of 9-valent HPV L1 VLP vaccine to GARDASIL G /r P = 1-VE[historic cohort] is the observed relative risk of GARDASIL to placebo from the historic cohort. A 95% confidence interval for VE N/P will be estimated from the 95% confidence interval for the natural log of the relative risk RR N/P. For the confidence interval calculation, the following estimate of the variance of RR N/P will be used: Var(Ln(RR N/P ))= Var(Ln(r N /r G )) + Var(Ln(r G /r P )) This approach will be used to address the objectives evaluating the impact of the 9-valent HPV L1 VLP vaccine on the incidence of a) Pap test abnormalities(s6); b) HPV 16/18- related persistent infection or disease(e2); c) HPV 6/11-related cervical, vulvar, and vaginal disease(e3); d) all CIN, AIS, and cervical cancer due to any HPV type(e4); e) all vulvar and vaginal disease due to any HPV type(e5); and f) cervical biopsy and cervical definitive therapy procedures(e9). No hypothesis testing will be conducted. Non-inferiority Assessment of the 9-Valent HPV L1 VLP Vaccine and That of GARDASIL for HPV Types 16/18, Both Relative to An Historical Placebo It is expected that the active control, GARDASIL and the 9-valent HPV L1 VLP vaccine will be both highly efficacious against the original HPV types. For example, based upon the high efficacy observed in the GARDASIL program, given a total of 10,000 subjects (5,000 per group) with a median follow-up of 30 months post randomization, only 3 cases of type 16/18 related persistent infection or disease are expected in each group, with a non-trivial probability that no case is observed in the GARDASIL group. If no case is observed in the GARDASIL group at the end of study, it will be difficult to draw any conclusion in terms of the relative incidence (9- valent/gardasil since the denominator will be zero. Moreover, even if a non-zero number of cases are observed in the GARDASIL group, and a slightly higher number of cases is observed in the 9-valent HPV L1 VLP vaccine group, the relative incidence may appear "large" but not be clinically significant. For this reason, and because of the expected small number of cases, a traditional case-driven study where a relatively large number of cases are needed to test a non-inferiority hypothesis involving a relative incidence is not appropriate; even with a very large number of subjects; the study may never reach its target within a reasonable time frame. V503_001-02_ProtDet APPROVED 16-Aug-2011

192 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Given the importance of protection against HPV 16 and 18 persistent infection and disease (alone these types are responsible for approximately 70% of identified cases of cervical cancer), and to ensure that GARDASIL cancer coverage for HPV 16/18 is maintained in the 9-valent HPV L1 VLP vaccine, we propose a statistical non-inferiority efficacy analysis in which the observed difference in the incidence rate for HPV 16/18 cases are normalized to a standard placebo rate. A margin of 15 percentage points will be used to define the non inferiority of the 9-valent vaccine versus GARDASIL efficacy against HPV 16- and/or 18-related persistent infection and disease compared to historical placebo, for which the incidence rate is conservatively assumed to be 4 per 100 person-years as observed from GARDASIL program. The 9-valent vaccine will be concluded to be non-inferior to GARDASIL if the lower bound of the two-sided 95% exact CI for the difference in vaccine efficacy compared to historical placebo (9-valent vaccine GARDASIL ) remains above -15 percentage points. The justification for this non-inferiority margin is based on broadening, from a public health perspective, the overall protection against cervical cancer as follows: HPV 16 and 18 are responsible for 70% of cervical cancer cases and GARDASIL against HPV 16/18 persistent infection and disease versus placebo is ~99% Thus GARDASIL cancers. has the potential to prevent 69% (0.99 X 70%) of cervical Assuming V503 has at least 90-95% efficacy preventing HPV 31, 33, 45, 52, 58 persistent infection and knowing that together these types cause another 17% of cervical cancer worldwide, V503 would then provide 15-16% potential cervical cancer coverage in addition to coverage the vaccine provides against HPV 16/18. Applying a non-inferiority margin of -15% and assuming observed GARDASIL efficacy in the trial is 99%, the lower bound of the 95% confidence interval for the observed V503 vaccine efficacy for type 16/18 would need to be 84% or greater and the minimum projected cervical cancer coverage of the vaccine for these types would be 59% (versus 69% for GARDASIL Thus accepting a non-inferiority margin of -15% allows that the minimum overall public health benefit for cervical cancer coverage, 74-75%, provided by V503, (59% from HPV 16/18 and 15-16% from HPV 31, 33, 45, 52, 58) would still exceed that currently provided by GARDASIL Immunogenicity Part A The hypotheses of noninferiority of GMTs for HPV types 6, 11, 16 and 18 will be based on one-sided tests of non-inferiority comparing GMTs for each component. Four ANOVA models (one per HPV type) with a response of log individual titers and a fixed V503_001-02_ProtDet APPROVED 16-Aug-2011

193 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: effect for vaccination group will be used. The hypotheses to be tested are: H 0 : GMT N /GMT G 1/2 versus H 1 : GMT N /GMT G > 1/2, where GMT G represents the GMTs in subjects receiving GARDASIL and GMT N represents the GMTs in subjects receiving the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B. Part A is considered a pilot study, and the statistical criterion for noninferiority in these tests corresponds to the lower bound of the confidence interval for the fold-difference in GMTs between the 2 groups, (9-valent HPV L1 VLP vaccine group/gardasil group), excluding a decrease of 2-fold or more for each component. Part B The hypotheses of noninferiority of GMTs for HPV types 6, 11, 16 and 18 will be based on one-sided tests of non-inferiority comparing GMTs for each component. Four ANOVA models (one per HPV type) with a response of log individual titers and a fixed effect for vaccination group will be used. The hypotheses to be tested are: H 0 : GMT N /GMT G 0.67 versus H 1 : GMT N /GMT G > 0.67, where GMT G represents the GMTs in subjects receiving GARDASIL and GMT N represents the GMTs in subjects receiving the 9-valent HPV L1 VLP vaccine. Each hypothesis will be tested at the α=0.025 level (1-sided). Because the confirmatory immunogenicity evaluation of the selected 9-valent HPV L1 VLP vaccine formulation will be conducted in Part B, the statistical criterion for non-inferiority in these tests corresponds to the lower bound of the 95% confidence interval for the fold-difference in GMTs between the 2 groups, (9-valent HPV L1 VLP vaccine group/gardasil group), excluding a decrease of 1.5-fold or more for each component. The secondary hypothesis of noninferiority of seroconversion percentages for each of HPV types 6, 11, 16, and 18 will be addressed by 4 one-sided tests of noninferiority (one corresponding to each HPV type) conducted at the α=0.025 level (1-sided). For each HPV type, the hypotheses to be tested are H 0 : p 1 -p H a : p 1 -p 2 > where p 1 is the proportion of subjects who seroconvert at Week 4 Postdose 3 in recipients 9-valent HPV L1 VLP vaccine and p 2 is the proportion of subjects who seroconvert at Week 4 Postdose 3 in GARDASIL The tests above will be conducted based on methods developed by Miettinen and Nurminen [28] for testing the equivalence of 2 proportions. The secondary hypothesis of acceptability of anti-hpv seroconversion rates (for HPV types 31, 33, 45, 52 and 58) will be tested by one-sided tests of the proportions of subjects seroconverting (1 test per HPV type). These tests will be conducted based on the exact 95% confidence interval for a binomial proportion. The hypotheses to be tested are: H 0 : P N 0.9 versus H 1 : P N > 0.9 (for each of the HPV Types 31, 33, 45, 52, and 58), where P N represents the true response rate of subjects receiving a 3-dose regimen of a V503_001-02_ProtDet APPROVED 16-Aug-2011

194 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: formulation of 9-valent HPV L1 VLP vaccine. Rejection of the null hypothesis in these tests corresponds to the lower bound of the 95% confidence interval for the percentage of subjects seroconverting in the 9-valent HPV L1 VLP vaccine group being greater than 90% for the given HPV type. The primary time point at which immune response will be evaluated is Week 4 Postdose 3. Anti-HPV responses for each of the 9 vaccine HPV types will be summarized by vaccination groups in terms of geometric mean titers (GMTs) with 95% confidence intervals at each time point when serum samples are collected. Anti-HPV responses during the period of vaccination and persistence from Month 7 onwards will be investigated using longitudinal plots of the GMTs. Reverse cumulative distribution (RCD) plots and a summary of seropositivity percentages, including 95% confidence intervals, will be presented for each of the 9 vaccine HPV types at all time points at which serum samples are collected. The geometric mean titers (GMTs) to HPV 6 and HPV 11 in peripartum maternal blood and cord blood of infants born to women vaccinated at sites participating in cord blood data collection will be summarized descriptively Safety Safety and tolerability will be assessed by statistical and clinical review of all safety data collected throughout the study. All subjects who are vaccinated and who have safety follow-up data will be included in the safety analyses and summaries. All safety analyses and summaries will be provided separately for: (1) all subjects in Part A; and (2) all subjects in Part B combined with subjects from Part A who received the selected 9-valent HPV L1 VLP vaccine formulation or GARDASIL To provide an overall assessment, safety measures such as the incidence of 1. any adverse experiences; 2. any injection-site experiences; 3. any systemic adverse experiences; and 4. any vaccine-related adverse experiences that occurred throughout the study will be summarized in both groups. Adverse experiences will be summarized as frequencies and percentages by vaccination group by vaccination visit and across all vaccination visits. To assess the risks of adverse experiences temporally associated with vaccination, a multi-tiered approach will be used for the analysis of safety parameters. Tier-1 adverse experiences include (1) injection-site adverse experiences prompted for on the VRC, such as redness, swelling, and pain/tenderness/soreness occurring Day 1 through Day 5 following any vaccination, and (2) elevated temperature ( F [ 37.8ºC]), from Day 1 to Day 5 following any vaccination. For Tier-1 adverse experiences, the risk difference between each 9-valent HPV L1 VLP vaccination group V503_001-02_ProtDet APPROVED 16-Aug-2011

195 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: and GARDASIL in Part A, the corresponding two-sided 95% CI on the risk difference, and the p-value for the test of significance of the risk difference will be provided. For the analyses combining data from Part A and Part B, the risk difference between the selected 9-valent HPV L1 VLP vaccination group and the GARDASIL group, 95% CI on the risk difference, and the p-value for the test of significance will also be provided. The corresponding p-values will be based on the asymptotic normal approximation for testing two independent binomial proportions at the two-sided 0.05 level. All risk differences and 95% CIs will be calculated using the methods proposed by Miettinen and Nurminen [28]. The Tier-2 adverse experience summaries include (1) specific systemic adverse experiences within 14 days following any vaccination occurring in 1% of subjects in any vaccination group, (2) injection-site adverse experiences not prompted for on the VRC occurring Day 1 to Day 5 following any vaccination in 1% of subjects in any vaccination group, (3) serious adverse experiences occurring within 14 days following any vaccination, (4) serious vaccine-related adverse experiences observed at any time during the study, and (5) severe injection-site adverse experiences Day 1 through Day 5 following any vaccination visit. Risk differences and 95% confidence intervals between each 9-valent HPV L1 VLP vaccination group compared to the GARDASIL group for all Part A subjects will be estimated for all Tier-2 adverse experiences using the methodology proposed by Miettinen and Nurminen [28]. Tier-3 adverse experiences will include summaries (counts and proportions) by vaccination group for any other adverse experiences, including all injection-site adverse experiences occurring from Day 1 to Day 5 following each vaccination visit and all systemic adverse experiences occurring within 14 days of each vaccination visit Multiplicity Considerations Part A Success in Part A will be declared if a formulation of 9-valent HPV VLP vaccine is selected for use in Part B. For the Month 7 hypothesis test of noninferiority of GMTs of the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B compared with GARDASIL, multiplicity with respect to the multiple hypotheses for HPV types 6, 11, 16 and 18 is addressed by requiring success for all 4 HPV types. The hypothesis for each of HPV types 6, 11, 16, and 18 will be tested at the nominal 1-sided α= level (with minor adjustment for the interim summary). As described in section 3.5.1, there will be an interim summary of immunogenicity in the study when ~100% of all serology data are available through Week 4 Postdose 2 for subjects in Part A. There will be no formal hypothesis testing done at the time of the interim analysis. Because it is recognized that informal reviews of unblinded data compromise the integrity of the final analysis of the data in Part A, a nominal 1-sided alpha level of will be used to test the hypothesis at Month 7 for each of HPV V503_001-02_ProtDet APPROVED 16-Aug-2011

196 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: types 6, 11, 16, and 18 as if an extreme alpha-spending rule had been used for the interim summary, such as the Haybittle-Peto rule [29]. This would theoretically apply a nominal significance level of to any interim summary (even though no hypothesis testing will be conducted). The use of the slightly reduced nominal alpha level at the final analysis in Part A for hypothesis testing will protect the overall study type 1 error at not more than 5% 2-sided. Part B Efficacy Because no efficacy data will be summarized for the interim analysis in Part A, no multiplicity adjustment will be made to account for the subjects from Part A that will be included in the efficacy evaluation for Part B. Immunogenicity The primary immunogenicity analysis in Part B will be conducted independently from the analysis in Part A. Thus, no multiplicity adjustment is required. Multiplicity with respect to the multiple hypotheses for HPV types 6, 11, 16 and 18 is addressed by requiring success for all 4 HPV types thereby controlling the overall alpha level. Thus, no multiplicity adjustment is needed for the overall immunogenicity hypotheses. Success in this study will be declared if the primary efficacy hypothesis and the primary immunogenicity hypothesis in Part B are achieved. Thus, no multiplicity adjustment is needed to control the overall type I error rate for the study. The supplementary analyses of efficacy to be carried out when full follow-up is complete will not require a multiplicity adjustment since no inference will be drawn at that time Sample Size and Power Calculations Efficacy The sample size for efficacy in Part B is determined by the primary efficacy endpoint of the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer. This study is powered based on a fixed event design, with a target of 30 cases. In planning the study, the phenomenon of cross-protection was considered. It is possible that the HPV types contained in GARDASIL may afford some protection against infection and disease related to other HPV types, and in particular HPV types 31, 33, 45, 52 and 58. This may reduce the incidence of the primary endpoint in subjects receiving GARDASIL relative to completely untreated subjects. Based on existing cross protection data from the GARDASIL program with respect to HPV infection and related disease, it has been assumed that this reduction may be up to 30%. V503_001-02_ProtDet APPROVED 16-Aug-2011

197 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: If assuming the vaccine efficacy for the 9-valent HPV L1 VLP versus placebo is 88%, taking the 30% cross protection into account, the true vaccine efficacy for the 9-valent HPV L1 VLP versus GARDASIL will be 83%. With the lower bound of the confidence interval for the true vaccine efficacy must be >25% under the alternative hypothesis, 30 primary efficacy cases are needed to provide at least 90% power to declare the vaccine efficacious (α=0.025, one-sided). To ensure that at least 30 cases are accrued based on a median follow-up of 30 months post-vaccination 1, approximately 14,000 subjects are needed for the efficacy evaluation (13,380 subjects enrolled for Part B combined with the ~620 subjects in the GARDASIL arm and selected 9-valent dose formulation arm from Part A). The assumptions are: exclusion rate due to seropositivity at Day 1 and/or PCR positive between Day 1 and Month 7 to HPV types 31, 33, 45, 52, and 58 is not greater than 23% attrition rate between Day 1 and Month 7 is not greater than 10% annual attrition rate post-month 7 is not greater than 5% the annual incidence rates for the primary efficacy endpoint is 0.35 per 100 personyears median follow-up time is 30 months post randomization Immunogenicity Part A With the enrollment of 310 subjects per group (1,240 subjects enrolled total) to obtain around 209 to 230 evaluable subjects per arm and assuming a nominal α= level (1- sided), this study has an overall power of >99%. The calculations are based on the following assumptions: 1. 15%, 15%, 18% and 10% of the subjects are initially seropositive or PCR positive to HPV Types 6, 11, 16, and 18, respectively. 2. The expected attrition rate through the Month 7 time point is 10%. 3. The expected percentage of subjects excluded due to vaccinations or serology samples out of range is no more than 8% 4. The standard deviations of the natural logarithm of the Month 7 titers are no more than 1.2 for each HPV type when expressed as mmu/ml, the unitage used in previous GARDASIL studies. 5. HPV Types 6, 11, 16, and 18 responses are identical in the GARDASIL group and at least one 9-valent HPV L1 VLP vaccine formulation. V503_001-02_ProtDet APPROVED 16-Aug-2011

198 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: The expected responses rates and noninferiority margins are as shown in Table 3-5 for each antigen. For GARDASIL, these are based on data available from previous studies. Table 3-5 Power for Part A Immunogenicity Hypotheses of Protocol V Antigen Parameter Expected rates/sd Non-inferiority Margin N evaluable Power (1) Primary Hypothesis - Non-Inferiority of 9-valent HPV L1 VLP Vaccine to GARDASIL HPV 6 GMT σ = fold decrease 218 >0.99 HPV 11 GMT σ = fold decrease 218 >0.99 HPV 16 GMT σ = fold decrease 209 >0.99 HPV 18 GMT σ = fold decrease 230 >0.99 Overall Power for GMT hypothesis >0.99 Part B The overall sample size in the study is determined by the efficacy analyses. With approximately 13,380 subjects in Part B, this study has over 99% power for the primary immunogenicity hypothesis of noninferiority with respect to geometric mean titers for HPV types 6, 11, 16, and 18. The calculations are based on the same assumptions as in the Part A immunogenicity analysis. The expected standard deviations and power for the primary immunogenicity hypotheses are as shown in Table 3-6 for each antigen. The study will still have an overall power of > 99% even when GMT N /GMT G =0.9 or 0.8. Table 3-6 Power for Part B Primary Immunogenicity Hypothesis of Protocol V Antigen Parameter Expected rates/sd Non-inferiority Margin N evaluable Power (1) Primary Hypothesis - Non-Inferiority of 9-valent HPV L1 VLP Vaccine to GARDASIL HPV 6 GMT σ = fold decrease 4708 >0.99 HPV 11 GMT σ = fold decrease 4708 >0.99 HPV 16 GMT σ = fold decrease 4542 >0.99 HPV 18 GMT σ = fold decrease 4985 >0.99 Overall Power for the primary immunogenicity hypotheses >0.99 V503_001-02_ProtDet APPROVED 16-Aug-2011

199 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: This study also has >99% power for the secondary immunogenicity hypothesis of noninferiority with respect to seroconversion percentages for HPV types 6, 11, 16, and 18. In addition, this study has >99% power for the secondary immunogenicity hypothesis of acceptability with respect to seroconversion percentages for HPV types 31, 33, 45, 52, and 58. For the secondary seroconversion hypothesis, it is assumed that for each of HPV 31, 33, 45, 52 and 58, no more than 23%, 22%, 16%, 22%, and 20%, respectively, are initially seropositive or PCR positive for any given type. All statistical tests will be conducted at the α=0.025 level (1-sided). The expected response rates and power for the secondary immunogenicity hypotheses are shown in Table 3-7. Table 3-7 Power for Part B Secondary Immunogenicity Hypothesis of Protocol V Antigen Parameter Expected rates/sd Non-inferiority Margin N evaluable Power (1) Secondary Immunogenicity Hypothesis - Non-Inferiority of 9-valent HPV L1 VLP Vaccine to GARDASIL HPV 6 Titer Seropositive Cutoff 98% 5% 4708 >0.99 HPV 11 Titer Seropositive Cutoff 98% 5% 4708 >0.99 HPV 16 Titer Seropositive Cutoff 98% 5% 4542 >0.99 HPV 18 Titer Seropositive Cutoff 98% 5% 4985 >0.99 Overall Power for the seroconversion hypothesis >0.99 (2) Secondary Hypothesis - Acceptability of 9-valent HPV L1 VLP Vaccine HPV 31 Titer Seropositive Cutoff 98% Lower bound>90% 4456 >0.99 HPV 33 Titer Seropositive Cutoff 98% Lower bound>90% 4513 >0.99 HPV 45 Titer Seropositive Cutoff 98% Lower bound>90% 4861 >0.99 HPV 52 Titer Seropositive Cutoff 98% Lower bound>90% 4513 >0.99 HPV 58 Titer Seropositive Cutoff 98% Lower bound>90% 4629 >0.99 Overall Power for the secondary immunogenicity hypothesis >0.99 Measured by Competitive Luminex immunoassay (clia). Seropositivity cutoff values for HPV 6, 11, 16, 18, 31, 33, 45, 52, 58 will be determined. Safety The probability of observing at least 1 SAE in this study depends on the number of subjects enrolled and the incidence rate of SAEs in the general population. If the incidence rate of an SAE is 1 of every 1240 (0.08%) subjects who received any formulation of 9-valent HPV L1 VLP vaccine in Part A, then there is a 53% chance of observing at least 1 such SAE among the 930 subjects who receive any of the 3 formulations of 9-valent HPV L1 VLP vaccine in Part A. If the incidence rate of an SAE is 1 of every 2880 (0.03%) subjects who received a formulation of 9-valent HPV L1 VLP vaccine, then there is a 28% chance of observing at least 1 such SAE among the 930 subjects who receive any of the 3 formulations of 9-valent HPV L1 VLP vaccine in Part A. If no SAEs are observed in the 930 subjects who received a 9-valent HPV L1 VLP V503_001-02_ProtDet APPROVED 16-Aug-2011

200 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: vaccine formulation in Part A, then this study will provide 95% confidence that the true incidence rate for SAEs is <0.4%. In Part B, there is a 99.6% chance of observing at least 1 SAE among the 7000 subjects who will receive the selected 9-valent HPV L1 VLP vaccine formulation if the incidence rate of an SAE is 1 of every 1280 subjects (0.08%). If the incidence rate is 1 of every 2880 recipients (0.03%), then there is a 87.8% chance of observing at least 1 SAE among the subjects who will receive the selected 9-valent HPV L1 VLP vaccine formulation. If no SAEs are observed in 7000 subjects, this study will provide 95% confidence that the true incidence rate for SAEs is <0.05% Interim Analyses Part A A preliminary summary of tolerability and immunogenicity data will be produced, by unblinded Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. personnel, when ~100% of subjects have Week 4 Postdose 2 serology data available. The interim immunogenicity analysis in Part A will be conducted in a subset of subjects who 1) are seronegative at Day 1 to the relevant HPV type, without regard to PCR status at Day 1; 2) receive the first 2 doses of 9-valent HPV L1 VLP vaccine or GARDASIL day ranges; and 3) have post-vaccination serum samples collected in acceptable day ranges. This will present summary statistics of immune responses by vaccination group (GMTs, seroconversion rates, confidence intervals, RCDF graphs). No serology or randomization data for individual study participants will be disclosed. No formal hypothesis testing will be done at the time of the interim analysis. It is important to note that the summary will focus on Postdose 2 data, while the study hypotheses concern Week 4 Postdose 3 (Month 7) responses in Part A subjects who receive the 9-valent HPV L1 VLP formulation that is selected for use in Part B or GARDASIL. Part B No interim analyses are planned for Part B of this study. The primary efficacy analysis will be conducted after at least 30 primary efficacy cases have been observed. Conclusions regarding the success of the study will be drawn from the results of both the primary efficacy analysis and the primary immunogenicity analysis in Part B. Estimates of vaccine efficacy and immunogenicity will be updated after Month 42 follow-up is complete in all subjects, if the primary efficacy analysis is conducted at an earlier time Definition of Compliance Measure Compliance is defined in this study as receipt of all scheduled study vaccinations. To summarize compliance, the numbers of subjects who receive each vaccination in Part A and Part B will be tabulated by vaccination group. Subjects who do not complete the full vaccination regimen will be excluded from the per-protocol primary analyses of efficacy and immunogenicity in Part B, but will be included in the supportive analyses. V503_001-02_ProtDet APPROVED 16-Aug-2011

201 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Another important aspect of compliance is the degree of compliance with follow-up examinations during follow-up visits after the vaccination series has been completed. The numbers of subjects who complete each follow-up visit will be tabulated by vaccination group. The numbers of subjects with biopsies or excision procedures performed outside of the study and the numbers of these subjects for whom the tissue samples were not available to Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., will be tabulated by vaccination group. Differences in these numbers will be assessed observationally and the potential impact on the efficacy analyses noted. 3.6 LABELING, PACKAGING, STORAGE, DISPENSING, AND RETURN OF CLINICAL SUPPLIES Subject and Replacements Information Clinical supplies will be packaged to support enrollment of approximately 14,620 subjects. Study personnel will have access to an Interactive Voice Response System (IVRS) to allocate patients, to assign study vaccine to patients, to notify sites of those subjects from Part A who will continue beyond the Month 7 visit, and to manage the distribution of clinical supplies. Clinical supplies will be packaged according to a component schedule generated by the SPONSOR. Each person accessing the IVRS system must be assigned an individual unique PIN. They must use only their assigned PIN to access the system and they must not share their assigned PIN with anyone Product Descriptions Investigational materials will be provided by the SPONSOR as summarized in Table 3-8. V503_001-02_ProtDet APPROVED 16-Aug-2011

202 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Table 3-8 Product Descriptions Product Name and Potency (HPV Types 6/11/16/18/31/33/45/52/58) Dosage Form Comments (if applicable) Quadrivalent HPV VLP Vaccine 20/40/40/20/0/0/0/0/0 mcg/ 0.5 ml dose Sterile solution for IM injection. Store at 2-8 C. DO NOT FREEZE. Protect from light. 9-valent HPV VLP Vaccine 20/40/40/20/20/20/20/20/20 mcg/ 0.5 ml dose 9-valent HPV VLP Vaccine 30/40/60/40/20/20/20/20/20 mcg/ 0.5 ml dose 9-valent HPV VLP Vaccine 30/40/80/55/30/30/30/30/30 mcg/ 0.5 ml dose Sterile solution for IM injection Sterile solution for IM injection Sterile solution for IM injection Store at 2-8 C. DO NOT FREEZE. Protect from light. Store at 2-8 C. DO NOT FREEZE. Protect from light. Store at 2-8 C. DO NOT FREEZE. Protect from light. GARDASIL = QUADRIVALENT HPV VLP VACCINE 9-VALENT HPV L1 VLP VACCINE = 9-VALENT HPV VLP VACCINE Primary Packaging and Labeling Information Quadrivalent / 9-valent HPV VLP Vaccine will be labeled as HPV VLP Vaccine and will be packaged in 3 ml glass vials containing approximately 0.75 ml of study vaccine for withdrawal of a single 0.5 ml dose. The appearance of all 9-valent HPV L1 VLP vaccine and quadrivalent HPV VLP vaccine vials will be indistinguishable. The tear-off portion of the clinical label will contain at least the packaging identification number and product/protocol. The tear-off portion must be affixed to the Vaccine Inventory and Label LV CRF. V503_001-02_ProtDet APPROVED 16-Aug-2011

203 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: Container label text may include the following: Packaging Control # / Lot Trace ID # Component ID # Dose Volume (0.5 ml Dose for IM injection) Re-evaluation date (If applicable) Product name AN: (If applicable) Dosing Instructions (Administer Per Protocol V ) Storage Conditions Country Regulatory requirements SPONSOR address (If applicable) Translation key (If applicable) Secondary Packaging and Labeling Information- Supplies Will Not be Kitted Clinical Supplies-Vaccine Disclosure IVRS should be used in order to unblind patients and to unmask drug identity. The SPONSOR will not provide disclosure envelopes with the clinical supplies. Vaccine group identification information is to be unmasked ONLY if necessary for the welfare of the patient. Every effort should be made not to unblind the patient unless necessary. Prior to unblinding, the investigator will attempt to contact the clinical monitor. Any unblinding that occurs at the site must be documented Storage Requirements All vaccine supplies will be shipped to the sites as a refrigerated solution to be stored at 2 C to 8 C (35.6 F to 46.4 F). Upon receipt at the investigational site, the vaccine should be removed from the outer secondary shipping box and placed immediately into the refrigerator. The TEMPTALE must be deactivated upon receipt of the shipment. Directions for inactivation are specified in the Instructions to Site, which are enclosed with each shipment. The TEMPTALE will indicate whether the shipment has remained within the specified temperatures. Return the TEMPTALE according to instructions accompanying the shipment. Notify the SPONSOR (Clinical Research Associate [CRA]) and IVRS technical support immediately if the TEMPTALE is in alarm. Store and hold product until instructed otherwise. The clinical supplies storage area at the site must be monitored by the site staff for temperature consistency with the acceptable storage temperature range specified in this protocol or in the product label attached to the protocol. Documentation of temperature monitoring should be maintained. Supplies should be stored in the original nested box with the lid closed to minimize exposure to light. If the refrigerator in which the study vaccine is stored deviates from the 2 C to 8 C (35.6 F to 46.4 F) range, study vaccinations should be suspended and the SPONSOR (CRA) and IVRS technical support should be contacted immediately. Vaccine must NOT be frozen. It is strongly recommended that a non-frost free laboratory grade refrigerator is used to store the study vaccine. This type of refrigerator is less likely to have wide temperature fluctuations, so it will be more likely to stay within the 2 C to 8 C (35.6 F to 46.4 F) temperature range. A daily refrigerator temperature log must be maintained at the site. V503_001-02_ProtDet APPROVED 16-Aug-2011

204 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: The refrigerator must be equipped with an appropriately calibrated min/max thermometer and/or circular chart temperature recorder. The temperature log will be reviewed by the CRA throughout the study. An appropriate back up system (i.e. alarm, generator) and study site personnel telephone numbers should be in place in the event of a refrigerator failure Standard Policies / Return of Clinical Supplies For studies using Controlled Substances, all Federal, State, Province, Country, etc., regulations must be adhered to in regard to the shipping, storage, handling, and dispensing of controlled substances. Additionally, the investigator should have the appropriate controlled drug license(s) as mandated by Federal, State, Province, Country, etc. laws in which the study is being conducted. Investigational clinical supplies must be received by a designated person at the study site, handled and stored safely and properly, and kept in a secured location to which only the investigator and designated assistants have access. Clinical supplies are to be administered only in accordance with the protocol. The investigator is responsible for keeping accurate records of the clinical supplies received from the SPONSOR, the amount administered to the subjects/patients, and the amount remaining at the conclusion of the study. The Clinical Monitor should be contacted with any questions concerning investigational products where special or protective handling is indicated. At the end of the study, all unused clinical supplies must be returned as indicated on the Contact Information page(s). U.S. sites should follow instructions for the Clinical Supplies Return Form (V464) and contact your SPONSOR representative for review of shipment and form before shipping. Sites outside of the United States should check with local country Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. personnel for appropriate documentation that needs to be completed for vaccine accountability. Used vials may be discarded per site s handling of biohazardous waste after documentation of vaccine administration/accountability guidelines have been met. All of the unused sealed vaccine that remain at the conclusion of the study should be retained at the site until the SPONSOR representative is able to account for all of the vials originally shipped to the sites. The unused vials should then be returned to the SPONSOR by the SPONSOR representative at the completion of the study Transport of the Vaccine In the event the vaccine needs to be transported, either between study sub-sites that are under the direction of the same principal investigator or because the location of the study site changes, the SPONSOR must be notified. If vaccine needs to be transported, a vaccine transport standard operating procedure that is acceptable to the study site and the SPONSOR will be followed and a copy of the procedure will be placed in the Administrative Binder. V503_001-02_ProtDet APPROVED 16-Aug-2011

205 V503, Protocol Issue Date: 16-Aug Product: V Protocol/Amendment No.: DATA MANAGEMENT Information regarding Data Management procedures for this protocol will be provided by the SPONSOR. 3.8 BIOLOGICAL SPECIMENS The laboratory that is analyzing any clinical samples should be blinded to the subject s vaccination group. It is the responsibility of the primary investigator to ensure that all staff personnel who will be handling, packaging, and/or shipping clinical specimens act in conformance with International Air Transport Association (IATA) regulations relating to the handling and shipping of hazardous goods. All specimens will be labeled with preprinted computer-generated labels provided by the SPONSOR and/or by the SPONSOR s Central Laboratory. Labels can only be affixed to dry surfaces. However, the labels can be used on polypropylene or polyethylene and will survive freezing and thawing. Information regarding biological specimens and sample labeling for this protocol will be provided by the SPONSOR/Central Laboratory. V503_001-02_ProtDet APPROVED 16-Aug-2011

206 V503, Protocol Issue Date: 16-Aug Product: V503 1 Protocol/Amendment No.: ADMINISTRATIVE AND REGULATORY DETAILS 4.1 CONFIDENTIALITY Confidentiality of Data For Studies Conducted Under the U.S. IND Particular attention is drawn to the regulations promulgated by the Food and Drug Administration under the Freedom of Information Act providing, in part, that information furnished to clinical investigators and Institutional Review Boards will be kept confidential by the Food and Drug Administration only if maintained in confidence by the clinical investigator and Institutional Review Board. For All Studies By signing this protocol, the investigator affirms to the SPONSOR that information furnished to the investigator by the SPONSOR will be maintained in confidence and such information will be divulged to the Institutional Review Board, Ethics Review Committee, or similar or expert committee; affiliated institution; and employees only under an appropriate understanding of confidentiality with such board or committee, affiliated institution and employees. Data generated by this study will be considered confidential by the investigator, except to the extent that it is included in a publication as provided in the Publications section of this protocol Confidentiality of Subject/Patient Records For All Studies By signing this protocol, the investigator agrees that the SPONSOR (or SPONSOR representative), Institutional Review Board/Independent Ethics Committee (IRB/IEC), or Regulatory Agency representatives may consult and/or copy study documents in order to verify worksheet/case report form data. By signing the consent form, the subject/patient agrees to this process. If study documents will be photocopied during the process of verifying worksheet/case report form information, the subject/patient will be identified by unique code only; full names/initials will be masked prior to transmission to the SPONSOR. For Studies Conducted Under the U.S. IND By signing this protocol, the investigator agrees to treat all patient data used and disclosed in connection with this study in accordance with all applicable privacy laws, rules and regulations, including all applicable provisions of the Health Insurance Portability and Accountability Act and its implementing regulations, as amended from time to time. ( PAA Confidentiality of Investigator Information For All Studies By signing this protocol, the investigator recognizes that certain personal identifying information with respect to the investigator, and all subinvestigators and study site V503_001-02_ProtA&R APPROVED 16-Aug-2011

207 V503, Protocol Issue Date: 16-Aug Product: V503 2 Protocol/Amendment No.: personnel, may be used and disclosed for study management purposes, as part of a regulatory submissions, and as required by law. This information may include: name, address, telephone number, and address; hospital or clinic address and telephone number; curriculum vitae or other summary of qualifications and credentials; and other professional documentation. Consistent with the purposes described above, this information may be transmitted to the SPONSOR, and subsidiaries, affiliates and agents of the SPONSOR, in your country and other countries, including countries that do not have laws protecting such information. Additionally, the investigator s name and business contact information may be included when reporting certain serious adverse events to regulatory agencies or to other investigators. By signing this protocol, the investigator expressly consents to these uses and disclosures. For Multicenter Studies In order to facilitate contact between investigators, the SPONSOR may share an investigator s name and contact information with other participating investigators upon request. 4.2 COMPLIANCE WITH LAW, AUDIT, AND DEBARMENT By signing this protocol, the investigator agrees to conduct the study in an efficient and diligent manner and in conformance with this protocol; generally accepted standards of Good Clinical Practice; and all applicable federal, state, and local laws, rules and regulations relating to the conduct of the clinical study. The Code of Conduct, a collection of goals and considerations that govern the ethical and scientific conduct of clinical investigations sponsored by Merck, is attached. The investigator also agrees to allow monitoring, audits, Institutional Review Board/Independent Ethics Committee review, and regulatory agency inspection of trialrelated documents and procedures and provide for direct access to all study-related source data and documents. The investigator agrees not to seek reimbursement from subjects/patients, their insurance providers, or from government programs for procedures included as part of the study reimbursed to the investigator by the SPONSOR. The Investigator shall prepare and maintain complete and accurate study documentation in compliance with Good Clinical Practice standards and applicable federal, state, and local laws, rules and regulations; and, for each subject/patient participating in the study, provide all data, and upon completion or termination of the clinical study submit any V503_001-02_ProtA&R APPROVED 16-Aug-2011

208 V503, Protocol Issue Date: 16-Aug Product: V503 3 Protocol/Amendment No.: other reports to the SPONSOR as required by this protocol or as otherwise required pursuant to any agreement with the SPONSOR. Study documentation will be promptly and fully disclosed to the SPONSOR by the investigator upon request and also shall be made available at the investigator s site upon request for inspection, copying, review, and audit at reasonable times by representatives of the SPONSOR or any regulatory agencies. The investigator agrees to promptly take any reasonable steps that are requested by the SPONSOR as a result of an audit to cure deficiencies in the study documentation and worksheets/case report forms. International Conference of Harmonization Good Clinical Practice guidelines (Section 4.3.3) recommend that the investigator inform the subject s primary physician about the subject s participation in the trial if the subject has a primary physician and if the subject agrees to the primary physician being informed. According to European legislation, a SPONSOR must designate a principal or coordinating investigator (CI) to review the report (summarizing the study results) and confirm that to the best of his/her knowledge the report accurately describes conduct and results of the study. The SPONSOR may consider one or more factors in the selection of the individual to serve as the CI (e.g., thorough understanding of clinical trial methods, appropriate enrollment of subject/patient cohort, timely achievement of study milestones, availability of the CI during the anticipated review process). The investigator will promptly inform the SPONSOR of any regulatory agency inspection conducted for this study. Persons debarred from conducting or working on clinical studies by any court or regulatory agency will not be allowed to conduct or work on this SPONSOR s studies. The investigator will immediately disclose in writing to the SPONSOR if any person who is involved in conducting the study is debarred, or if any proceeding for debarment is pending or, to the best of the investigator s knowledge, threatened. In the event the SPONSOR prematurely terminates a particular trial site, the SPONSOR will promptly notify that site s IRB/IEC. 4.3 COMPLIANCE WITH FINANCIAL DISCLOSURE REQUIREMENTS By signing this protocol, the investigator agrees to provide to the SPONSOR accurate financial information to allow the SPONSOR to submit complete and accurate certification and disclosure statements as required by U.S. Food and Drug Administration regulations (21 CFR Part 54). The investigator further agrees to provide this information on a Financial Disclosure/Certification Form that is provided by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. This requirement also extends to subinvestigators. The investigator also consents to the transmission of this information to Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. in the United States for these purposes. This may involve the transmission of information to countries that do not have laws protecting personal data. V503_001-02_ProtA&R APPROVED 16-Aug-2011

209 V503, Protocol Issue Date: 16-Aug Product: V503 4 Protocol/Amendment No.: QUALITY CONTROL AND QUALITY ASSURANCE By signing this protocol, the SPONSOR agrees to be responsible for implementing and maintaining quality control and quality assurance systems with written SOPs to ensure that trials are conducted and data are generated, documented, and reported in compliance with the protocol, accepted standards of Good Clinical Practice, and all applicable federal, state, and local laws, rules and regulations relating to the conduct of the clinical study. 4.5 COMPLIANCE WITH INFORMATION PROGRAM ON CLINICAL TRIALS FOR SERIOUS OR LIFE THREATENING CONDITIONS Under the terms of The Food and Drug Administration Modernization Act (FDAMA), the SPONSOR of the study is solely responsible for determining whether the study is subject to the requirements for submission to the Clinical Trials Data Bank, Merck, as SPONSOR of this study, will review this protocol and submit the information necessary to fulfill this requirement. Merck entries are not limited to FDAMA mandated trials. Merck s voluntary listings, beyond those mandated by FDAMA, will be in the same format as for treatments for serious or life-threatening illnesses. Information posted will allow patients to identify potentially appropriate trials for their disease conditions and pursue participation by calling a central contact number for further information on appropriate study locations and site contact information. By signing this protocol, the investigator acknowledges that the statutory obligation under FDAMA is that of the SPONSOR and agrees not to submit any information about this study to the Clinical Trials Data Bank. 4.6 PUBLICATIONS As this study is part of a multicenter trial, publications derived from this study should include input from the investigator(s) and SPONSOR personnel. Such input should be reflected in publication authorship, and whenever possible, preliminary agreement regarding the strategy for order of authors names should be established before conducting the study. Subsequent to the multicenter publication, or 24 months after completion of the study, whichever comes first, an investigator and/or his/her colleagues may publish the results for their study site independently. However, the SPONSOR does not recommend separate publication of individual study site results due to scientific concerns. The SPONSOR must have the opportunity to review all proposed abstracts, manuscripts, or presentations regarding this study 60 days prior to submission for publication/presentation. Any information identified by the SPONSOR as confidential must be deleted prior to submission. SPONSOR review can be expedited to meet publication guidelines. V503_001-02_ProtA&R APPROVED 16-Aug-2011

210 V503, Protocol Issue Date: 16-Aug Product: V503 1 Protocol/Amendment No.: LIST OF REFERENCES 1. Munoz, N, Bosch, XF, Catellsague, X, Diaz, M, De Sanjose, S, Hammouda, D, Shah, KV, and Meijer, C. Against which human papillomavirus types shall we vaccinate and screen? The international perspective. Int J Cancer 2004;111: Pagliusi SR, Aguado MT. Efficacy and other milestones for human papillomavirus vaccine introduction. Vaccine 2004;23: Jansen KU, Shaw AR. Human papillomavirus vaccines and prevention of cervical cancer. Annu Rev Med 2004;55: Chuang T-Y, Perry HO, Kurland LT, Ilstrup DM. Condyloma acuminatum in Rochester, Minn, :II anaplasias and unfavorable outcomes. Arch Dermatol 1984;120: Stone KM. Human papillomavirus infection and genital warts: update on epidemiology and treatment. Clin Infec Dis 1995;20(Suppl 1):S91-S Bosch FX, Munoz N. The viral etiology of cervical cancer. Virus Research 2002;89(2): Christensen ND. Emerging human papillomavirus vaccines. Expert Opin Emerg Drugs 2005;10(1): Glikman D, Baroody FM. Recurrent respiratory papillomatosis with lung involvement. N Engl J Med 2005;352(24):e Szeps M, Dahlgren L, Aaltonen LM, Ohd J, Kanter-Lewenshon L, Dahlstrand H, et al. Human papillomavirus, viral load and proliferation rate in recurrent respiratory papillomatosis in response to alpha interferon treatment. J Gen Virol 2005;86: Lin T-S, Lee H, Chen R-A, Ho M-L, Lin C-Y, Chen Y-H, et al. An association of DNMT3b protein expression with P16INK4a promoter hypermethylation in nonsmoking female lung cancer with human papillomavirus infection. Cancer Lett 2005;226: Lyronis ID, Baritaki S, Bizakis I, Tsardi M, Spandidos DA. Evaluation of the prevalence of human papillomavirus and Epstein-Barr virus in esophageal squamous cell carcinomas. Int J Biol Markers 2005;20(1): Yang H, Yang K, Khafagi A, Tang Y, Carey TE, Opipari AW, et al. Sensitive detection of human papillomavirus in cervical, head/neck, and schistosomiasisassociated bladder malignancies. Proceedings of the National Academy of Sciences 2005;102(21): Howley PM. Papillomavirinae: The viruses and their replication. In: Fields BN, Knipe DM, Howley PM, et al., eds. Virology. 3rd ed. Philadelphia: Lippincott-Raven Publishers, 1996: V503_001-02_ProtApp APPROVED 16-Aug-2011

211 V503, Protocol Issue Date: 16-Aug Product: V503 2 Protocol/Amendment No.: Auvinen E, Crusius K, Steuer B, Alonso A. Human papillomavirus type 16 E5 protein (review). Int J Oncol 1997;11: Stern PL. Immune control of human papillomavirus (HPV) associated anogenital disease and potential for vaccination. J Clin Virol 2005;Suppl 32:S72-S Oriel JD. Natural history of genital warts. Brit J Vener Dis 1971;47: Xi LF, Koutsky LA. Epidemiology of genital human papillomavirus infections. Bull Inst Pasteur 1997;95: Koutsky L. Epidemiology of genital human papillomavirus infection. Am J Med 1997;102(5A): Walboomers JMM, Jacobs MV, Manos MM, Bosch FX, Kummer A, Shah KV, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189: Nobbenhuis MAE, Walboomers JMM, Helmerhorst TJM, Rozendaal L, Remmink AJ, Risse EKJ, et al. Relation of human papillomavirus status to cervical lesions and consequences for cervical-cancer screening: a prospective study. Lancet 1999;354: Wallin K-L, Wiklund F, Ångström T, Bergman F, Stendahl U, Wadell G, et al. Typespecific persistence of human papillomavirus DNA before the development of invasive cervical cancer. N Engl J Med 1999;341(22): Melbye M, Frisch M. The role of human papillomaviruses in anogenital cancers. Semin Cancer Biol 1998;8: Consensus Development Panel. Consensus statement: National Institutes of Health Consensus Development Conference statement on cervical cancer. Gynecol Oncol 1997;66: Anttila A, Pukkala E, Söderman B, Kallio M, Nieminen P, Hakama M. Effect of organised screening on cervical cancer incidence and mortality in Finland, : recent increase in cervical cancer incidence. Int J Cancer 1999;83: Östör AG. Natural history of cervical intraepithelial neoplasia: a critical review. Int J Gynecol Pathol 1993;12(2): Chan ISF, Bohidar NR. Exact power and sample size for vaccine efficacy studies. Commun Statist-Theory Meth 1998;27(6): Hasselblad V, Kong, DF. Statistical methods for comparison to placebo in activecontrol trials. Drug Inf J 2001; 35 (2): Miettinen O, Nurminen M. Comparative analysis of two rates. Stat Med 1985;4: Haybittle JL. Repeated assessment of results in clinical trials of cancer treatment. Brit J Rad 1971;44: V503_001-02_ProtApp APPROVED 16-Aug-2011

212 V503 Statistical Analysis Plan Statistical Analysis Plan A Randomized, International, Double-Blinded (With In-House Blinding), Controlled With GARDASIL, Dose-Ranging, Tolerability, Immunogenicity, and Efficacy Study of a Multivalent Human Papillomavirus (HPV) L1 Virus-Like Particle (VLP) Vaccine Administered to 16- to 26-Year-Old Women (Study 001) Amendment Prepared by Oliver M. Bautista, Ph.D. CONFIDENTIAL, PROPRIETARY PROPERTY OF MERCK & CO., INC. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

213 V503 Statistical Analysis Plan 2 SUMMARY OF CHANGES PRIMARY REASON FOR AMENDMENT: August 2011 Amendment This SAP was amended according to a major change in the protocol. In particular, the success criterion relating to the secondary efficacy objective on the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart, which is changed from demonstrating vaccine efficacy greater than 0% to demonstrating vaccine efficacy greater than 25%. January 2009 Amendments This SAP was amended according to the major changes in the protocol. The primary reason for the amendment was to revise the primary objective and hypothesis for measuring efficacy of the 9-valent HPV L1 VLP vaccine against infection and disease caused by HPV types 31, 33, 45, 52, and 58. The new primary objective and hypothesis is designed to study the efficacy of the 9-valent HPV L1 VLP vaccine against the composite endpoint of HPV 31-, 33-, 45-, 52-, and 58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer, compared with GARDASIL in 16- to 26-year-old adolescent and young adult women. Based on the new primary objective and hypothesis, the following significant changes have been made: The original primary objective and associated hypothesis designed to demonstrate that administration of 9-valent HPV L1 VLP vaccine reduces the incidence of the composite endpoint of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection for a duration of 6 months or longer compared with GARDASIL in the study population has been moved to a secondary objective. The original secondary objective and associated hypothesis designed to demonstrate that administration of 9-valent HPV L1 VLP vaccine reduces the incidence of the composite endpoint of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection for a duration of 12 month or longer compared with GARDASIL in the study population has been moved to an exploratory objective. Based on the anticipated incidence of the new primary endpoint, the study size increased from 3,720 (enrolled in Part A & Part B combined) to 14,620 total subjects. Of the 14,620 total subjects enrolled, 620 subjects from Part A who did not receive the selected V503 formulation or GARDASIL will complete the study at Month 7 and not be included in the efficacy phase of the study. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

214 V503 Statistical Analysis Plan 3 OTHER CHANGES INCLUDED IN THE AMENDMENT: SAP Section Revision Executive Summary and Section Section August 2011 Amendments Deleted the word ELISA in objective (E10). Clarified that the formal statistical test of the Part A hypothesis based upon screened and cleaned Month 7 data will be provided in the final study report, as opposed to when all Month 7 data are available. Section 2.3 Provided the serostatus cutoffs used in the definition of HPV positive by serology. Section In Table 5: - Clarified the timepoints at which summaries will be provided. - Deleted the indications that Part B summaries will be provided for the S0P0 population. The S0P0 is the primary analysis population for assessing Part A immune response Postdose 2 whereas the PPI is the primary analysis population for assessing Part A and Part B immune response Postdose 3. Section Section Deleted the indications that summaries will be provided for the S1 (without regard for PCR-status) population. Summaries are provided for populations defined based on serology and PCR status. - Defined the S0P0 analysis population in the body of this section, as opposed to the previous version of the SAP where the S0P0 population definition was provided only in a footnote in Table 5. - Deleted The subjects Month 7 PCR samples do not need to be collected within 14 to 72 days post dose 3 from the definition of the PPI population. Deleted references to Part B summaries being provided for S0P0 and for S1 (without regard for PCRstatus) populations. Only Day 1 GMTs, without RCDs, of anti-hpv types 31, 33, 45, 52, and 58 will be provided for the populations defined by Day 1 serology and PCR status. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

215 V503 Statistical Analysis Plan 4 SAP Section Section 7.3 Revision Clarified that the formal statistical test of the Part A hypothesis based upon screened and cleaned Month 7 data will be conducted at the time the database is unblinded for the primary efficacy analysis. January 2009 Amendments Section Study Design Section Primary Objectives Section Secondary Objectives - Revised the method by which sites would receive notification of which subjects from Part A received the selected 9-valent HPV L1 VLP vaccine formulation or GARDASIL. The process employs the use of IVRS rather than sending lists to the study sites. - Revised timing and number of cases required for the primary efficacy analysis. -Revised number of serum samples to be evaluated for immunogenicity hypothesis in Part B. In addition to the main changes to the objectives/hypotheses outlined at the beginning, the statistical criteria for success have been added to the primary efficacy and immunogenicity hypothesis. - The original secondary objectives, (1), (2), and (3) have moved to exploratory objectives (1), (2) and (3). - In the first (1) secondary objective: Clarified the description of 6 month persistent infection; Removed disease from the endpoint since that is captured in the new primary efficacy endpoint; Revised the criteria for success such that the lower bound of the 95% CI for vaccine efficacy will be greater than 0. - In the second (2) secondary objective: Clarified the definition of acceptability. - In the third (3) secondary objective: Clarified the definition of noninferiority. - Added a new objective (as (4)) to evaluate the impact of 9-valent HPV L1 VLP vaccine on HPV 31-, 33-, 45-, 52-, and 58- related cervical, vulvar and vaginal disease. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

216 V503 Statistical Analysis Plan 5 SAP Section Section Exploratory Objectives Revision - In the first (1) exploratory objective: Clarified the description of 12 month persistent infection; Removed disease from the endpoint since that is captured in the new primary efficacy endpoint; Revised the primary approach for this analysis to be per protocol; Clarified the criteria for success. - In the second (2) exploratory objective: Added a hypothesis. Section Efficacy Endpoint - In the seventh (7) exploratory objective: Clarified the description of 6 month persistent infection. - Added a new exploratory objective to assess humoral immune responses to HPV type 6, 11, 16, 18, 31, 33, 45, 52, and 58 using a type specific anti-hpv L1 VLP IgG ELISA upon availability of a validated assay. - Added the endpoints associated with the new primary efficacy endpoint of disease and the definition of a case. - Moved the endpoints associated with the 6 month Persistent Infection to the "Endpoints Associated with Secondary Efficacy Hypothesis" section. - Added the definition of the endpoint of HPV 31-, 33-, 45-, 52-, 58-related cervical, vulvar, and vaginal disease. - Revised the definition of 12 month Persistent Infection to reflect that cases will be counted after Month 7 instead of Day 1. Table 4 Planned Analyses -Modified and re-arranged endpoints based on new efficacy objectives /hypothesis. -Removed 3 rd sensitivity analysis of "including subjects with PCR (+) at last visit as cases" because this only related to persistent infection which is no longer in the primary efficacy endpoint. -Revised to use PPE as the primary approach for evaluating persistent infection 12 month. Section 5.4 -Added Summary of the conduct of the Part A dose selection. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

217 V503 Statistical Analysis Plan 6 SAP Section Section 6.1 DSMB Section 7.1 Sample Size and Power Revision -Added two additional milestones for review of tolerability data by the DSMB since the size of the study and the duration of enrollment has increased. -Recalculated Part B efficacy sample size based on the new primary efficacy endpoint. -Revised power calculation for Part B immunogenicity hypothesis based on new sample size. -Revised probability of observing SAE in Part B given new sample size. Section 7.4 Statistical Methods -Revised the lower bound of vaccine efficacy to 25% for the primary efficacy hypothesis. -Clarified the approach for determining vaccine efficacy of the 9-valent HPV L1 VLP vaccine relative to historical placebo. -Added details for the non-inferiority assessment of 9- valent HPV L1 VLP vaccine and GARDASIL for HPV types 16/18, both relative to a historical placebo. -Removed case imputation in sensitivity analysis with respect to persistent infection to match planned analyses in Table 4. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

218 V503 Statistical Analysis Plan 7 Statistical Analysis Plan A Randomized, International, Double-Blinded (With In-House Blinding), Controlled With GARDASIL, Dose-Ranging, Tolerability, Immunogenicity, and Efficacy Study of a Multivalent Human Papillomavirus (HPV) L1 Virus-Like Particle (VLP) Vaccine Administered to 16- to 26-Year-Old Women (Study 001) TABLE OF CONTENTS PAGE Summary of Changes 2 1. Introduction Objective of the Statistical Analysis Plan Description of the Study and Objectives/Hypotheses Study Design Objectives/Hypotheses Study Participant Characteristics Subject Accounting Subject Characteristics Baseline HPV Status Prior and Concomitant Medications New Medical Conditions Sexual Activity and Contraceptive Use Efficacy Analyses Efficacy/Immunogenicity Endpoints Efficacy Endpoints Immunogenicity Endpoints Study Participant Populations Efficacy Populations Immunogenicity Populations Other Efficacy and Immunogenicity Populations Approaches to Efficacy Analysis Efficacy Analyses Immunogenicity Analyses Analysis of Correlates of Protection Definition of Compliance Measure Safety Analyses Study Participant Population Extent of Exposure to Drug Adverse Experiences Clinical Safety Interim Analyses Timing and Procedure 55 BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

219 V503 Statistical Analysis Plan 8 Statistical Analysis Plan (Study 001) TABLE OF CONTENTS (CONT.) PAGE 5.2 Dose Selection Criteria Analysis Details Summary of the conduct of the Part A dose selection Data/Study Reviewing Committees Data Safety and Monitoring Board (DSMB) Merck Senior Management Committee (sidmc) Scientific Advisory Committee (SAC) Statistical Technical Issues Planned Statistical Power and Sample Size Sample Size Considerations for Primary and Secondary Efficacy Hypotheses Sample Size Considerations for Primary and Secondary Immunogenicity Hypotheses Sample Size Considerations for Safety Analyses Method of Assigning Study Participants to Treatment Groups Blinding/Unblinding Statistical Methods Vaccine Efficacy of the 9-Valent HPV L1 VLP Vaccine Relative to GARDASIL Evaluation of Efficacy of the 9-Valent HPV L1 VLP Vaccine Relative to Historic Placebo Non-inferiority Assessment of the 9-Valent HPV L1 VLP Vaccine and That of GARDASIL for HPV Types 16/18, Both Relative to a Historical Placebo Counting of Efficacy Endpoints Computation of Follow-Up Time Potential Bias Handling of Missing Data Case Imputation Analyses of Immune Responses Multiplicity Subgroup Analyses Ground Rules and Data Handling Conventions Baseline Definitions or Conventions Time Points, Day Ranges, and Phasing of Study Periods 73 BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

220 V503 Statistical Analysis Plan 9 Statistical Analysis Plan (Study 001) TABLE OF CONTENTS (CONT.) PAGE 8.3 Description of Data Handling Procedures Prior to Unblinding Description of Protocol Violations Datasets/Programs/Variables References 78 BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

221 V503 Statistical Analysis Plan 10 EXECUTIVE SUMMARY Background HPV types 6, 11, 16, and 18 are in GARDASIL, a prophylactic HPV L1 VLP vaccine that has been marketed in several countries for such indications as the prevention of cervical, vulvar, and vaginal cancers and related precancers, and external genital lesions. The addition of HPV 31, 33, 45, 52, and 58 to the HPV types in GARDASIL has the potential to increase overall cervical cancer coverage from approximately 70% to 87%. V503 is Merck s first vaccine that includes VLPs for all 9 of the aforementioned types. Protocol V is the first study of this vaccine. The study is designed to first evaluate a dose, among those studied, based on tolerability and immunogenicity, of the prophylactic 9-valent HPV (Types 6, 11, 16, 18, 31, 33, 45, 52, 58) L1 virus-like particle (VLP) vaccine and to subsequently study this dose for additional tolerability and immunogenicity and for efficacy. Because a dose-ranging pilot will be conducted before expanded enrollment can proceed for the primary immunogenicity and efficacy objectives, this study will be conducted in two parts. The first part of this study, Part A, will evaluate the tolerability and immunogenicity of three 9-valent HPV L1 VLP vaccine formulations versus GARDASIL. The second part of this study, Part B, is a larger scale evaluation of the tolerability, immunogenicity, and efficacy of the 9-valent HPV L1 VLP vaccine formulation selected from Part A versus GARDASIL. Design Part A (dose-ranging and tolerability): Approximately 1,240 healthy 16- to 26-yearold women will be randomized in equal numbers to one of three 9-valent HPV L1 VLP vaccine formulations or the comparator GARDASIL in a three dose regimen (Day 1, Month 2, and Month 6). A 9-valent HPV L1 VLP vaccine dose, among those doses tested, will be selected for Part B after approximately 100% of the Part A post-dose 2 immunogenicity and adverse experience data are available in the clinical database. The dose for Part B will be chosen by a senior management committee (sidmc) at Merck & Co., Inc. (Merck & Co., Inc. hereafter referred to as the SPONSOR), whose members are not directly involved with study conduct based on a review of the immunogenicity summaries by vaccination group and manufacturing considerations. This committee will relay the dose selection to the Data Safety Monitoring Board (DSMB), who will ratify the decision based on unblinded tolerability data. For the summary of the conduct of the Part A dose selection, see Section 5.4 for detail. Part B (tolerability, immunogenicity, and efficacy): Approximately 13,380 additional healthy 16- to 26-year-old women will be randomized in equal numbers to the 9-valent HPV L1 VLP vaccine dose selected from Part A or the comparator GARDASIL. The formal assessments of the tolerability, immunogenicity, and efficacy of the 9-valent vaccine will be the tests of the Part B primary hypotheses. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

222 V503 Statistical Analysis Plan 11 The formal assessment of the effectiveness of the HPV 6, 11, 16 and 18 components of the vaccine will be the test of the primary immunogenicity hypothesis. This hypothesis compares the selected 9-valent formulation from Part A to the GARDASIL control. The hypothesis test will be conducted in individual HPV type-specific per protocol immunogenicity populations using only subjects accrued in Part B of the study. The formal assessment of the efficacy of the 9-valent HPV L1 VLP vaccine with respect to the HPV 31, 33, 45, 52 and 58 components will be the test of reduction of the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer, compared with GARDASIL. The subjects enrolled in Part A who received the selected 9-valent HPV L1 VLP vaccine dose or the comparator GARDASIL will be combined with all subjects enrolled in Part B for the assessment of efficacy with respect to HPV 31, 33, 45, 52 and 58. The primary efficacy analysis is case driven and will be conducted after at least 30 primary efficacy cases have been observed. In addition, in order to have a sufficient duration of follow-up of vaccinated subjects, the primary efficacy analysis will not be performed before the median follow-up of PART B subjects is approximately 24 months after initial vaccination. The primary evaluation of tolerability will also combine the subjects enrolled in Part A who received the selected 9-valent HPV L1 VLP vaccine dose or the comparator GARDASIL with subjects enrolled in Part B. The schedule of vaccinations and assessments for efficacy, immunogenicity, and safety are as shown in Table 1. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

223 V503 Statistical Analysis Plan 12 Procedures/Assessments Table 1 Study Measurements to Evaluate Tolerability, Immunogenicity, and Efficacy Vaccination Phase Month Vaccination Safety Assessments Elevated temperatures and injection site adverse experiences within 4 days after the vaccination visit Systemic adverse experiences within 14 days after the vaccination visit. Serious adverse experiences within 14 days following any vaccination visit Follow-up Phase Month Day (D1-5) (D1-15) (D1-15) (D1-5) (D1-15) (D1-15) (D1-5) (D1-15) (D1-15) Serious vaccine-related adverse experiences Procedures Relating to Assessment of Immunogenicity Serum collection for anti-hpv antigens Procedures Relating to Assessment of Efficacy External genital wart inspection LVPP swabs for HPV PCR EEC swab for HPV PCR Pap test (ThinPrep ) (D1-5) denotes the follow-up period from Day 1 to Day 5 following the respective vaccination visit (within 4 days following each vaccination visit). (D1-15) denotes the follow-up period from Day 1 to Day 15 following the respective vaccination visit (within 14 days following each vaccination visit). EEC = Endo/Ectocervical; HPV = Human Papillomavirus; LVPP = Labial, vulvar, perineal, and perianal. BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

224 V503 Statistical Analysis Plan 13 Objectives / Hypotheses Part A (dose-ranging and tolerability): Primary Part A Objectives/Hypotheses The primary objectives and hypotheses in Part A of the study are as follows: (1) Tolerability Objective: To evaluate the tolerability of the 9-valent HPV L1 VLP vaccine when administered to 16- to 26-year-old women. Hypothesis: 9-valent HPV L1 VLP vaccine administered to 16- to 26-year-old women is generally well-tolerated. (2) Immunogenicity Objective: To evaluate a formulation of 9-valent HPV L1 VLP vaccine for use in the efficacy evaluation in Part B. Hypothesis: 9-valent HPV L1 VLP vaccine selected for use in Part B generates anti-hpv 6, 11, 16, and 18 GMTs 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The GMTs for each of HPV types 6, 11, 16, and 18 will be tested separately. Noninferiority for GMTs is defined as statistically less than a 2-fold decrease.) Part B (tolerability, immunogenicity, and efficacy): Primary Part B Objectives/Hypotheses The primary objectives and hypotheses of Part B of the study are as follows: (1) Tolerability Objective: To evaluate the tolerability of the 9-valent HPV L1 VLP vaccine when administered to 16- to 26-year-old women. Hypothesis: 9-valent HPV L1 VLP vaccine administered to 16- to 26-year-old women is generally well-tolerated. (2) Efficacy Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine will reduce the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer, compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

225 V503 Statistical Analysis Plan 14 Hypothesis: Administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer, compared with GARDASIL in 16- to 26- year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type. (The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the vaccine efficacy be greater than 25%). (3) Immunogenicity Objective: To demonstrate that the 9-valent HPV L1 VLP vaccine induces noninferior GMTs for anti-hpv 6, 11, 16, and 18 compared to GARDASIL. Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 6, 11, 16, and 18 GMTs 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The GMTs for each of HPV types 6, 11, 16, and 18 will be tested separately. Noninferiority for GMTs is defined as the lower bound of the two-sided 95% confidence interval for the geometric mean ratio of 9-valent vaccine versus GARDASIL being greater than 0.67.) Secondary Part B Objectives/Hypotheses The secondary objectives and hypotheses of Part B of the study are as follows: (S1) Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine will reduce the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type. Hypothesis: Administration of 9-valent HPV L1 VLP vaccine to 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart compared with GARDASIL. (The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the vaccine efficacy be greater than 25%.) (S2) Objective: To demonstrate that 9-valent HPV L1 VLP vaccine is immunogenic with respect to HPV types 31, 33, 45, 52, and 58. Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 31, 33, 45, 52 and 58 seroconversion rates 4 weeks postdose 3 that are acceptable in 16- to 26-year-old adolescent and young adult women who are PCR negative and seronegative at baseline and PCR negative through one month after completion of the vaccination series for the relevant HPV type. (The seroconversion rate for each of HPV types BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

226 V503 Statistical Analysis Plan 15 31, 33, 45, 52, and 58 will be tested separately. Acceptability is defined as the lower bound of the two-sided 95% confidence interval for the seroconversion rate is greater than 90%.) (S3) Objective: To demonstrate that the 9-valent HPV L1 VLP vaccine induces noninferior immune responses with respect to seroconversion percentages for HPV 6, 11, 16, and 18 compared to GARDASIL. Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 6, 11, 16, and 18 seroconversion percentages by 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The seroconversion rates for each of HPV types 6, 11, 16, and 18 will be tested separately. Noninferiority is defined as the lower bound of the two-sided 95% confidence interval for the difference (9-valent vaccine minus GARDASIL ) in seroconversion rate being greater than 5 percentage points.) (S4) Objective: To quantify the amount by which the administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58- related cervical, vulvar and vaginal disease compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (S5) Objective: To evaluate the persistence of anti-hpv 6, 11, 16, 18, 31, 33, 45, 52, and 58 immune responses generated by 9-valent HPV L1 VLP vaccine. (S6) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of Pap test abnormalities (ASC-US [Positive for High Risk HPV] or worse). Exploratory Part B Objectives The exploratory objectives of Part B of the study are as follows: (E1) Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 12 months or longer apart compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). Hypothesis: Administration of 9-valent HPV L1 VLP vaccine to 16- to 26-yearold adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 12 months or longer, compared with GARDASIL. ( The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the vaccine efficacy be greater than 0.) BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

227 V503 Statistical Analysis Plan 16 (E2) (E3) (E4) (E5) (E6) (E7) (E8) Objective: To evaluate whether the administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of persistent HPV 16 and 18 infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart and HPV 16- and 18-related cervical, vulvar, and vaginal disease. Hypothesis: 9-valent HPV L1 VLP vaccine is non-inferior to GARDASIL in reducing the combined incidence of HPV 16- and/or 18-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart and HPV 16- and 18-related cervical, vulvar, and vaginal disease compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are PCR negative and seronegative at baseline and PCR negative through one month after completion of the vaccination series for the relevant HPV type. (The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the difference (9-valent vaccine minus GARDASIL ) in vaccine efficacy relative to historical placebo, for which the incidence rate is conservatively assumed to be 4 per 100 person-years as observed from the GARDASIL program, is greater than 15 percentage points.) Objective: To evaluate whether the administration of 9-valent HPV L1 VLP vaccine results in a combined incidence of HPV 6- and 11-related cervical, vulvar, and vaginal disease that is comparable to the combined incidence observed with GARDASIL. Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the combined incidence of CIN, AIS, and cervical cancer caused by any HPV type. Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the combined incidence of vulvar and vaginal disease caused by any HPV type. Objective: To characterize the titers of anti-hpv type 6/11 in both peripartum maternal blood and in cord blood of infants born to subjects for evaluating the potential impact of 9-valent HPV L1 VLP vaccine on recurrent respiratory papillomatosis. Objective: To evaluate the efficacy of the selected formulation of 9-valent HPV L1 vaccine against persistent HPV 35-, 39-, 51-, 56- and 59-related infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart and HPV 35-, 39-, 51-, 56-, and 59-related cervical, vulvar, and vaginal disease. Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of Pap test abnormalities (ASC-US [Positive for High Risk HPV] or worse) related to HPV Types 31, 33, 45, 52, & 58. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

228 V503 Statistical Analysis Plan 17 (E9) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of cervical biopsy and cervical definitive therapy treatments. (E10) Objective: To assess humoral immune responses to HPV type 6, 11, 16, 18, 31, 33, 45, 52, and 58 using a type specific anti-hpv L1 VLP IgG upon availability of a validated assay. Table 2 provides a summary of the principal endpoints for which the three 9-valent HPV L1 VLP vaccine formulation groups will be compared to GARDASIL to address the primary tolerability and immunogenicity hypotheses in Part A, and a summary of endpoints for which the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B and GARDASIL will be compared to address the efficacy, safety, and immunogenicity objectives of the study. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

229 V503 Statistical Analysis Plan 18 Table 2 Principal Safety, Efficacy, and Immunogenicity Endpoints to Address Study Objectives and Hypotheses Part A Study Phase Hypothesis or Objective Endpoints Part A Primary Safety Hypothesis Incidence of injection-site adverse experiences (Day 1 to Day 5 after any vaccination visit) Incidence of serious adverse experiences (Day 1 to Day 15 following any vaccination visit) Primary Immunogenicity Anti-HPV 6, anti-hpv 11, anti-hpv 16, and anti-hpv 18 geometric mean titers (GMTs) at Month 7 Hypothesis BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

230 V503 Statistical Analysis Plan 19 Principal Safety, Efficacy, and Immunogenicity Endpoints to Address Study Objectives and Hypotheses Part B (Cont.) Study Phase Hypothesis or Objective Endpoints Part B Primary Objectives/Hypotheses Primary Safety Hypothesis Incidence of injection-site adverse experiences (Day 1 to Day 5 after any vaccination visit). Incidence of elevated temperatures ( 100 F, oral equivalent) (Day 1 to Day 5 after any vaccination visit) Incidence of serious adverse experiences (Day 1 to Day 15 following any vaccination visit) Incidence of severe systemic adverse experiences (Day 1 to Day 15 following any vaccination visit) Incidence of serious vaccine related adverse experiences occurring at any time during the study Primary Efficacy Hypothesis Combined incidence of HPV 31/33/45/52/58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer (starting after Month 7) Primary Immunogenicity Hypothesis Anti-HPV 6, anti-hpv 11, anti-hpv 16, and anti-hpv 18 GMTs at Month 7 BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

231 V503 Statistical Analysis Plan 20 Principal Safety, Efficacy, and Immunogenicity Endpoints to Address Study Objectives and Hypotheses Part B (Cont.) Study Phase Hypothesis or Objective Endpoints Part B Secondary Objectives/Hypotheses (S1) Secondary Efficacy Hypothesis Combined incidence of HPV 31/33/45/52/58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart (starting after Month 7). (S2) Secondary Immunogenicity Hypothesis Proportion of subjects with HPV 31, 33, 45, 52, and 58 titers serostatus cutoff for the respective type at Month 7 (S3) Secondary Immunogenicity Hypothesis Proportion of subjects with HPV 6, 11, 16, and 18 titers serostatus cutoff for the respective type at Month 7 (S4) Secondary Efficacy Objective Combined incidence of HPV 31/33/45/52/58-related CIN(any grade), AIS, cervical cancer, genital wart, VIN(any grade), VaIN(any grade), vulvar cancer, or vaginal cancer (starting after Month 7) (S5) Secondary Immunogenicity Objective Proportion of subjects with HPV 6, 11, 16, 18, 31, 33, 45, 52, and 58 titers serostatus cutoff for the respective type at months 7, 12, 24, 36, and 42 Anti-HPV 6, anti-hpv 11, anti-hpv 16, anti-hpv 18, anti-hpv 31, anti-hpv 33, anti-hpv 45, anti- HPV 52, and anti-hpv 58 GMTs at months 7, 12, 24, 36, and 42 (S6) Secondary Efficacy Objective Combined incidence of Pap diagnoses ASC-US with positive high-risk probe, LSIL, HSIL, ASC-H, AGC, and cancer (starting after Day 1) BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

232 V503 Statistical Analysis Plan 21 Principal Safety, Efficacy, and Immunogenicity Endpoints to Address Study Objectives and Hypotheses Part B (Cont.) Study Phase Hypothesis or Objective Endpoints Exploratory Objectives/ Hypotheses (E1) Exploratory Efficacy Hypotheses Combined incidence of HPV 31/33/45/52/58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 12 months or longer apart (starting after Month 7) (E2) Exploratory Efficacy Hypotheses Combined incidence of HPV 16/18-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart and, HPV 16/18-related genital warts, VIN, VaIN, vulvar cancer, vaginal cancer, CIN, cervical AIS, and cervical cancer (starting after Month 7) (E3) Exploratory Efficacy Objective Combined incidence of HPV 6/11-related genital warts, VIN, VaIN, vulvar cancer, vaginal cancer, CIN, cervical AIS, and cervical cancer (starting after Month 7) Part B (E4) Exploratory Efficacy Objective Combined incidence of all CIN, AIS, and cervical cancer caused by any HPV type (starting after Day 1) (E5) Exploratory Efficacy Objective Combined incidence of all vulvar and vaginal disease caused by any HPV type (starting after Day 1) (E6) Exploratory Immunogenicity Objective Anti-HPV 6 and anti-hpv 11 titers in peripartum maternal blood and in cord blood of infants born to subjects (applicable only to sites that are able to collect this information) (E7) Exploratory Efficacy Objective Combined incidence of HPV 35/39/51/56/59-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart, and HPV 35/39/51/56/59-related genital warts, VIN, VaIN, vulvar cancer, vaginal cancer, CIN, cervical AIS, and cervical cancer (starting after Day 1) (E8) Exploratory Efficacy Objective Combined incidence of Pap diagnoses of HPV 31/33/45/52/58-related ASC-US with positive high-risk probe, LSIL, HSIL, ASC-H, AGC, and cancer (starting after Month 7) (E9) Exploratory Efficacy Objective Combined incidence of cervical biopsy and cervical definitive therapy treatments (starting after Day 1) AGC = Atypical glandular cells; AIS = Adenocarcinoma in situ; ASC-H = Atypical squamous cells, cannot exclude HSIL; ASC-US = Atypical squamous cells of undetermined significance; CIN = Cervical intraepithelial neoplasia; GMT = Geometric mean titer; HPV = Human papillomavirus; HSIL = High-grade squamous intraepithelial lesion; LSIL = Low-grade squamous intraepithelial lesion; VaIN = Vaginal intraepithelial neoplasia; VIN = Vulvar intraepithelial neoplasia. BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

233 V503 Statistical Analysis Plan 22 Criterion for Study Success The evaluation of primary safety and efficacy objectives for the 9-valent HPV L1 VLP vaccine will be based on combined data from subjects enrolled in Part A who received the selected 9-valent HPV L1 VLP vaccine dose or the comparator GARDASIL and subjects enrolled in Part B. The evaluation of primary immunogenicity objective for Part B will include those subjects enrolled in Part B only. V will be considered successful if success is achieved on all primary objectives/hypotheses (tolerability, efficacy, and immunogenicity) in Part B of the study. Success on the secondary efficacy hypothesis is not a requisite criterion to declare V successful. However, success on the secondary efficacy hypothesis provides supportive evidence for a claim that administration of a 3-dose regimen of the selected 9-valent HPV L1 VLP vaccine reduces the combined incidence of HPV 31/33/45/52/58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart compared with GARDASIL. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

234 V503 Statistical Analysis Plan Introduction 1.1 Objective of the Statistical Analysis Plan This Statistical Analysis Plan (SAP) is intended to be a comprehensive and detailed description of the strategy, rationale, and statistical techniques that will be used to evaluate the tolerability, efficacy and immunogenicity of Merck s first 9-valent HPV L1 VLP vaccine. This SAP covers V and subsequent amendments, provided that such amendments do not materially change the objectives, hypotheses, efficacy and immunogenicity endpoints, or the data analysis section of V Description of the Study and Objectives/Hypotheses Study Design V is a randomized, international, multi-centered, double-blinded, dose-ranging, tolerability, immunogenicity, and efficacy study of the 9-valent HPV L1 VLP vaccine operating under in-house blinding procedures. This study will be conducted in 2 parts (Part A and Part B). Part A (dose-ranging and tolerability): In Part A (dose-ranging and tolerability), approximately 1,240 healthy 16- to 26-year-old women will be randomized in equal numbers to one of three 9-valent HPV L1 VLP vaccine formulations or the comparator GARDASIL in a three-dose regimen (Day 1, Month 2, Month 6). The Part A vaccine formulations are listed in Table 3. Table 3 Part A Vaccine Formulations (0.5-mL Dose) Arm 1 HPV 6/ 11/ 16/ 18 20/ 40/ 40/ 20 (GARDASIL ) HPV 31 (mcg) HPV 33 (mcg) HPV 45 (mcg) HPV 52 (mcg) HPV 58 (mcg) Total VLP (mcg) AAHS (mcg) Estimated Subjects / 40/ 40/ / 40/ 60/ / 40/ 80/ AAHS=Merck Aluminum Adjuvant (amorphous aluminum hydroxyl phosphate sulfate) The primary goal in Part A is to identify a 9-valent HPV L1 VLP vaccine dose for further formal efficacy and immunogenicity evaluations (Part B). For details regarding the process for dose selection, see Section 5, Interim Analyses. The selected formulation will be communicated to the study sites by letter before Part B vaccine supplies are shipped to the sites. In the unlikely event that the SPONSOR is BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

235 V503 Statistical Analysis Plan 24 unable to determine an acceptable 9-valent HPV L1 VLP vaccine formulation for use in Part B, the study may not continue to Part B. After a dose is selected, the Merck study team will notify the Interactive Voice Response System (IVRS) of the selected 9-valent HPV L1 VLP vaccine formulation. For subjects enrolled in Part A, as each subject completes their Month 7 visit and all study data has been collected, study sites will call IVRS to receive notification of each subject's status. Part A subjects who received the formulations that were not chosen for further evaluation in Part B will be discontinued from the study after the Month 7 visit. In addition, subjects who received a 9-valent HPV L1 VLP dose with a lower total VLP amount than the formulation that is selected for use in Part B may be offered a single dose of GARDASIL after all Month 7 visits in Part A are complete. For the subjects in Part A who receive the selected 9-valent HPV L1 VLP vaccine dose or the comparator GARDASIL, the total duration of follow-up will be approximately 42 months from enrollment. The integrity of the blinding will be maintained in Part B because no vaccination group information for the continuing subjects will be released to the study sites, and the subjects continuing beyond Month 7 will remain randomized between the two continuing vaccination groups. The primary immunogenicity analysis in Part A will be conducted based on the Week 4 Postdose 3 (Month 7) immunogenicity results of the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B and the GARDASIL comparator. The formal statistical test of the Part A hypothesis based upon screened and cleaned Month 7 data will be provided in the final study report. An important goal in Part A of the study is to evaluate the tolerability of the 9-valent formulations relative to that of GARDASIL. HPV L1 VLP vaccines formulated with Merck s adjuvant AAHS have been well-studied clinically, have been found to be generally well-tolerated, and are currently approved for use as a vaccine in the form of GARDASIL. Because, however, this is the first time these specific formulations have been studied clinically, Merck & Co., Inc. personnel will perform an early and expedited review of blinded Vaccine Report Card (VRC) data once postdose 1 data for the first 40 subjects (approximately 10 subjects per arm) in Part A are available in the clinical database, and will request an unblinded review by the chairman of the Data Safety Monitoring Board (DSMB) if the rates of severe and/or serious adverse experiences are not consistent with the known adverse experience profile in the GARDASIL program. Subsequently, the DSMB, which consists of members that are independent of Merck Research Laboratories and a non-voting Merck & Co., Inc. statistician that is not associated with the study, will review VRC data when approximately 25% of Part A Day 1 vaccination visits are available in the study database, and will perform additional reviews at milestones specified in Section 6.1. Part B (tolerability, immunogenicity, and efficacy): In Part B (tolerability, immunogenicity, and efficacy), approximately 13,380 additional healthy 16- to 26-year-old women will be randomized in equal numbers to the 9-valent HPV L1 VLP vaccine dose selected from Part A or the comparator GARDASIL. The primary immunogenicity analyses for Part B will be the formal comparison of the selected 9-valent formulation to the GARDASIL control for HPV 6, 11, 16 and 18 and BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

236 V503 Statistical Analysis Plan 25 will include those subjects enrolled in Part B only. Because the immunogenicity data for Part A will have been analyzed previously, the Part A immunogenicity data for the selected 9-valent HPV L1 VLP vaccine dose and the comparator GARDASIL will not be used to address the Part B immunogenicity hypothesis. However, since neither the unblinded per subject safety nor efficacy data for Part A subjects will have been viewed previously by Merck personnel involved in the conduct of the study, the formal evaluations of tolerability and efficacy related to HPV 31, 33, 45, 52 and 58 will combine the data from subjects enrolled in Part A who received the selected 9-valent HPV L1 VLP vaccine dose or the comparator GARDASIL and subjects enrolled in Part B. The primary efficacy analysis is case driven and will be conducted after at least 30 primary efficacy cases have been observed. In addition, in order to have a sufficient duration of follow-up of vaccinated subjects, the primary efficacy analysis will not be performed before the median follow-up of Part B subjects is approximately 24 months after initial vaccination. For the subjects in Part B, the total duration of follow-up will be approximately 42 months from enrollment. Data Utilized in Immunogenicity and Efficacy Evaluations Serum samples collected at enrollment and Months 7, 12, 24, 36, and 42 will be tested for antibody to HPV vaccine types 6, 11, 16, 18, 31, 33, 45, 52 and 58 using a competitive luminex immunoassay (clia). The clia results at enrollment will be used to exclude subjects who test positive for vaccine HPV types from the per-protocol evaluation of vaccine efficacy and immunogenicity. The postvaccination clia results will be used to assess the immunogenicity of the vaccine. The primary evaluation of immune responses to a 3-dose regimen of 9-valent HPV L1 VLP vaccine in Part A and Part B will be based on serology data from serum sample collection at Month 7. The primary immunogenicity analyses will be done with the per protocol immunogenicity populations. All subjects in Part A will have clia measurements summarized for dose selection and the Part A primary immunogenicity analysis. All subjects enrolled in Part B will be included in the Part B immunogenicity hypothesis testing at Month 7. For the purpose of demonstrating vaccine induced anti-hpv antibody persistence at Months 12, 24, 36, and 42, a random sample of 2800 subjects (1400 each vaccine group) of Part B subjects will be tested and analyzed. Efficacy endpoints are based on data resulting from study procedures performed at both scheduled and unscheduled visits. The procedures performed at scheduled visits (see Table 1) for the purpose of efficacy data collection include serum sample at enrollment; and external genital inspection, collection of cervicovaginal specimens, and Pap testing at enrollment, Month 7, and every subsequent scheduled visit thereafter. The procedures that are typically performed at unscheduled visits for the purpose of efficacy data collection include repeat Pap tests, colposcopy, and biopsy. The cervicovaginal specimens collected at enrollment, Month 7, and all subsequent visits will be tested for HPV deoxyribonucleic acid (DNA) for types 6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59 using type-specific PCR assays. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

237 V503 Statistical Analysis Plan 26 The PCR results from the cervicovaginal specimens collected at enrollment and Month 7 will be used along with the clia results at enrollment to exclude subjects, who are positive for vaccine HPV types at baseline or who develop infection before 1 month following the administration of the third dose of the vaccination regimen, from the perprotocol evaluation of vaccine efficacy and immunogenicity. The PCR results for cervicovaginal swabs collected post-month 7 will be used for efficacy assessments. The results of the Pap tests performed throughout the study will be used to identify subjects with HPV disease. Mandatory protocol-specified guidelines will be used to triage subjects with Pap abnormalities to colposcopy [1]. Colposcopy is performed by an experienced colposcopist who has been trained in protocol-specific colposcopy procedures. If a lesion is seen during the colposcopy, it is biopsied. All biopsies will be read by a pathologist at a central laboratory (DCL) for the purpose of patient management. An independent panel of pathology experts (from hereon referred to as the Pathology Panel ) will also read the biopsies. The consensus diagnosis of the Pathology Panel will serve as the official diagnosis for endpoint determination. To determine the causal HPV type within a biopsy for efficacy assessments, Merck s typespecific HPV localizing assay (HPV Type-Specific Thinsection PCR Assay) will be performed on paraffin-embedded tissue samples. The Pathology Panel is blinded to the vaccination group assignment of study subjects, and to results of HPV typing analysis of specimens of study subjects, throughout the duration of the study. Though not a protocol-approved practice, it is possible that some subjects will have colposcopies performed by private physicians outside the context of the study rather than by the study sites. If biopsies or tissue specimens are collected during such colposcopies, these tissue specimens will not be collected according to protocol-specified procedures and will be read by the local laboratory used by the private physician rather than the central laboratory for patient management. Nevertheless, if these procedures have been conducted, every effort is made by the study site to obtain permission from the subject to access: (1) the Pap, colposcopy, or operative report; (2) the diagnostic slides; (3) the local pathology diagnosis of the tissue specimen; and (4) the tissue block itself for diagnostic slide preparation and HPV typing. If the slides are obtained, they will be sent to the pathology panel for diagnosis. If the tissue block is obtained, slides will be prepared by DCL and sent to the pathology panel and sections will be sent for PCR analysis. Data resulting from such specimens will not be used in the primary per-protocol efficacy analysis, but will be included in the analysis of the Full Analysis Set (see Section for definition of Full Analysis Set population) if a pathology panel diagnosis and/or PCR results for the specimen are available. Though the study duration is 42 months, Part B of the study employs a fixed event design whereby the primary efficacy analysis will be conducted after at least 30 cases of the primary efficacy endpoint (HPV 31/33/45/52/58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer) have been BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

238 V503 Statistical Analysis Plan 27 observed. In addition, in order to have a sufficient duration of follow-up of vaccinated subjects, the primary efficacy analysis will not be performed before the median follow-up of PART B subjects approximately 24 months after initial vaccination. At that time, the primary efficacy hypothesis will be tested. Success on this hypothesis will be achieved if the lower bound of a 2-sided 95% confidence interval (CI) for vaccine efficacy (VE) of 9-valent HPV L1 VLP vaccine relative to GARDASIL is greater than 25%. It is anticipated that at least 30 subjects will develop a case of the primary efficacy endpoint after an average of ~2.5 years of follow-up post-dose 1. If the primary analysis is conducted prior to the end of the study, an additional analysis of the vaccine efficacy will be conducted at the end of the study. The purpose of this analysis will be to provide information on the duration of the vaccine efficacy Objectives/Hypotheses Part A (dose-ranging and tolerability): Primary Part A Objectives/Hypotheses The primary objectives and hypotheses in Part A of the study are as follows: (1) Tolerability Objective: To evaluate the tolerability of the 9-valent HPV L1 VLP vaccine when administered to 16- to 26-year-old women. Hypothesis: 9-valent HPV L1 VLP vaccine administered to 16- to 26-year-old women is generally well-tolerated. (2) Immunogenicity Objective: To evaluate a formulation of 9-valent HPV L1 VLP vaccine for use in the efficacy evaluation in Part B. Hypothesis: 9-valent HPV L1 VLP vaccine selected for use in Part B generates anti-hpv 6, 11, 16, and 18 GMTs 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). Part B (tolerability, immunogenicity, and efficacy): Primary Part B Objectives/Hypotheses The primary objectives and hypotheses of Part B of the study are as follows: (1) Tolerability Objective: To evaluate the tolerability of the 9-valent HPV L1 VLP vaccine when administered to 16- to 26-year-old women. Hypothesis: 9-valent HPV L1 VLP vaccine administered to 16- to 26-year-old women is generally well-tolerated. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

239 V503 Statistical Analysis Plan 28 (2) Efficacy Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine will reduce the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer, compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type Hypothesis: Administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer, compared with GARDASIL in 16- to 26- year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type. (The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the vaccine efficacy be greater than 25%). (3) Immunogenicity Objective: To demonstrate that the 9-valent HPV L1 VLP vaccine induces noninferior GMTs for anti-hpv 6, 11, 16, and 18 compared to GARDASIL. Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 6, 11, 16, and 18 GMTs 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The GMTs for each of HPV types 6, 11, 16, and 18 will be tested separately. Noninferiority for GMTs is defined as the lower bound of the two-sided 95% confidence interval for the geometric mean ratio of 9-valent vaccine versus GARDASIL being greater than 0.67.) Secondary Part B Objectives/Hypotheses The secondary objectives and hypotheses of Part B of the study are as follows: (S1) Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine will reduce the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type. Hypothesis: Administration of 9-valent HPV L1 VLP vaccine to 16- to 26-yearold adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

240 V503 Statistical Analysis Plan 29 (S2) or longer compared with GARDASIL. (The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the vaccine efficacy be greater than 25%.) Objective: To demonstrate that 9-valent HPV L1 VLP vaccine is immunogenic with respect to HPV types 31, 33, 45, 52, and 58. Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 31, 33, 45, 52 and 58 seroconversion rates 4 weeks postdose 3 that are acceptable in 16- to 26- year-old adolescent and young adult women who are PCR negative and seronegative at baseline and PCR negative through one month after completion of the vaccination series for the relevant HPV type. (The seroconversion rate for each of HPV types 31, 33, 45, 52, and 58 will be tested separately. Acceptability is defined as the lower bound of the two-sided 95% confidence interval for the seroconversion rate is greater than 90%.) (S3) Objective: To demonstrate that the 9-valent HPV L1 VLP vaccine induces noninferior immune responses with respect to seroconversion percentages for HPV 6, 11, 16, and 18 compared to GARDASIL. (S4) Hypothesis: 9-valent HPV L1 VLP vaccine generates anti-hpv 6, 11, 16, and 18 seroconversion percentages by 4 weeks post dose 3 that are noninferior to those generated by GARDASIL in 16- to 26- year old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (The seroconversion rates for each of HPV types 6, 11, 16, and 18 will be tested separately. Noninferiority is defined as the lower bound of the two-sided 95% confidence interval for the difference (9-valent vaccine minus GARDASIL ) in seroconversion rate being greater than -5 percentage points.) Objective: To quantify the amount by which the administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related cervical, vulvar and vaginal disease compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). (S5) Objective: To evaluate the persistence of anti-hpv 6, 11, 16, 18, 31, 33, 45, 52, and 58 immune responses generated by 9-valent HPV L1 VLP vaccine. (S6) Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of Pap test abnormalities (ASC-US [Positive for High Risk HPV] or worse). Exploratory Part B Objectives The exploratory objectives of Part B of the study are as follows: (E1) Objective: To demonstrate that administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 12 month or longer apart compared with GARDASIL in BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

241 V503 Statistical Analysis Plan 30 (E2) (E3) (E4) (E5) (E6) 16- to 26-year-old adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type(s). Hypothesis: Administration of 9-valent HPV L1 VLP vaccine to 16- to 26-yearold adolescent and young adult women who are seronegative at Day 1 and PCR negative Day 1 through Month 7 to the relevant HPV type reduces the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 12 months or longer, compared with GARDASIL. ( The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the vaccine efficacy be greater than 0.) Objective: To evaluate whether the administration of 9-valent HPV L1 VLP vaccine reduces the combined incidence of persistent HPV 16 and 18 infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart and HPV 16- and 18-related cervical, vulvar, and vaginal disease. Hypothesis: 9-valent HPV L1 VLP vaccine is non-inferior to GARDASIL in reducing the combined incidence of HPV 16- and/or 18-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart and HPV 16- and 18-related cervical, vulvar, and vaginal disease compared with GARDASIL in 16- to 26-year-old adolescent and young adult women who are PCR negative and seronegative at baseline and PCR negative through one month after completion of the vaccination series for the relevant HPV type. (The statistical criterion for success requires that the lower bound of the two-sided 95% confidence interval for the difference (9-valent vaccine minus GARDASIL ) in vaccine efficacy relative to historical placebo, for which the incidence rate is conservatively assumed to be 4 per 100 person-years as observed from the GARDASIL program, is greater than 15 percentage points.) Objective: To evaluate whether the administration of 9-valent HPV L1 VLP vaccine results in a combined incidence of HPV 6- and 11-related cervical, vulvar, and vaginal disease that is comparable to the combined incidence observed with GARDASIL. Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the combined incidence of CIN, AIS, and cervical cancer caused by any HPV type. Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the combined incidence of vulvar and vaginal disease caused by any HPV type. Objective: To characterize the titers of anti-hpv type 6/11 in both peripartum maternal blood and in cord blood of infants born to subjects for evaluating the potential impact of 9-valent HPV L1 VLP vaccine on recurrent respiratory papillomatosis. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

242 V503 Statistical Analysis Plan 31 (E7) (E8) (E9) Objective: To evaluate the efficacy of the selected formulation of 9-valent HPV L1 vaccine against persistent HPV 35-, 39-, 51-, 56- and 59-related infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart and HPV 35-, 39-, 51-, 56-, and 59-related cervical, vulvar, and vaginal disease. Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of Pap test abnormalities (ASC-US [Positive for High Risk HPV] or worse) related to HPV Types 31, 33, 45, 52, & 58. Objective: To evaluate the impact of administration of 9-valent HPV L1 VLP vaccine on the incidence of cervical biopsy and cervical definitive therapy treatments. (E10) Objective: To assess humoral immune responses to HPV type 6, 11, 16, 18, 31, 33, 45, 52, and 58 using a type specific anti-hpv L1 VLP IgG upon availability of a validated assay. 2. Study Participant Characteristics Balance across vaccination groups with respect to study participant characteristics will be assessed observationally. For all summaries of baseline study participant characteristics, separate summaries will be provided for Part A and Part B. It is important to note that all Part B summaries will include the subjects from Part A who received GARDASIL or the selected 9-valent HPV L1 VLP vaccine formulation. 2.1 Subject Accounting The total number of subjects enrolled, randomized, vaccinated, and discontinued will be summarized overall and by vaccination group. The total number of women screened (i.e., subjects who signed an informed consent and were given a baseline number) who were not randomized will be presented over all vaccination groups combined. 2.2 Subject Characteristics The following subject characteristics will be summarized overall and by vaccination group: Age Gender Race/ethnicity Number of lifetime male and female sexual partners Tobacco use Pregnancy history Age at first sexual intercourse Lifetime gynecologic medical history History of sexually transmitted diseases (STDs) Contraceptive use Baseline (Day 1) Pap test results Baseline (Day 1) presence of other STDs BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

243 V503 Statistical Analysis Plan 32 All summaries listed above will be provided for all subjects randomized, for the set of subjects who are in the primary per-protocol efficacy population for at least one HPV type (for Part B only), for the set of subjects who are in the primary Part B per-protocol immunogenicity population for at least one HPV type, and by geographic region (North America, Latin America, Europe, Asia-Pacific). 2.3 Baseline HPV Status As shown in Table 1, all subjects enrolled in the study are scheduled to have an endo/ectocervical (EEC) swab and labial/vulvar/perineal and perianal (LVPP) swabs collected at the Day 1 visit for the purpose of assessing baseline PCR positivity to HPV 6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58 and 59. In addition, a serum sample will be collected from all subjects at the Day 1 visit for the purpose of assessing baseline seropositivity to the nine 9-valent HPV L1 VLP vaccine HPV types (HPV 6, 11, 16, 18, 31, 33, 45, 52, and 58). A subject will be considered PCR positive and/or seropositive at baseline (Day 1) as follows: Definition of HPV Positive by PCR: A subject will be classified as positive to a given HPV type by PCR if the subject has at least one Day 1 LVPP or EEC swab, or a biopsy, that is PCR positive to the given HPV type regardless of whether or not any other swab results are missing. HPV PCR positivity at Day 1 is indicative of current infection at Day 1. Definition of HPV Positive by Serology: A subject will be classified as positive to a given HPV type if the subject s anti-hpv titer is greater than or equal to the serostatus cutoffs 30, 16, 20, 24, 10, 8, 8, 8, and 8 mmu/ml corresponding to the HPV types 6, 11, 16, 18, 31, 33, 45, 52, and 58 respectively. For Part A subjects randomized to the 9-valent HPV L1 VLP vaccine formulation not selected for efficacy evaluation, the Day 1 HPV 6 serostatus cutoff is 20 mmu/ml. Baseline (Day 1) serostatus and PCR status for each HPV type will be presented by vaccination group, overall, and by geographic region for all subjects randomized into the study. For each HPV type, the proportions of subjects who are seronegative and PCR negative, seropositive and PCR negative, seronegative and PCR positive, or seropositive and PCR positive will be presented by vaccination group, overall, and by geographic region In addition, baseline (Day 1) PCR status for each of the 14 tested HPV types will be tabulated by each of the following: Baseline Pap test abnormality Age Geographic Region Lifetime Number of Sexual Partners (None, 1-2, 3 or more) Within each baseline category, PCR will be classified according to the following groupings: - PCR Negative to all 14 tested HPV types (Types 6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59) BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

244 V503 Statistical Analysis Plan 33 - PCR Positive to 1 or more HPV types - PCR Positive to 1 or more 9-valent HPV types - PCR Positive to 1 or more A9 species HPV type (includes HPV 16, 31, 33, 35, 52, and 58) - PCR Positive to 1 or more A7 species HPV type (includes HPV 18, 39, 45, and 59) - PCR Positive to 1 or more Other Tested High-Risk HPV type (includes HPV 51 and 56) - PCR Positive to HPV 6 or 11 - Positive to 2 or more Tested HPV types - Positive to 3 or more Tested HPV types 2.4 Prior and Concomitant Medications The number and percentage of subjects with specific prior medications or vaccinations prior to each study vaccination will be summarized by vaccination group. The relevant pre-injection day ranges for summarization will be 3 days for medications, 14 days for non-live virus vaccines and 21 days for live virus vaccines. Similarly, the number and percent of subjects with specific concomitant medications and vaccinations within 14 days following any study vaccination will be summarized by vaccination group. 2.5 New Medical Conditions For subjects in Part B and subjects who were randomized to the selected 9-valent HPV L1 VLP vaccine or GARDASIL groups in Part A, new medical conditions will be summarized by vaccination group during the vaccination period (post Day 1 to Month 7) and efficacy follow-up period (post Month 7). A separate summary of new medical history during the vaccination period (post Day 1 to Month 7) will also be provided for all subjects in Part A by vaccination group. 2.6 Sexual Activity and Contraceptive Use Since HPV infection is sexually transmitted, it is important to evaluate sexual behavior during the vaccination period (post Day 1 to Month 7) and efficacy follow-up period (post Month 7) by vaccination group. The number of new sexual partners and the rate of new sexual partners in the 6 months prior to study start, post Day 1 to Month 7, and post Month 7 will be summarized by vaccination group for subjects in Part B (including those subjects who were randomized to the selected 9-valent HPV L1 VLP vaccine or GARDASIL groups in Part A). The rate of new sexual partners in the 6 months prior to study start is defined as the number of new sexual partners in the last 6 months reported by each subject at the Day 1 visit, divided by the person-time (0.5 years) for each subject. The rate of new sexual partners after the Day 1 visit is defined as the number of new sexual partners reported during the indicated time interval divided by the person-time accrued during the time interval. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

245 V503 Statistical Analysis Plan 34 For the same subjects as above, contraceptive use will be summarized by vaccination group for the post Day 1 to Month 7 period and post Month 7. The number and percent of subjects with specific contraceptive use by contraceptive category for the vaccination period (post Day 1 to Month 7) and the efficacy follow-up period (after Month 7) will be reported by vaccination group. 3. Efficacy Analyses 3.1 Efficacy/Immunogenicity Endpoints Efficacy Endpoints Endpoints Associated with Primary Efficacy Hypothesis (Part B) The primary efficacy endpoint is the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer. To be classified as an endpoint case for the primary analysis, a subject must develop at least one of the following after the completion of the Month 7 visit: 1. High-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma related to HPV 31, 33, 45, 52 or 58. This endpoint is defined to have occurred if on a single cervical biopsy, ECC, LEEP or Conization (cold knife/laser) specimen, there is: (a) a HPV Vaccine Program Pathology Panel consensus diagnosis of CIN (grade 2 or 3), AIS, or cervical cancer; AND (b) detection of at least 1 of HPV types 31, 33, 45, 52 or 58 by Thinsection PCR in an adjacent section from the same tissue block. 2. High-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer related to HPV 31, 33, 45, 52 or 58. This endpoint is defined to have occurred if on a single biopsy or excised tissue, there is: (a) a HPV Vaccine Program Pathology Panel consensus diagnosis of VIN (grade 2 or 3), VaIN (grade 2 or 3), vulvar cancer, or vaginal cancer; AND (b) detection of at least 1 of HPV types 31, 33, 45, 52 or 58 by thin section PCR in an adjacent section from the same tissue block. An additional efficacy endpoint will be evaluated in a sensitivity analysis. In the sensitivity analysis, a case of disease will require a consensus diagnosis of HPV disease by the pathology panel, the appropriate HPV type detected in an adjacent section of the same tissue block AND at least 1 specimen immediately prior to or subsequent to the biopsy or definitive therapy sample that is positive to the same HPV type that is found in the biopsy. Endpoints Associated with Secondary Efficacy Hypothesis (Part B) S1: Incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart. This endpoint is defined to have occurred if a subject who, after completion of the Month 7 visit, is positive for the same HPV type by the HPV 31/33/45/52/58 PCR assay to at least 1 common gene in 2 or more consecutive cervicovaginal/external genital swab, BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

246 V503 Statistical Analysis Plan 35 biopsy, or definitive therapy samples obtained at two or more consecutive visits 6 months (± 1 month) or longer. Persistent infection is defined based on the protocol visit intervals. Scheduled visits for the collection of LVPP/EEC swabs are at 6-monthly intervals, so most cases of persistent infection will be based on consecutive specimens obtained 6 months apart; however, due to protocol-allowable deviations for scheduled visit intervals, the minimum length of time between samples for a subject to be counted as a case of persistent infection will be 4 months. If a subject has 2 consecutive samples that are PCR positive to at least one common gene for the same HPV type and at least one of the samples is a biopsy or definitive therapy sample showing pathologic evidence of HPV disease, then the subject will be considered a case of persistent infection without regard to the length of time between the samples. S4: Incidence of HPV 31-, 33-, 45-, 52-, and 58-related cervical, vulvar and vaginal disease. This endpoint is defined as a subject who, after completion of the Month 7 visit, is found to have a biopsy or excised tissue specimen with (a) a HPV Vaccine Program Pathology Panel consensus diagnosis of CIN(any grade), AIS, cervical cancer, genital wart, VIN(any grade), VaIN(any grade), vulvar cancer, or vaginal cancer, AND (b)detection of at least 1 of HPV types 31, 33, 45, 52, or 58 by Thinsection PCR in an adjacent section from the same tissue block. S6: Incidence of Pap test abnormalities. A case is defined as a subject who, after completion of the Day 1 visit, is found to have a Pap test result of atypical squamous cells of undetermined significance (ASC-US) with positive HPV probe or worse (ASC-US [positive for High Risk HPV], atypical squamous cells suggestive of a high-grade intraepithelial lesion [ASC-H], HSIL, LSIL, atypical glandular cells, or adenocarcinoma in situ [AIS]). Endpoints Associated with Exploratory Efficacy Hypothesis E1: Incidence of HPV 31-, 33-, 45-, 52-, and 58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 12 months or longer apart. This endpoint is defined to have occurred if a subject who, after completion of the Month 7 visit, is positive for the same HPV type by the HPV 31/33/45/52/58 PCR assay to at least 1 common gene in 2 or more consecutive cervicovaginal/external genital swab, biopsy, or definitive therapy samples obtained for over a period of at least 12 months (within ± 1 month windows) or longer. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

247 V503 Statistical Analysis Plan 36 E2: Combined incidence of HPV 16- and 18-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart or HPV 16- and 18-related cervical, vulvar, and vaginal disease. To be classified as a case for this endpoint, a subject must develop at least one of the following after the completion of the Month 7 visit: 1. Persistent HPV 16 or 18 Infection. The definition of this endpoint is the same as those given for the incidence of HPV 31-, 33-, 45-, 52- and 18-related persistent infection in the secondary efficacy hypothesis (S1), except in relation to HPV types 16 and 18 instead of types 31, 33, 45, 52 and HPV 16- and 18-related cervical, vulvar, and vaginal disease. The definition of this endpoint is the same as those given for the incidence of HPV 31-, 33-, 45-, 52- and 58-related cervical, vulvar, and vaginal disease in the secondary efficacy hypothesis (S4), except in relation to HPV types 16 and 18 instead of types 31, 33, 45, 52 and 58. E3: Incidence of HPV 6- and 11-related cervical, vulvar and vaginal Disease. The definition of this endpoint is the same as those given for the incidence of HPV 31-, 33-, 45-, 52- and 58-related cervical, vulvar, and vaginal disease in the secondary efficacy hypothesis (S4), except in relation to HPV types 6 and 11 instead of types 31, 33, 45, 52 and 58. E4: Incidence of CIN, AIS, or cervical cancer caused by any HPV type (vaccine or nonvaccine). This endpoint is defined as a subject who, after completion of the Day 1 visit, has a consensus diagnosis of CIN (any grade), AIS, or cervical cancer on a single cervical biopsy, ECC, LEEP or Conization (cold knife/laser) specimen. E5: Incidence of vulvar or vaginal disease caused by any HPV type (vaccine or nonvaccine). This endpoint is defined as a subject who, after completion of the Day 1 visit, has a consensus diagnosis of genital wart, VIN (any grade), VaIN (any grade), vulvar cancer, or vaginal cancer on a single biopsy or excised tissue. E7: Combined incidence of HPV 35-, 39-, 51-, 56- and 59-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart and HPV 35-, 39-, 51-, 56- and 59-related cervical, vulvar, and vaginal disease This endpoint is defined as a subject who, after completion of the Day 1 visit, is found to have a HPV 35-, 39-, 51-, 56-, or 59-related persistent infection or HPV 35-, 39-, 51-, 56-, or 59- related cervical, vulvar and vaginal disease. The definition for persistent infection and disease is the same as those given for the combined incidence of HPV 16 and 18-related persistent infection and disease in the exploratory efficacy hypothesis (E2), except in relation to HPV types 35, 39, 51, 56 and 59 instead of types 16 and 18. E8: Incidence of HPV 31-, 33-, 45-, 52-, and 58-related Pap test abnormalities. A case is defined as a subject who, after completion of the Month 7 visit, is found to have a Pap test diagnosis of ASC-US [with positive for High Risk HPV ], atypical squamous cells suggestive of a high-grade intraepithelial lesion [ASC-H], HSIL, LSIL, atypical glandular BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

248 V503 Statistical Analysis Plan 37 cells, or adenocarcinoma in situ [AIS], and at least one cervicovaginal/external genital swab that is PCR positive for HPV 31, 33, 45, 52, or 58 at the same study visit. E9: Incidence of cervical biopsy and cervical definitive therapy treatments. A subject is defined to have an incident cervical biopsy or definitive therapy procedure if she has a cervical biopsy or definitive therapy after completion of the Day 1 visit. All primary, secondary and exploratory efficacy endpoints that are being evaluated in this study are summarized in Table 4. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

249 V503 Statistical Analysis Plan 38 Table 4 Planned Primary, Secondary and Exploratory Analyses of Efficacy in V (Part B Subjects Combined With Part A Subjects Randomized to Receive Selected 9-Valent HPV L1 VLP Vaccine or GARDASIL ) Analysis Population Endpoints PPE HN-TS All-HN FAS S0P1 S1P0 S1P1 Analysis Method VE Estimation Time-to-Event Endpoints to Support the Primary Efficacy Hypothesis HPV 31/33/45/52/58-Related CIN 2/3, AIS, invasive cervical carcinoma, VIN (P) 2/3, VaIN 2/3, vulvar cancer, or vaginal cancer Sensitivity Analysis of the Primary Efficacy Endpoint HPV 31/33/45/52/58- Related CIN 2/3, AIS, invasive cervical carcinoma, VIN (P) 2/3, VaIN 2/3, vulvar cancer, or vaginal cancer with Additional PCR Positivity HPV 31/33/45/52/58- Related CIN 2/3, AIS, invasive cervical carcinoma, VIN (P) 2/3, VaIN 2/3, vulvar cancer, or vaginal cancer -Imputing Case Status for Subjects Lost to Follow-up Endpoints to Support the Secondary Efficacy Hypothesis HPV 31/33/45/52/58-Related Persistent Infection 6 months (±1 month) (P) Endpoints for Supplementary Analyses of the Primary and Secondary Efficacy Endpoints HPV 31/33/45/52/58-Related CIN 2/3, AIS, invasive cervical carcinoma, VIN 2/3, VaIN 2/3, vulvar cancer, or vaginal cancer - By HPV Type - By Lesion Type (P) (P) HPV 31/33/45/52/58- Related Persistent Infection 6 months (±1 month) - By HPV Type (P) Endpoints to Support Secondary Efficacy Objectives HPV 31/33/45/52/58-Related CIN 1/2/3, AIS, invasive cervical carcinoma, genital wart, VIN 1/2/3, VaIN 1/2/3, vulvar cancer, or vaginal cancer - Overall - By HPV Type - By Lesion Type Pap Test Abnormalities (ASC-US [+ for HR HPV probe] or worse) - Overall - High Grade (ASC-H, HSIL or worse) - By Specific Abnormality (P) (P) (P) (P) (P) (P) BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

250 V503 Statistical Analysis Plan 39 Planned Primary, Secondary and Exploratory Analyses of Efficacy in V Part B Subjects Combined With Part A Subjects Randomized to Receive Selected 9-Valent HPV L1 VLP Vaccine or GARDASIL ) (Cont.) Analysis Population Endpoints PPE HN-TS All-HN FAS S0P1 S1P0 S1P1 Endpoints to Support Exploratory Efficacy Objectives HPV 31/33/45/52/58-Related Documented Infection 12 months (±1 month) - Overall - By HPV Type HPV 16/18-Related Persistent Infection 6 months (±1 month) or Disease - Overall - By HPV Type - By Lesion Type HPV 6/11-Related Disease - Overall - By HPV Type - By Lesion Type All CIN, AIS and Cervical Cancer Caused By Any HPV type - Overall - By Lesion Type All Vulvar and Vaginal Disease Caused By Any HPV Type - Overall - By Lesion Type HPV 35/39/51/56/59-Related Persistent Infection 6 months (±1 month) or Disease - Overall - By HPV Type - By Lesion Type Pap Test Abnormalities Potentially Related to HPV 31/33/45/52/58 - Overall - By Specific Abnormality - By HPV Type (P) (P) (P) (P) (P) (P) (P) (P) (P) (P) (P) (P) (P) (P) (P) (P) (P) (P) Analysis Method VE Estimation Time-to-Event BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

251 V503 Statistical Analysis Plan 40 Planned Primary, Secondary and Exploratory Analyses of Efficacy in V Part B Subjects Combined With Part A Subjects Randomized to Receive Selected 9-Valent HPV L1 VLP Vaccine or GARDASIL ) (Cont.) Analysis Population Analysis Method Endpoints PPE HN-TS All-HN FAS S0P1 S1P0 S1P1 VE Estimation Time-to-Event Endpoints to Support Exploratory Efficacy Objectives Cervical Biopsy and Cervical Definitive Therapy Procedures - Overall - By Procedure Type (P) (P) Endpoint cases are counted starting after Month 7 in the PPE population. For all other analysis populations, endpoint cases are counted starting after Day 1. Disease is defined as any of genital warts, VIN, VaIN, vulvar cancer, vaginal cancer, CIN, cervical AIS, or cervical cancer. In addition to meeting all criteria for defining a case of the primary efficacy endpoint, subjects who are a case of the primary efficacy endpoint based on a cervical, vulvar, or vaginal biopsy, ECC, or definitive therapy specimen are also required to be PCR positive to the same HPV type for at least 1 specimen immediately prior to or immediately after the HPV 31/33/45/52/58-related disease specimen. In addition to the estimation of vaccine efficacy of 9-valent HPV L1 VLP vaccine relative to GARDASIL, efficacy of the 9-valent HPV L1 VLP vaccine will be estimated relative to the combined placebo population of Protocols 007 and 012 from the GARDASIL (V501) clinical program using the methods in Hasselblad and Kong [3]. (P) indicates that the analysis population is the primary approach for evaluating the given endpoint. AIS = Adenocarcinoma in-situ; All-HN = All HPV naïve; CIN = Cervical intraepithelial neoplasia; FAS = Full analysis set; HN-TS = HPV-naïve to the relevant type; HPV = Human papillomavirus; HR = High-risk; PPE = Per protocol efficacy; S0P1 = Day 1 seronegative and PCR positive; S1P0 = Day 1 seropositive and PCR negative; S1P1 = Day 1 seropositive and PCR positive. BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

252 V503 Statistical Analysis Plan Immunogenicity Endpoints Endpoints Associated with Primary Immunogenicity Hypotheses (Parts A and B) The primary immunogenicity endpoints are geometric mean titers (GMTs) corresponding to HPV types 6, 11, 16, and 18 at Week 4 Postdose 3 (Month 7). The secondary immunogenicity endpoints are a) the seroconversion percentages to each of HPV 6, 11, 16, and 18 by Week 4 Postdose 3; and b) the percentages of subjects who seroconvert for each HPV type (31, 33, 45, 52 and 58) by Week 4 Postdose 3. Seroconversion is defined as changing serostatus from seronegative to seropositive, by Week 4 Postdose 3. A subject with a clia titer at or above the serostatus cutoff for a given HPV type is considered seropositive for that type. Additional secondary immunogenicity endpoints include the geometric mean titers (GMTs) for HPV 31, 33, 45, 52 and 58 at Week 4 Postdose 3 and the immune responses for HPV 6, 11, 16, 18, 31, 33, 45, 52, and 58 at the persistence time points (Months 12, 24, 36, and 42). The exploratory immunogenicity endpoints are the geometric mean titers (GMTs) to HPV 6 and HPV 11 measured in peripartum maternal blood and cord blood of infants born to women vaccinated at sites participating in cord blood data collection. All primary, secondary and exploratory immunogenicity endpoints that will be evaluated in this study are summarized in Table 5. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

253 V503 Statistical Analysis Plan 42 Table 5 Primary, Secondary and Exploratory Analyses of Immunogenicity in Part A and Part B of V Time Point(s) Analysis Population PPI ANSS S0P0 S0P1 S1P0 S1P1 Noninferiority/ Acceptability margin Point Estimate and 95% CI Part A: Endpoints to Support Interim Month 3 Immunogenicity Analysis Anti-HPV 6 GMT D1, M3 Anti-HPV 11 GMT D1, M3 Anti-HPV 16 GMT D1, M3 Anti-HPV 18 GMT D1, M3 Anti-HPV 31 GMT D1, M3 Anti-HPV 33 GMT D1, M3 Anti-HPV 45 GMT D1, M3 Anti-HPV 52 GMT D1, M3 Anti-HPV 58 GMT D1, M3 Percent Seroconversion to HPV 6 D1, M3 Percent Seroconversion to HPV 11 D1, M3 Percent Seroconversion to HPV 16 D1, M3 Percent Seroconversion to HPV 18 D1, M3 Percent Seroconversion to HPV 31 D1, M3 Percent Seroconversion to HPV 33 D1, M3 Percent Seroconversion to HPV 45 D1, M3 Percent Seroconversion to HPV 52 D1, M3 Percent Seroconversion to HPV 58 D1, M3 Part A: Endpoints to Support Primary Part A Immunogenicity Hypotheses Anti-HPV 6 GMT D1, M7 2-fold Anti-HPV 11 GMT D1, M7 2-fold Anti-HPV 16 GMT D1, M7 2-fold Anti-HPV 18 GMT D1, M7 2-fold p- value RCD Longitudinal Plot BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

254 V503 Statistical Analysis Plan 43 Primary, Secondary and Exploratory Analyses of Immunogenicity in Part A and Part B of V (Cont.) Time Point(s) Analysis Population PPI ANSS S0P0 S0P1 S1P0 S1P1 Noninferiority/ Acceptability margin Point Estimate and 95% CI Part B : Endpoints to Support Primary Part B Immunogenicity Hypotheses Anti-HPV 6 GMT D1, M7 1.5-fold Anti-HPV 11 GMT D1, M7 1.5-fold Anti-HPV 16 GMT D1, M7 1.5-fold Anti-HPV 18 GMT D1, M7 1.5-fold Part B : Endpoints to Support Secondary Part B Immunogenicity Hypotheses/Objectives Percent Seroconversion to HPV 6 D1, M7 5% Percent Seroconversion to HPV 11 D1, M7 5% Percent Seroconversion to HPV 16 D1, M7 5% Percent Seroconversion to HPV 18 D1, M7 5% Percent Seroconversion to HPV 31 D1, M7 LCB > 90% Percent Seroconversion to HPV 33 D1, M7 LCB > 90% Percent Seroconversion to HPV 45 D1, M7 LCB > 90% Percent Seroconversion to HPV 52 D1, M7 LCB > 90% Percent Seroconversion to HPV 58 D1, M7 LCB > 90% Anti-HPV 31 GMT D1, M7 Anti-HPV 33 GMT D1, M7 Anti-HPV 45 GMT D1, M7 Anti-HPV 52 GMT D1, M7 Anti-HPV 58 GMT D1, M7 p- value RCD Longitudinal Plot BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

255 V503 Statistical Analysis Plan 44 Primary, Secondary and Exploratory Analyses of Immunogenicity in Part A and Part B of V (Cont.) Time Point(s) Analysis Population PPI ANSS S0P0 S0P1 S1P0 S1P1 Noninferiority/ Acceptability margin Point Estimate and 95% CI p- value RCD Longitudinal Plot Part B : Endpoints to Support Secondary Part B Immunogenicity Hypotheses/Objectives (Cont.) Anti-HPV 6 GMT D1, M7;12;24;36;42 Anti-HPV 11 GMT D1, M7;12;24;36;42 Anti-HPV 16 GMT D1, M7;12;24;36;42 Anti-HPV 18 GMT D1, M7;12;24;36;42 Anti-HPV 31 GMT D1, M7;12;24;36;42 Anti-HPV 33 GMT D1, M7;12;24;36;42 Anti-HPV 45 GMT D1, M7;12;24;36;42 Anti-HPV 52 GMT D1, M7;12;24;36;42 Anti-HPV 58 GMT D1, M7;12;24;36;42 Percent Seropositivity to HPV 6 D1, M7;12;24;36;42 Percent Seropositivity to HPV 11 D1, M7;12;24;36;42 Percent Seropositivity to HPV 16 D1, M7;12;24;36;42 Percent Seropositivity to HPV 18 D1, M7;12;24;36;42 Percent Seropositivity to HPV 31 D1, M7;12;24;36;42 Percent Seropositivity to HPV 33 D1, M7;12;24;36;42 Percent Seropositivity to HPV 45 D1, M7;12;24;36;42 Percent Seropositivity to HPV 52 D1, M7;12;24;36;42 Percent Seropositivity to HPV 58 D1, M7;12;24;36;42 Part B : Endpoints to Support Exploratory Part B Immunogenicity Objective Relating to Peripartum Maternal Blood and Infant Cord Blood Anti-HPV 6 GMT At time of infant birth All evaluable subjects Anti-HPV 11 GMT At time of infant birth All evaluable subjects Tested at the α= level (1-sided) with minor multiplicity adjustment for the Month 3 interim immunogenicity analysis. The first 60% of subjects randomized at each site in Part B will be selected for the Part B immunogenicity evaluations. ANSS = All Naïve Subjects with Serology; D1 = Day 1; GMT = Geometric mean titer; M3 = Month 3; M7 = Month 7; M7;12;24;36;42 = Months 7, 12, 24,36, and 42; PPI = Per-protocol immunogenicity; RCD = Reverse Cumulative Distribution; S0P0 = Day 1 seronegative and PCR negative; S0P1 = Day 1 seronegative and PCR positive; S1P0 = Day 1 seropositive and PCR negative; S1P1 = Day 1 seropositive and PCR positive. BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

256 V503 Statistical Analysis Plan Study Participant Populations Efficacy Populations Efficacy evaluations will be conducted in the populations indicated in Table 4. All efficacy evaluations will include subjects enrolled in Part B of the study combined with subjects from Part A who were randomized to receive either the formulation of the 9- valent HPV L1 VLP vaccine selected for efficacy evaluation or GARDASIL. Per Protocol Efficacy (PPE) Populations The approach for the primary efficacy hypothesis will be per-protocol. To be included in the primary efficacy evaluation, subjects must have received all 3 vaccinations with the correct clinical material within 1 year, have 1 or more follow-up visits following Month 7, and have Month 7 PCR swab samples collected within 14 to 72 days postdose 3. Subjects must also be seronegative to the appropriate HPV type(s) at Day 1 and PCR negative to the appropriate HPV type(s) on all cervicovaginal swabs and biopsies from Day 1 through Month 7. For evaluations of efficacy with respect to either HPV type 6 or HPV type 11, subjects must be negative at the aforementioned time points for both HPV 6 and 11. For evaluations of efficacy with respect to any other vaccine HPV type, subjects need only be negative at the aforementioned time points for the HPV type under evaluation. Modified Intent-to-Treat Analyses Three modified intent-to-treat (MITT) populations will be analyzed to support the analyses in the per protocol population, to explore the robustness of the vaccine efficacy and to evaluate the secondary and exploratory efficacy objectives. Subjects who received at least 1 vaccination and have any follow-up visit following the first vaccination will be included in all 3 populations. In all 3 populations, events occurring anytime after Day 1 are eligible to be counted as endpoint cases if they meet the appropriate case definitions. The populations are defined as follows: 1. HPV Type-Specific Naïve Population (HN-TS) - This population will include only subjects who are seronegative and PCR-negative at enrollment to the appropriate HPV types, as defined in the Primary Per Protocol Analysis. 2. Full Analysis Set (FAS) - This population will not be restricted to subjects who are seronegative and PCR-negative at enrollment to the appropriate HPV types, as defined in the Primary Per Protocol Analysis. This analysis in essence includes all subjects randomized, except those who did not receive any vaccination or those without any follow-up. 3. All HPV Naïve Population (All-HN) This population includes only subjects who (a) are seronegative and PCR-negative at enrollment to all 9 vaccine HPV types; (b) are PCR-negative at enrollment to all other non-vaccine HPV types for which PCR assays will be available; and (c) have a normal Pap test result at enrollment. In all the MITT populations, subjects will be counted in the vaccination group to which they were originally randomized, regardless of the actual clinical material received at each of the three vaccination visits. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

257 V503 Statistical Analysis Plan Immunogenicity Populations Immunogenicity evaluations will be conducted in the populations indicated in Table 5. Part A immunogenicity evaluations will include all subjects enrolled in Part A. All subjects enrolled in Part B will be included in Part B immunogenicity hypothesis testing at Month 7. A random sample of 2800 subjects (1400 each vaccine group) of Part B will be evaluated at Months 12, 24, 36, and 42 for the purpose of demonstrating vaccine induced anti-hpv antibody persistence. Day 1 Seronegative and PCR Negative (S0P0) Population The analysis of Postdose 2 immune response among Part A subjects will be conducted on the S0P0 population, defined as subjects who receive both the Day 1 and Month 2 vaccinations within acceptable day ranges, are seronegative and PCRnegative at Day 1 to the relevant HPV type, and have serum samples for evaluation of Postdose 2 immune response collected within acceptable day ranges. Per Protocol Immunogenicity (PPI) Population The primary approach to the analyses of immunogenicity will also be per protocol (immunogenicity). The per-protocol immunogenicity population is defined in the same way as the per protocol efficacy population with the following exceptions: The subjects vaccination visits must each occur within acceptable day ranges relative to Day 1. The subjects must have at least 1 Month 7 serology result within 21 to 49 days Postdose 3. All Type-Specific Naïve Subjects with Serology Populations (ANSS) A supportive immunogenicity analysis will be carried out on the all type-specific naïve subjects with serology population. To be included in this analysis subjects must be seronegative and PCR negative to the relevant HPV type(s) as described above, and must have provided serology data Other Efficacy and Immunogenicity Populations Three additional analysis populations will be used to explore efficacy and immune responses among subjects in different stages of infection with HPV. These populations are: Day 1 Seronegative and PCR Positive (S0P1): Efficacy analyses conducted in this subgroup are intended to explore the therapeutic efficacy of 9-valent HPV L1 VLP vaccine among a population of women with ongoing HPV infection who have no detectable immune response to the infection (based on HPV titers) at the start of the vaccination regimen. Day 1 Seropositive and PCR Negative (S1P0): Efficacy analyses conducted in this subgroup are intended to explore the efficacy of 9-valent HPV L1 VLP vaccine among a population of women who have prior history of HPV infection but are HPV infection-free at the start of the vaccination regimen. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

258 V503 Statistical Analysis Plan 47 Day 1 Seropositive and PCR Positive (S1P1): Efficacy analyses conducted in this subgroup are intended to explore the therapeutic efficacy of 9-valent HPV L1 VLP vaccine among a population of women with ongoing HPV infection who already have detectable immune response to the infection (based on HPV titers) at the start of the vaccination regimen. 3.3 Approaches to Efficacy Analysis Efficacy Analyses A listing of planned analyses of efficacy is given in Table 4. Estimation of Vaccine Efficacy: Analyses of efficacy endpoints will generally involve computations of point and confidence interval (CI) estimates of vaccine efficacy (VE), defined as 100%*(1-relative risk). Relative risk is the ratio of risk of becoming a case of a particular efficacy endpoint in the 9-valent HPV L1 VLP vaccine group over the risk in the GARDASIL group. VE, as defined, is equivalent to the percent reduction in the risk of becoming a case of a particular endpoint in the group vaccinated with 9-valent HPV L1 VLP vaccine relative to the risk in the GARDASIL group. VE will be computed using exact methods [2] and as described in Section Time-to-Event Plots: In addition to computing point and CI estimates of VE, Kaplan- Meier time-to-event plots showing cumulative incidence distributions of endpoints in each of the vaccine groups will be provided for selected efficacy endpoints, in selected analysis populations. These plots will provide supportive information Immunogenicity Analyses A listing of analyses planned for addressing the study immunogenicity objectives is given in Table 5. Primary Immunogenicity Hypothesis (Part A) The primary hypothesis concerns non-inferiority of anti-hpv 6, 11, 16, and 18 GMTs at Week 4 Postdose 3. The hypothesis may be written H 0 : GMT N /GMT G 1/2 versus H 1 : GMT N /GMT G > 1/2 where GMT G represents the GMT at Week 4 Postdose 3 in subjects receiving GARDASIL and GMT N represents the GMT at Week 4 Postdose 3 in subjects receiving the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B. The hypotheses of noninferiority of GMTs for HPV types 6, 11, 16 and 18 will be based on one-sided tests of non-inferiority comparing GMTs for each component. Four ANOVA models (one per HPV type) with a response of log individual titers and a fixed effect for vaccination group will be used. Part A is considered a pilot study, and the statistical criterion for noninferiority in these tests corresponds to the lower bound of the confidence interval for the fold-difference in GMTs between the 2 groups, (9-valent HPV L1 VLP vaccine group/gardasil group), excluding a decrease of 2-fold or more for each component (i.e., the lower bound of the confidence interval for the fold-difference being greater than 0.5). BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

259 V503 Statistical Analysis Plan 48 Primary Immunogenicity Hypothesis (Part B) The primary hypothesis concerns non-inferiority of anti-hpv 6, 11, 16, and 18 GMTs at Week 4 Postdose 3. The hypothesis may be written H 0 : GMT N /GMT G 0.67 versus H 1 : GMT N /GMT G > 0.67 where GMT G represents the GMT at Week 4 Postdose 3 in subjects receiving GARDASIL and GMT N represents the GMT at Week 4 Postdose 3 in subjects receiving the 9-valent HPV L1 VLP vaccine. The hypotheses of noninferiority of GMTs will be tested using the same methodology as for Part A. Because the confirmatory immunogenicity evaluation of the selected 9-valent HPV L1 VLP vaccine formulation will be conducted in Part B, the statistical criterion for non-inferiority in these tests corresponds to the lower bound of the 95% confidence interval for the fold-difference in GMTs between the 2 groups, (9-valent HPV L1 VLP vaccine group/gardasil group), excluding a decrease of 1.5-fold or more for each component (i.e., the lower bound of the confidence interval for the fold-difference being greater than 0.67). Secondary Immunogenicity Hypotheses (Part B) The secondary hypothesis of noninferiority of seroconversion percentages for each of HPV types 6, 11, 16, and 18 will be addressed by 4 one-sided tests of noninferiority (one corresponding to each HPV type) conducted at the α=0.025 level (1-sided). For each HPV type, the hypotheses to be tested are H 0 : p 1 -p versus H a : p 1 -p 2 > where p 1 is the proportion of subjects who seroconvert at Week 4 Postdose 3 in recipients of the 9-valent HPV L1 VLP vaccine and p 2 is the proportion of subjects who seroconvert at Week 4 Postdose 3 in GARDASIL recipients. The tests above will be conducted based on methods developed by Miettinen and Nurminen [4] for testing the equivalence of 2 proportions. The secondary hypothesis of acceptability of anti-hpv seroconversion rates (for HPV types 31, 33, 45, 52 and 58) will be tested by one-sided tests of the proportions of subjects seroconverting (1 test per HPV type). The hypotheses to be tested are: H 0 : P N 0.9 versus H 1 : P N > 0.9 (for each of the HPV Types 31, 33, 45, 52, and 58), where P N represents the true response rate at Week 4 Postdose 3 of subjects receiving a 3- dose regimen of a formulation of 9-valent HPV L1 VLP vaccine. These tests will be conducted based on the exact 95% confidence interval for a binomial proportion. Rejection of the null hypothesis in these tests corresponds to the lower bound of the 95% confidence interval for the percentage of subjects seroconverting in the 9-valent HPV L1 VLP vaccine group being greater than 90% for the given HPV type. Estimation of GMTs: Anti-HPV responses for each of the 9 vaccine HPV types will be summarized by vaccination groups in terms of geometric mean titers (GMTs) with 95% confidence intervals at each time point when serum samples are collected. Anti-HPV responses during the period of vaccination and persistence from Month 7 onwards will be BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

260 V503 Statistical Analysis Plan 49 investigated using longitudinal plots of the GMTs. The geometric mean titers (GMTs) to HPV 6 and HPV 11 in peripartum maternal blood and cord blood of infants born to women vaccinated at sites participating in cord blood data collection will be summarized descriptively. Estimation of Percent Seroconversions: For each vaccination group, the percentages and the corresponding 95% CIs of subjects who seroconvert to HPV types 6, 11, 16, 18, 31, 33, 45, 52, and 58 will be provided. Confidence intervals associated with the percentages of subjects who seroconvert will be computed using exact methods. Reverse Cumulative Distribution (RCD) plots will be used for cross-sectional graphical comparisons of distributions of HPV titers between the vaccination groups. Additional Analyses of Immunogenicity by Day 1 Serostatus: Anti-HPV 6, 11, 16, and 18 GMTs following administration of GARDASIL or 9-valent HPV L1 VLP vaccine will be summarized separately for subjects who are: - Seropositive and PCR negative to the relevant HPV type at Day 1 - Seronegative and PCR positive to the relevant HPV type at Day 1 - Seropositive and PCR positive to the relevant HPV type at Day 1 Anti-HPV 31, 33, 45, 52, and 58 GMTs in subjects with natural HPV infection will also be measured, although the capacity to interpret the immune response in these subjects over time will be more limited than in the previous placebo controlled studies of GARDASIL. GMTs of Day 1 anti-hpv 31, 33, 45, 52, and 58 levels for subsets of subjects who are (1) seropositive and PCR negative to the relevant HPV type; and (2) seropositive and PCR-positive to the relevant HPV type. The Day 1 GMTs for each HPV type in each of these subgroups will be observationally compared to the each postvaccination GMT in subjects randomized to the selected 9-valent HPV L1 VLP vaccine formulation in Part B who are seronegative and PCR negative to the relevant HPV type at Day Analysis of Correlates of Protection An exploratory analysis of any breakthrough cases of the primary endpoint will be performed to try to establish the relationship between immune markers for HPV types 31, 33, 45, 52, and 58 and protection from persistent infection and disease endpoints. Specifically, the immune responses in the non-breakthrough vaccinees will be compared with the immune responses in the breakthrough cases at all available time points (i.e., at Months 7, 24, 36 and 42). However, if there are few cases among the vaccine recipients, there may be insufficient information to establish an immune correlate of protection in this study. An additional exploratory analysis of any breakthrough cases of HPV 6/11/16/18-related persistent infection or disease will be performed in subjects who develop an endpoint in either the 9-valent HPV L1 VLP vaccine group or the GARDASIL group. Immune markers will be explored using the per-protocol efficacy population. However, if the vaccine is highly efficacious, very few primary endpoints will be observed in the vaccine group in the per-protocol analysis. If this is the case, it may be more informative BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

261 V503 Statistical Analysis Plan 50 to explore correlates of protection in the HN-TS population. This population still includes only initially HPV naïve subjects but is likely to yield more endpoints in the relevant vaccination group. 3.4 Definition of Compliance Measure Compliance is defined in this study as receipt of all scheduled study vaccinations. To summarize compliance, the numbers of subjects who receive each vaccination will be tabulated by vaccination group, separately for Part A and Part B. Compliance with the planned vaccination schedule (0, 2, 6 months) will be described by histograms of actual intervals between vaccinations relative to the expected interval. Another important compliance measure in this study is the degree of subject compliance with follow-up visits following the completion of the vaccination series. Since colposcopy and external genital lesion inspection are required procedures for the identification of potential study endpoints of disease, subjects who miss study follow-up visits or who visit their private gynecologists for examinations and/or treatment rather than the study investigators represent missed opportunities to observe endpoints. To summarize compliance with respect to follow-up study visits, the numbers of subjects who complete each follow-up visit will be tabulated by vaccination group. To provide measures of compliance with respect to follow-up study visits, the average interval between scheduled study visits will be computed for each subject. In addition, the numbers and percentages of subjects in each group who had biopsies or excision procedures performed outside of the study will be summarized, and the numbers and percentages of subjects in each group who had biopsies or excision procedures performed outside of the study for whom the tissue samples were unavailable to the SPONSOR will be summarized. Differences in these measures between the vaccination groups will be assessed observationally and the potential impact on the efficacy analyses results will be noted. 4. Safety Analyses The primary safety objective of Part A and Part B of V is to evaluate the tolerability of the 9-valent HPV L1 VLP vaccine when administered to 16- to 26-year-old women. Assessments of safety will be conducted using statistical analyses as well as by clinical review. All safety analyses and summaries will be provided separately for: (1) all subjects in Part A; and (2) all subjects in Part B combined with subjects from Part A who received the selected 9-valent HPV L1 VLP vaccine formulation or GARDASIL. For the Part A summaries and analyses that involve computing (incidence) risk differences between vaccination groups, the risk difference between each 9-valent HPV L1 VLP vaccination group and GARDASIL will be computed. For the Part B summaries and analyses that involve computing (incidence) risk differences between vaccination groups, the risk difference between the selected 9-valent HPV L1 VLP vaccination group and GARDASIL will be computed. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

262 V503 Statistical Analysis Plan Study Participant Population All safety analyses will be performed on the All-Subjects-As-Treated (ASaT) population. The ASaT population includes all randomized subjects who receive at least 1 injection and have follow-up data, and assigns subjects to the vaccination group corresponding to the actual clinical material received. For subjects who received injections of study vaccines corresponding to a sequence that does not correspond to any of the protocoldefined vaccination groups (i.e., cross-treated subjects), a safety profile listing will be created separate from the safety reports that will be provided for the protocol-defined vaccination groups. 4.2 Extent of Exposure to Drug Exposure to vaccine will be summarized by tabulating the number of subjects receiving each of the study vaccinations. 4.3 Adverse Experiences Statistical analyses of adverse experiences will follow the 3-tiered analysis approach commonly used by the SPONSOR when conducting safety assessments. Tier 1 analysis compares vaccination groups by providing incidence summaries by vaccination group, computing (incidence) risk differences (RDs) and 95% CIs of RDs between vaccination groups, and computing p-values corresponding to tests of significance of the RDs. Tier-1 adverse experiences include (1) injection-site adverse experiences prompted for on the VRC (i.e., redness, swelling, and pain/tenderness/soreness) occurring Day 1 through Day 5 following any vaccination, and (2) elevated temperature ( F [ 37.8ºC]) from Day 1 to Day 5 following any vaccination. Tier 2 analysis follows the tier 1 analysis approach, except p-values are not computed. The Tier-2 adverse experience summaries include (1) specific systemic adverse experiences within 14 days following any vaccination occurring in 1% of subjects in any vaccination group, (2) injection-site adverse experiences not prompted for on the VRC occurring Day 1 to Day 5 following any vaccination in 1% of subjects in any vaccination group, (3) serious adverse experiences occurring within 14 days (Day 1 to Day 15) following any vaccination, (4) serious vaccine-related adverse experiences observed at any time during the study, and (5) severe injection-site adverse experiences Day 1 through Day 5 following any vaccination visit. Tier 3 analysis compares vaccination groups by providing only incidence summaries by vaccination group. Tier-3 adverse experiences include summaries (counts and percentages) by vaccination group for any other adverse experiences, including all injection-site adverse experiences occurring from Day 1 to Day 5 following each vaccination visit and all systemic adverse experiences occurring within 14 days (Day 1 to Day 15) of each vaccination visit. Regardless of incidence, all adverse experiences will be summarized as frequencies and percentages by vaccination group, by vaccination visit and across all vaccination visits. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

263 V503 Statistical Analysis Plan 52 Specific adverse experiences for which summaries will be provided are identified in Table 6. Incidence is defined as (number of subjects with the indicated endpoint divided by the total number of subjects with follow-up over the relevant period)*100%. Confidence intervals (95%) and 2-sided p-values will be provided without adjusting for multiplicity. The method of Miettinen and Nurminen [4] will be used for all comparisons of vaccination groups. Risk differences, confidence intervals, and p-values will be computed across all sites combined (i.e., no stratification by study sites will be performed). Risk differences will be provided for any systemic adverse experience occurring in at least 1% of subjects in either vaccination group. For the measured adverse experiences of redness and swelling, 0 to 1 inch will be categorized as mild intensity, >1 inch to 2 inches will be categorized as moderate intensity, and >2 inches will be categorized as severe intensity. 4.4 Clinical Safety Study subjects are required to be using adequate contraception through Month 7 (1 month after completion of the vaccination regimen) of the study. Thus, pregnancies are expected to be rare. However, all serious adverse experiences occurring during pregnancy, all serious adverse experiences occurring during lactation, all pregnancy outcomes, and all serious adverse experiences occurring in infants who were exposed to study vaccination (either during conception or via breast milk) will be summarized and compared descriptively between the vaccination groups. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

264 V503 Statistical Analysis Plan 53 Adverse Experience Endpoint Table 6 Planned Analyses of Adverse Experience Endpoints Follow-Up Period After Any Vaccination Visit Day 1 Day 1 to Day 5 to Day 15 Day 1 through Month 7 Any Time During Study Summaries/Analyses Risk Difference and 95% CI P-Value Incidence Clinical AEs Any AE Deaths Injection site AEs AEs of pain/tenderness, swelling, and redness Severe injection site AEs Number (%) of subjects by maximum intensity rating, across the categories of pain/tenderness, swelling, and redness Maximum intensity ratings of injection site AEs, across the categories of pain/tenderness, swelling, and redness Maximum intensity rating of injection site AEs, within each of the categories of pain/tenderness, swelling, and redness Systemic AEs Systemic AEs Number (%) of subjects by maximum intensity rating, over all systemic AEs Maximum intensity rating of systemic AEs, over all systemic AEs Temperatures Elevated temperatures # Maximum temperatures AEs of Special Interest Serious AEs Serious VR AEs New medical conditions Serious AEs of infants exposed to study vaccine during conception BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

265 V503 Statistical Analysis Plan 54 Adverse Experience Endpoint AEs of Special Interest Serious AEs of infants exposed to study vaccine during lactation Planned Analyses of Adverse Experience Endpoints (Cont.) Follow-Up Period After Any Vaccination Visit Day 1 Day 1 to Day 5 to Day 15 Day 1 through Month 7 Any Time During Study Incidence Summaries/Analyses Risk Difference and 95% CI P-Value Serious AEs during pregnancy Pregnancy Outcomes The day of vaccination is counted as Day 1. Day 1 to Day 5 refers to the period within 4 days of a vaccination. Day 1 to Day 15 refers to the period within 14 days of a vaccination. Summaries will be provided by vaccination group separately for the subjects in Part A and subjects in Part B, including those from Part A who were received GARDASIL or the selected 9-valent HPV L1 VLP vaccine formulation. For the measured adverse experiences of redness and swelling 0 to 1 inch will be categorized as mild, >1 inch to 2 inches will be categorized as moderate and >2 inches will be categorized as severe. Defined as the difference in incidence between the vaccination groups being compared. Risk difference and 95% CIs will be provided only for systemic AEs occurring in at least 1% of subjects in either vaccination group. # Defined as maximum (over the follow-up period) temperature 37.8 C ( 100 F, oral equivalent). Distribution of maximum temperatures over the relevant follow-up period. Including new medical conditions considered potentially of an autoimmune nature. AEs = Adverse experiences; CI = Confidence interval; VR = Vaccine-related. BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

266 V503 Statistical Analysis Plan Interim Analyses The primary goal of Part A is to identify a 9-valent HPV L1 VLP vaccine formulation to further evaluate in the efficacy portion of the study (Part B). Because the 9-valent HPV L1 VLP vaccine represents an important medical advance, offering the potential of significant prophylactic cancer coverage in addition to that already provided by GARDASIL, a dose-selection procedure that allows the study to move to Part B prudently, but as efficiently as possible, will be employed. Specifically, the doseselection decision will be based on an interim analysis of the tolerability and immunogenicity data from Part A of the study. No interim analyses are planned for Part B of this study. Safety data will be reviewed periodically by the DSMB as discussed in Section Timing and Procedure The Part A interim analysis will be conducted when ~100% of subjects in Part A have Week 4 Postdose 2 serology data available. It is important to note that the interim analysis will focus on Postdose 2 data, while the study hypotheses concern Week 4 Postdose 3 (Month 7) responses in Part A subjects who receive the 9-valent HPV L1 VLP formulation that is selected for use in Part B or GARDASIL. However, based on previous studies, the ratio of the HPV 6, 11, 16, or 18 GMTs for the 9-valent HPV L1 VLP vaccine versus GARDASIL 1 month after the second dose of vaccine (post-dose 2) is expected to be predictive of the ratio after the completion of the 3-dose series. The procedure for the dose selection will be as follows: Immunogenicity and tolerability summaries of the postdose 2 data will be produced by unblinded Merck & Co., Inc. personnel A senior management committee (sidmc) within the SPONSOR, who is not directly involved with study conduct, will review the immunogenicity summaries by vaccination group, will choose a dose, and will relay the dose selection to the Data Safety Monitoring Board (DSMB). The DSMB will review the immunogenicity data and all tolerability data and will communicate to the SPONSOR senior management committee (sidmc) if, based on the clinical judgment of the DSMB, there is concern with respect to the incidence of serious adverse experiences, severe injection-site adverse experiences, or severe systemic adverse experiences for the selected dose formulation as compared to the incidence observed in the GARDASIL group. If no tolerability concerns are raised, the senior management committee (sidmc) will inform the study team of the selected dose formulation, and the study will proceed to Part B with the selected formulation and the control arm GARDASIL. If the DSMB expresses a concern about the tolerability profile of the selected dose, either the DSMB will offer a counter recommendation (subject to senior management committee approval) or the senior management committee will select another dose and again request approval from the DSMB. Once the senior management committee (sidmc) and the DSMB agree on a BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

267 V503 Statistical Analysis Plan 56 chosen formulation, the senior management committee at the SPONSOR will inform the study team of the formulation selected for Part B. After the senior management committee (sidmc) notifies the study team as to which 9-valent HPV L1 VLP vaccine formulation has been selected for use in Part B, the study team will communicate this information to the appropriate regulatory authorities. In the unlikely event that the SPONSOR is unable to determine an acceptable 9-valent HPV L1 VLP vaccine formulation for use in Part B, the study may not continue to Part B. For the summary of the conduct of the Part A dose selection, see Section 5.4 for detail. 5.2 Dose Selection Criteria The selected 9-valent HPV L1 VLP vaccine formulation should be generally welltolerated, maintain the immunogenicity of GARDASIL against HPV types 6, 11, 16, and 18, and extend prophylactic efficacy coverage to include new HPV types (31, 33, 45, 52, and 58). To identify such a dose, the senior management committee with the assistance of DSMB, will seek a formulation that: 1. produces immune responses that closely match those of the HPV types contained in GARDASIL, as measured by the anti-hpv 6, anti-hpv 11, anti-hpv 16, and anti-hpv 18 GMTs; 2. produces high immune responses for HPV 31, HPV 33, HPV 45, HPV 52, and HPV 58, as measured by the percent of subjects who seroconvert to each HPV type; AND 3. has a favorable tolerability profile. In the event that 2 or more 9-valent HPV L1 VLP vaccine formulations have immune responses with respect to HPV 6, HPV 11, HPV 16, and HPV 18 that are similar to those of GARDASIL, AND have high seroconversion percentages for HPV types 31, 33, 45, 52, and 58, the goal will be to select the formulation with the lowest total VLP concentration. 5.3 Analysis Details The interim immunogenicity analysis in Part A will be conducted in a subset of subjects who 1) are seronegative at Day 1 to the relevant HPV type, without regard to PCR status at Day 1; 2) receive the first 2 doses of 9-valent HPV L1 VLP vaccine or GARDASIL in acceptable day ranges; and 3) have post-vaccination serum samples collected in acceptable day ranges. Summary statistics of immune responses by vaccination group (GMTs, seroconversion rates, confidence intervals, RCDF graphs) will be provided. No serology or randomization data for individual study participants will be disclosed. No formal hypothesis testing will be done at the time of the interim analysis. To evaluate the tolerability of the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B by the sidmc, the DSMB will be provided with the number and percentage of subjects with: 1) serious adverse experiences; 2) severe injection-site adverse experiences; and 3) severe systemic adverse experiences. No confidence intervals or p- values will be provided for these tolerability summaries. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

268 V503 Statistical Analysis Plan Summary of the conduct of the Part A dose selection As planned, and described in Sections 5.1 through 5.3 of this Statistical Analysis Plan, summaries of immunogenicity data by vaccination group were to be reviewed by a senior management committee at the SPONSOR (sidmc) who is not directly involved with study conduct. The senior management committee at the SPONSOR selected the 9-valent HPV L1 VLP (30/40/60/40/20/20/20/20/20 mcg each of HPV Types 6/11/16/18/31/33/45/52/58) vaccine for use in Part B. The sidmc relayed the dose selection to the DSMB. The DSMB reviewed the unblinded tolerability profile of all three 9-valent HPV L1 VLP vaccine formulations and the GARDASIL control group, concluded there were no safety concerns with any of the formulations, and agreed with the sidmc s selected dose for study in Part B. Because immunogenicity was comparably high for each of the 9-valent vaccine formulations, the group level summaries were provided to the Merck study team who facilitated the selection of the preferred 9-valent formulation for pivotal efficacy study in Part B. The Merck study team members reviewed grouped immunogenicity data only and were not unblinded to any individual treatments, or any safety or efficacy data. An unblinded monitor and liaison not involved directly in the conduct of the study were also designated to review Part A safety and tolerability data with regulatory agencies as requested. The clinical study team remains blinded to any individual treatments or any safety or efficacy data. 6. Data/Study Reviewing Committees 6.1 Data Safety and Monitoring Board (DSMB) The Data and Safety Monitoring Board (DSMB) assesses the effects of the study vaccine during the trial. The DSMB will review all available tolerability data at the following milestones during the Part A and Part B vaccination periods: 1. Milestone 1: VRC data from approximately 25% of Part A Day 1 vaccination visits are available in the study database, 2. Milestone 2: VRC data from approximately 100% of Part A Day 1 vaccination visits are available in the study database, 3. Milestone 3: VRC data from approximately 25% of Part A Month 2 vaccination visits are available in the study database, 4. Milestone 4: VRC data from approximately 100% of Part A Month 2 vaccination visits are available in the study database (this milestone will involve review of the dose selected by the senior management committee, as discussed in Section ), 5. Milestone 5: VRC data from approximately 100% of Part A Month 6 vaccination visits are available in the study database (review will include all available VRC data from Part B), 6. Milestone 6: VRC data from approximately 25% of Part B Month 2 vaccination visits are available in the study database, 7. Milestone 7: VRC data from approximately 50% of Part B Month 2 vaccination visits are available in the study database, BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

269 V503 Statistical Analysis Plan Milestone 8: VRC data from approximately 100% of Part B Month 2 vaccination visits are available in the study database In addition to above milestones, the DSMB may also meet on an ad-hoc basis at the request of the DSMB chair, the protocol team, the site, or a study investigator. With the exception of a non-voting Merck & Co., Inc. statistician, the members of the DSMB are independent of Merck Research Laboratories and clinical investigators participating in this trial, and will not have any other involvement in the study, nor will they have any relation to the study participants. The DSMB monitors the trial for evidence of adverse effects of the study vaccine using the guidelines proposed by the protocol. The DSMB may recommend any steps to ensure the safety and integrity of the trial. Furthermore, the DSMB may recommend that the trial be terminated or that specific high-risk patient groups be withdrawn from the study, if any subgroup manifests serious or widespread side effects. To guarantee the unrestricted performance of its task, the DSMB may receive the individual study morbidity and mortality data from a designated MRL statistician. Unblinded reports of all serious adverse experiences in this study from the MRL Worldwide Adverse Experience database and the clinical database will be presented to the DSMB. The DSMB minutes will summarize the actions and deliberations of the DSMB and will be made available to Merck & Co., Inc. at the conclusion of the trial. A full manual of operations for the DSMB will be written and approved by the DSMB [5]. 6.2 Merck Senior Management Committee (sidmc) As described in Section 5, a Merck Senior Management Committee (sidmc), who is not directly involved with the conduct of the study, will review interim immunogenicity summaries from Part A of the study and, with the assistance of the DSMB, be responsible for selecting the dose for evaluation in Part B of the study. The sidmc will include a small number of Merck & Co., Inc. personnel who are experts in medical, regulatory, statistical and epidemiological aspects of clinical trials. Further details regarding the process for dose selection and the procedures used to maintain blinding will be provided in a guidance document, which will be approved before the senior management committee or the DSMB convenes to review the dose-selection data [6]. 6.3 Scientific Advisory Committee (SAC) A Scientific Advisory Committee (SAC) will be convened prior to the initiation of Part B of the trial. This committee is comprised of internal Merck representatives and external scientific leaders in the field of gynecology, oncology, and/or HPV research. The SAC will convene to provide advice on protocol direction, the statistical analyses, and interpretation and publication of study results. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

270 V503 Statistical Analysis Plan Statistical Technical Issues 7.1 Planned Statistical Power and Sample Size Sample Size Considerations for Primary and Secondary Efficacy Hypotheses The sample size for this study is determined by the primary efficacy endpoint of the combined incidence of HPV 31-, 33-, 45-, 52-, and 58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN 2/3), vulvar cancer, or vaginal cancer. This study is powered based on a fixed event design. The power for a given number of observed primary and secondary endpoints was computed using the method of Chan and Bohidar [2]. In planning the study, the potential for cross-protection was considered. It is possible that the HPV types contained in GARDASIL may afford some protection against infection and disease related to other HPV types, and in particular HPV types 31, 33, 45, 52 and 58. This may reduce the incidence of the primary and secondary endpoints in subjects receiving GARDASIL, relative to completely untreated subjects. Based on existing crossprotection data from the GARDASIL program with respect to HPV infection and related disease, it was assumed that this reduction may be up to 30% [7,8]. If assuming the vaccine efficacy for the 9-valent HPV L1 VLP versus placebo is 88%, taking the 30% cross protection into account, the true vaccine efficacy for the 9-valent HPV L1 VLP versus GARDASIL will be 83%. With the lower bound of the confidence interval for the true vaccine efficacy must be >25% under the alternative hypothesis, 30 primary efficacy cases are needed to provide at least 90% power to declare the vaccine efficacious (α=0.025, one-sided). To ensure that at least 30 cases are accrued based on a median follow-up of 30 months post-vaccination 1, approximately 14,000 subjects are needed for the efficacy evaluation (13,380 subjects enrolled for Part B combined with the ~620 subjects in the GARDASIL arm and selected 9-valent dose formulation arm from Part A). The assumptions used in sample size calculation are: exclusion rate due to seropositivity at Day 1 and/or PCR positive between Day 1 and Month 7 to HPV types 31, 33, 45, 52, and 58 is not greater than 23% attrition rate between Day 1 and Month 7 is not greater than 10% annual attrition rate post-month 7 is not greater than 5% the annual incidence rates for the primary efficacy endpoint is 0.35 per 100 personyears median follow-up time is 30 months post randomization BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

271 V503 Statistical Analysis Plan Sample Size Considerations for Primary and Secondary Immunogenicity Hypotheses Part A With the enrollment of 310 subjects per group to obtain around 209 to 230 evaluable subjects per arm and assuming a nominal α= level (1-sided) (See Section 7.5 for multiplicity considerations), this study has an overall power of >99%. The calculations are based on the following assumptions: 1. 15%, 15%, 18% and 10% of the subjects are initially seropositive or PCR positive to HPV Types 6, 11, 16, and 18, respectively. 2. The expected attrition rate through the Month 7 time point is 10%. 3. The expected percentage of subjects excluded due to vaccinations or serology samples out of range is no more than 8% 4. The standard deviations of the natural logarithm of the Month 7 titers are no more than 1.2 for each HPV type when expressed as mmu/ml, the unitage used in previous GARDASIL studies. 5. HPV Types 6, 11, 16, and 18 responses are identical in the GARDASIL group and at least one 9-valent HPV L1 VLP vaccine formulation. The expected responses rates and noninferiority margins are as shown in Table 7 for each antigen. For GARDASIL, these are based on data available from previous studies. Antigen Table 7 Power for Part A Immunogenicity Hypotheses of Protocol V Parameter Expected rates/sd Non-inferiority Margin N evaluable (1) Primary Hypothesis - Non-Inferiority of 9-valent HPV L1 VLP Vaccine to GARDASIL Power HPV 6 GMT σ = fold decrease 218 >0.99 HPV 11 GMT σ = fold decrease 218 >0.99 HPV 16 GMT σ = fold decrease 209 >0.99 HPV 18 GMT σ = fold decrease 230 >0.99 Overall Power for GMT hypothesis >0.99 BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

272 V503 Statistical Analysis Plan 61 Part B The overall sample size in the study is determined by the efficacy analyses. With approximately 13,380 subjects in Part B, this study has over 99% power for the primary immunogenicity hypothesis of noninferiority with respect to geometric mean titers for HPV types 6, 11, 16, and 18. The calculations are based on the same assumptions as in the Part A immunogenicity analysis. The expected standard deviations and power for the primary immunogenicity hypotheses are as shown in Table 8 for each antigen. "N evaluable" in the table reflects number of subjects that will be included in the per protocol immunogenicity population after accounting for various exclusion reasons under the assumption. The study will still have an overall power of > 99% even when GMT N /GMT G =0.9 or 0.8. Table 8 Power for Part B Primary Immunogenicity Hypothesis of Protocol V Antigen Parameter Expected rates/sd Non-inferiority Margin N evaluable (1) Primary Hypothesis - Non-Inferiority of 9-valent HPV L1 VLP Vaccine to GARDASIL Power HPV 6 GMT σ = fold decrease 4708 >0.99 HPV 11 GMT σ = fold decrease 4708 >0.99 HPV 16 GMT σ = fold decrease 4542 >0.99 HPV 18 GMT σ = fold decrease 4985 >0.99 Overall Power for the primary immunogenicity hypotheses >0.99 This study also has >99% power for the secondary immunogenicity hypothesis of noninferiority with respect to seroconversion percentages for HPV types 6, 11, 16, and 18. In addition, this study has >99% power for the secondary immunogenicity hypothesis of acceptability with respect to seroconversion percentages for HPV types 31, 33, 45, 52, and 58. For the secondary seroconversion hypothesis, it is assumed that for each of HPV 31, 33, 45, 52 and 58, no more than 23%, 22%, 16%, 22%, and 20%, respectively, are initially seropositive or PCR positive for any given type. All statistical tests will be conducted at the α=0.025 level (1-sided). The expected response rates and power for the secondary immunogenicity hypotheses are shown in Table 9. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

273 V503 Statistical Analysis Plan 62 Table 9 Power for Part B Secondary Immunogenicity Hypothesis of Protocol V Antigen Parameter Expected rates/sd Non-inferiority Margin N evaluable (1) Secondary Immunogenicity Hypothesis - Non-Inferiority of 9-valent HPV L1 VLP Vaccine to GARDASIL Power HPV 6 Titer Seropositive Cutoff 98% 5% 4708 >0.99 HPV 11 Titer Seropositive Cutoff 98% 5% 4708 >0.99 HPV 16 Titer Seropositive Cutoff 98% 5% 4542 >0.99 HPV 18 Titer Seropositive Cutoff 98% 5% 4985 >0.99 Overall Power for the seroconversion hypothesis >0.99 (2) Secondary Hypothesis - Acceptability of 9-valent HPV L1 VLP Vaccine HPV 31 Titer Seropositive Cutoff 98% Lower bound>90% 4456 >0.99 HPV 33 Titer Seropositive Cutoff 98% Lower bound>90% 4513 >0.99 HPV 45 Titer Seropositive Cutoff 98% Lower bound>90% 4861 >0.99 HPV 52 Titer Seropositive Cutoff 98% Lower bound>90% 4513 >0.99 HPV 58 Titer Seropositive Cutoff 98% Lower bound>90% 4629 >0.99 Overall Power for the secondary immunogenicity hypothesis >0.99 Measured by Competitive Luminex immunoassay (clia). Seropositivity cutoff values for HPV 6, 11, 16, 18, 31, 33, 45, 52, 58 will be determined Sample Size Considerations for Safety Analyses The probability of observing at least 1 SAE in this study depends on the number of subjects enrolled and the incidence rate of SAEs in the general population. If the incidence rate of an SAE is 1 of every 1240 (0.08%) subjects who received any formulation of 9-valent HPV L1 VLP vaccine in Part A, then there is a 53% chance of observing at least 1 such SAE among the 930 subjects who receive any of the 3 formulations of 9-valent HPV L1 VLP vaccine in Part A. If the incidence rate of an SAE is 1 of every 2880 (0.03%) subjects who received a formulation of 9-valent HPV L1 VLP vaccine, then there is a 28% chance of observing at least 1 such SAE among the 930 subjects who receive any of the 3 formulations of 9-valent HPV L1 VLP vaccine in Part A. If no SAEs are observed in the 930 subjects who received a 9-valent HPV L1 VLP vaccine formulation in Part A, then this study will provide 95% confidence that the true incidence rate for SAEs is <0.4%. In Part B, there is a 99.6% chance of observing at least 1 SAE among the 7000 subjects who will receive the selected 9-valent HPV L1 VLP vaccine formulation if the incidence rate of an SAE is 1 of every 1280 subjects (0.08%). If the incidence rate is 1 of every BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

274 V503 Statistical Analysis Plan recipients (0.03%), then there is a 87.8% chance of observing at least 1 SAE among the subjects who will receive the selected 9-valent HPV L1 VLP vaccine formulation. If no SAEs are observed in 7000 subjects, this study will provide 95% confidence that the true incidence rate for SAEs is <0.05%. 7.2 Method of Assigning Study Participants to Treatment Groups One allocation schedule for Part A and 3 allocation schedules for Part B were generated by the study SPONSOR under in-house blinding procedures using block randomization. The allocation schedule for Part A randomizes subjects in a 1:1:1:1 ratio to receive GARDASIL or one of the 3 dose formulations of 9-valent HPV L1 VLP vaccine. Each of the allocation schedules for Part B randomizes subjects in a 1:1 ratio to receive one of the 9-valent HPV L1 VLP vaccine formulations or GARDASIL. Complete randomization blocks are allocated to sites, in order to aim for balanced allocation among the vaccination groups within a site. An Interactive Voice Response System (IVRS) is used to allocate subjects to vaccination groups separately in Part A and Part B. Once a dose formulation has been selected for use in Part B, only the relevant allocation schedule corresponding to the selected 9-valent HPV L1 VLP vaccine formulation will be loaded into the IVRS for the purpose of enrolling the additional 13,380 subjects into Part B of the study. 7.3 Blinding/Unblinding The study subjects, investigators, site personnel, the MRL laboratory staff conducting HPV typing of clinical samples by serology and PCR, the Central laboratory staff who read and issue diagnoses of Pap tests, and the Pathology Panel are blinded to the vaccination group assignments of the subjects for the duration of the study. The clinical materials GARDASIL and the 3 doses of 9-valent HPV L1 VLP vaccine are visually indistinguishable, and are supplied to study sites in identical vials labeled with a doublepanel, blinded label. The SPONSOR is operating under in-house blinding procedures. The SPONSOR s clinical, regulatory, data management, and statistical staff involved with V are blinded to the treatment group assignments of the subjects until the time when the number of endpoint cases required for the primary efficacy hypothesis has been accrued. This study has a data and safety monitoring board (DSMB). An unblinded statistician at the SPONSOR who is otherwise unrelated to this protocol will serve as a non-voting member of the DSMB and will be responsible for providing summaries of safety results to the rest of the DSMB. The unblinded statistician will also be responsible for generating grouped summaries of immunogenicity data for the Week 4 Postdose 2 (Month 3) interim analysis for distribution to a senior management committee at the SPONSOR who is not directly involved with study conduct for the purpose of selecting a dose for the safety, immunogenicity, and efficacy evaluations in Part B. Part A The interim analysis of immunogenicity will be conducted when ~100% of all serology data are available through Week 4 Postdose 2 (Month 3) for subjects in Part A. At that time, the database will be made available only to the unblinded statistician in order to BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

275 V503 Statistical Analysis Plan 64 carry out the immunogenicity analysis. All available data related to the immunogenicity summaries will be screened and cleaned in a blinded manner before the database is made available to the unblinded statistician. Group-level summaries of immune responses (GMTs, seroconversion rates) will be provided to a senior management committee at the SPONSOR. No serology or randomization data for individual study participants will be disclosed. Unblinded summaries of all available tolerability data in Part A will be provided to the DSMB for review. No efficacy results will be summarized. The senior management committee with the assistance provided by the DSMB will select the dose using the process outlined in Section 5. After a dose is selected, the Merck study team will notify the Interactive Voice Response System (IVRS) of the selected 9-valent HPV L1 VLP vaccine formulation. For subjects enrolled in Part A, as each subject completes their Month 7 visit and all study data has been collected, study sites will call IVRS to receive notification of each subject's status. Part A subjects who received the formulations that were not chosen for further evaluation in Part B will be discontinued from the study after the Month 7 visit. For the subjects in Part A who receive the selected 9-valent HPV L1 VLP vaccine dose or the comparator GARDASIL, the total duration of follow-up will be approximately 42 months from enrollment. The integrity of the blinding will be maintained in Part B because no vaccination group information for the continuing subjects will be released to the study sites, and the subjects continuing beyond Month 7 will remain randomized between the two continuing vaccination groups. The formal statistical test of the Part A hypothesis based upon screened and cleaned Month 7 data will be conducted at the time the database is unblinded for the primary efficacy analysis, results of which will be provided in the final study report. Subjects who were enrolled in a 9-valent HPV L1 VLP vaccine formulation with a lower total VLP concentration than the formulation that was selected for use in Part B may be offered a single dose of GARDASIL. Although identifying subjects who will be offered a single dose of GARDASIL could potentially unblind the clinical team to their vaccination group assignment, these subjects will be discontinued after the Month 7 visit in Part A and thus will not participate in the planned efficacy component of the study. Thus, the integrity of the blind will be preserved for the subjects in the selected 9-valent HPV L1 VLP vaccine formulation and GARDASIL arms who will proceed to Part B. The unblinded statistician will provide a list of allocation numbers of subjects who may be offered a single dose of GARDASIL in the study to the Clinical Operations group, if applicable. Part B Part B employs a fixed event design. Although the study will continue for 42 months, the study conclusions regarding vaccine efficacy with respect to the primary endpoint will be based on the analyses conducted after approximately 50% of the subjects for PART B have had at least 24 months of follow-up and when at least 30 cases of the primary efficacy endpoint (HPV 31/33/45/52/58-related high-grade cervical abnormalities (CIN 2/3), Adenocarcinoma In Situ (AIS), invasive cervical carcinoma, high-grade Vulvar Intraepithelial Neoplasia (VIN 2/3), high-grade Vaginal Intraepithelial Neoplasia (VaIN BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

276 V503 Statistical Analysis Plan 65 2/3), vulvar cancer, or vaginal cancer) have been observed. Inference with respect to the primary efficacy hypothesis will be based on this analysis. The counting of efficacy cases will be done by the same unblinded statistician at the SPONSOR who will provide summaries of safety to the DSMB. If the primary efficacy analysis is conducted prior to the end of the study (i.e. before the Month 42 follow-up visit of the last subject enrolled), the remainder of study follow-up still to be accrued will be considered an extension, the purpose of which is to collect additional information in order to assess vaccine efficacy over the entire follow-up period. The primary immunogenicity analysis will be conducted at the time the database is unblinded for the primary efficacy analysis. The official clinical database for the primary efficacy analysis and primary immunogenicity analysis in part B will be unblinded only after medical/scientific review has been completed, and the database has been declared complete. After the primary efficacy analysis and the primary immunogenicity analysis are performed, data cleaning, data screening, and medical/scientific review of the database will continue unblinded until follow-up is complete. However, investigators, laboratory staff (including SPONSOR s laboratory staff), the members of the HPV Vaccine Program Pathology Panel, and study subjects will remain blinded until the conclusion of the study. All investigators and technicians who are responsible for the ascertainment and confirmation of efficacy endpoints will remain blinded for the duration of the study. 7.4 Statistical Methods Vaccine Efficacy of the 9-Valent HPV L1 VLP Vaccine Relative to GARDASIL To address the primary efficacy hypothesis, a one-sided test of the null hypothesis that the vaccine efficacy is 25% will be conducted. The alternative hypothesis states that the vaccine efficacy is >25%. The primary efficacy hypothesis will be tested at the α=0.025 (1-sided) level. Vaccine efficacy is defined as: VE = 100%*{1-(r N /r G )} where r N, the incidence rate among 9-valent HPV L1 VLP vaccine recipients, is defined as r N = C N /τ N. C N = number of primary efficacy cases among 9-valent HPV L1 VLP vaccine recipients and τ N = total person-years of follow-up among 9-valent HPV L1 VLP vaccine recipients. Similarly, r G = C G /τ G is the incidence rate among GARDASIL recipients, where C G is the number of primary efficacy cases among GARDASIL recipients and τ G is the total person-years of follow-up among GARDASIL recipients. The null hypothesis that vaccine is not efficacious (i.e., VE 25%) will be tested by constructing a two-sided exact confidence interval for VE. The statistical criterion for success with respect to the primary efficacy hypothesis will be met if the lower bound of the confidence interval for vaccine efficacy excludes 25%. Under the assumption that r N and r G are the means of independent Poisson processes, and given that there are a total of n = C N + C G primary efficacy cases observed on all subjects, the number of primary efficacy cases C N among 9-valent HPV L1 VLP vaccine recipients is distributed as Binomial(n,p), where the binomial probability p is defined as p= BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

277 V503 Statistical Analysis Plan 66 τ N r N /(τ N r N +τ G r G ). The probability p is a person-years-adjusted estimate of the probability that a given primary efficacy case is a 9-valent HPV L1 VLP vaccine recipient. The lower bound of the 100*(1 α)% exact confidence interval for the probability p is obtained by searching for the proportion p L such that the probability of observing C N or more primary efficacy cases out of n total primary efficacy cases is α/2. Similarly, the upper bound of the 100*(1 α)% exact confidence interval for the probability p is obtained by searching for the proportion p U such that the probability of observing C N or fewer primary efficacy cases out of n total primary efficacy cases is α/2 [2]. The upper and lower bounds of the 100*(1 α)% confidence interval for the vaccine efficacy can be computed from the upper and lower bounds of the confidence interval for p. To address the secondary efficacy hypothesis concerning the combined incidence of HPV 31/33/45/52/58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 6 month or longer apart and exploratory efficacy hypothesis concerning the combined incidence of HPV 31/33/45/52/58-related persistent infection detected in samples from two or more consecutive visits (±1 month visit windows) 12 months or longer apart, a one-sided test of the null hypothesis that the vaccine efficacy is 0% will be conducted using the same methodology as for the primary efficacy hypothesis. For every composite efficacy endpoint, an estimate of vaccine efficacy with 95% confidence intervals will be provided for each HPV type and/or by lesion type. No statistical testing will be conducted for the components Evaluation of Efficacy of the 9-Valent HPV L1 VLP Vaccine Relative to Historic Placebo Given the expected high efficacy of both the 9-valent HPV L1 VLP vaccine and GARDASIL against the original HPV types (6, 11, 16, and 18), a small number of cases may be observed in both groups, making it difficult to assess vaccine efficacy of the 9-valent HPV L1 VLP vaccine relative to GARDASIL. Therefore, additional analysis to estimate vaccine efficacy of the 9-valent HPV L1 VLP vaccine relative to placebo is needed. Because there is no placebo group in the current study, a historic cohort of subjects in Protocols 007 and 012 from the Phase II/III placebo-controlled efficacy studies for GARDASIL will be used. Specifically, the historic cohort will consist of subjects who meet the same eligibility criteria for inclusion in the Per-Protocol Efficacy (PPE) population for the relevant HPV type as in the current study. To provide an estimate of the vaccine efficacy of 9-valent HPV L1 VLP vaccine relative to placebo, an indirect method will be used [3]: VE N/P = 100%* (1-RR N/P ) = 100% {1-(r N /r G * r G /r P )} where r N /r G = 1-VE[current study], is the observed relative risk of 9-valent HPV L1 VLP vaccine to GARDASIL from the current study, and r G /r P = 1-VE[historic cohort] is the observed relative risk of GARDASIL to placebo from the historic cohort. A 95% confidence interval for VE N/P will be estimated from the 95% confidence interval for the BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

278 V503 Statistical Analysis Plan 67 natural log of the relative risk RR N/P. For the confidence interval calculation, the following estimate of the variance of RR N/P will be used: Var(Ln(RR N/P ))= Var(Ln(r N /r G )) + Var(Ln(r G /r P )) This approach will be used to address the objectives evaluating the impact of the 9-valent HPV L1 VLP vaccine on the incidence of a) Pap test abnormalities(s6); b) HPV 16/18- related persistent infection or disease(e2); c) HPV 6/11-related cervical, vulvar, and vaginal disease(e3); d) all CIN, AIS, and cervical cancer due to any HPV type(e4); e) all vulvar and vaginal disease due to any HPV type(e5); and f) cervical biopsy and cervical definitive therapy procedures(e9). No hypothesis testing will be conducted Non-inferiority Assessment of the 9-Valent HPV L1 VLP Vaccine and That of GARDASIL for HPV Types 16/18, Both Relative to a Historical Placebo It is expected that the active control, GARDASIL, and the 9-valent HPV L1 VLP vaccine will be both highly efficacious against the original HPV types. For example, based upon the high efficacy observed in the GARDASIL program, given a total of 10,000 subjects (5,000 per group) with a median follow-up of 30 months post randomization, only 3 cases of type 16/18 related persistent infection or disease are expected in each group, with a non-trivial probability that no case is observed in the GARDASIL group. If no case is observed in the GARDASIL group at the end of study, it will be difficult to draw any conclusion in terms of the relative incidence (9- valent/gardasil ) since the denominator will be zero. Moreover, even if a non-zero number of cases are observed in the GARDASIL group, and a slightly higher number of cases is observed in the 9-valent HPV L1 VLP vaccine group, the relative incidence may appear "large" but not be clinically significant. For this reason, and because of the expected small number of cases, a traditional case-driven study where a relatively large number of cases are needed to test a non-inferiority hypothesis involving a relative incidence is not appropriate; even with a very large number of subjects; the study may never reach its target within a reasonable time frame. Given the importance of protection against HPV 16 and 18 persistent infection and disease (alone these types are responsible for approximately 70% of identified cases of cervical cancer), and to ensure that GARDASIL cancer coverage for HPV 16/18 is maintained in the 9-valent HPV L1 VLP vaccine, we propose a statistical non-inferiority efficacy analysis in which the observed difference in the incidence rate for HPV 16/18 cases are normalized to a standard placebo rate. A margin of 15 percentage points will be used to define the non inferiority of the 9-valent vaccine versus GARDASIL in vaccine efficacy against HPV 16- and/or 18-related persistent infection and disease compared to historical placebo, for which the incidence rate is conservatively assumed to be 4 per 100 person-years as observed from GARDASIL program. The 9-valent vaccine will be concluded to be non-inferior to GARDASIL if the lower bound of the two-sided 95% exact CI for the difference in vaccine efficacy compared to historical placebo (9-valent vaccine GARDASIL remains above -15 percentage points. The justification for this non-inferiority margin is based on broadening, from a public health perspective, the overall protection against cervical cancer as follows: BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

279 V503 Statistical Analysis Plan 68 HPV 16 and 18 are responsible for 70% of cervical cancer cases and GARDASIL efficacy against HPV 16/18 persistent infection and disease versus placebo is ~99% Thus GARDASIL has the potential to prevent 69% (0.99 X 70%) of cervical cancers. Assuming V503 has at least 90-95% efficacy preventing HPV 31, 33, 45, 52, 58 persistent infection and knowing that together these types cause another 17% of cervical cancer worldwide, V503 would then provide 15-16% potential cervical cancer coverage in addition to coverage the vaccine provides against HPV 16/18. Applying a non-inferiority margin of -15% and assuming observed GARDASIL efficacy in the trial is 99%, the lower bound of the 95% confidence interval for the observed V503 vaccine efficacy for type 16/18 would need to be 84% or greater and the minimum projected cervical cancer coverage of the vaccine for these types would be 59% (versus 69% for GARDASIL ). Thus accepting a non-inferiority margin of -15% allows that the minimum overall public health benefit for cervical cancer coverage, 74-75%, provided by V503, (59% from HPV 16/18 and 15-16% from HPV 31, 33, 45, 52, 58) would still exceed that currently provided by GARDASIL Counting of Efficacy Endpoints Tests of the primary and secondary efficacy hypotheses will be conducted on composite efficacy endpoints, and point and interval estimates of VE will be provided for the components that contribute to the composite endpoints. The following procedure will be following in counting efficacy endpoints: A subject who meets the criteria for one or more of the components of a composite endpoint will be counted once as a case of the composite endpoint. When estimating VE relating to individual components, a subject who meets the criteria for the individual component for which VE is being computed will be counted once as a case of that component endpoint. A subject who meets the criteria for more than one component of a composite endpoint will be counted as a case once in each applicable category Computation of Follow-Up Time In the primary per-protocol analyses, follow-up for efficacy endpoints begins after the Month 7 visit. Therefore, each subject s follow-up time will be computed by calculating the time between the subject s Month 7 visit date and that subject s endpoint-specific endpoint case date or censoring/last visit date, whichever comes first. For persistent-infection endpoints, the endpoint case date is the earlier of the two or more visits that define persistent infection. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

280 V503 Statistical Analysis Plan 69 For clinical disease endpoints, the endpoint case date is the date of the first visit at which the disease is detected. For subjects who experience more than one component of a composite endpoint, the endpoint case date for the composite endpoint is the date of the visit at which the first endpoint component is detected. For analyses that are not per-protocol, follow-up of efficacy endpoints begins after the Day 1 visit. Computation of follow-up time is similar to that described above, except that it starts at the date of the Day 1 visit Potential Bias Handling of Efficacy Data Not Collected Per-Protocol Mandatory protocol-specified guidelines are to be used to triage subjects with Pap abnormalities to colposcopy. If a cervical lesion is seen during the colposcopy, it is biopsied. Colposcopies and cervical biopsies may be performed for reasons other than a Pap abnormality at a preceding visit. Cervical disease endpoints that are not a result of a Pap abnormality at a preceding visit will only be included in the analyses of the FAS population and will be excluded from all other efficacy analysis population analyses. Censoring Due to Definitive Therapy When a subject has a cervical biopsy performed, the subject is referred for definitive therapy based on the central laboratory s diagnosis of the biopsy (not the consensus diagnosis of the pathology panel). A subject is typically referred for a central laboratory diagnosis of CIN 2 or worse. Once a subject is referred for cervical definitive therapy, the subject is censored from being considered as a case of a cervical endpoint (based on EEC swabs, or one or more cervical biopsies, an endocervical curettage [ECC], or one or more cervical definitive therapy specimens) in the primary and secondary efficacy analyses. If the subject meets the case criteria for an endpoint based on any tissue samples collected up to the time of cervical definitive therapy, or if the tissue sample collected at the time of definitive therapy qualifies her as an endpoint, she will be counted as a case. In addition, if a subject meets the case criteria based on the results of LVPP swabs, an external genital lesion, a vulvar lesion, or a vaginal lesion that occur after cervical definitive therapy, she will be counted as a case Handling of Missing Data In the analysis of vaccine efficacy with respect to HPV 31/33/45/52/58-related persistent infection and disease endpoints, subjects with missing data will be handled as described in Table Case Imputation Imputation Sensitivity Analyses Robustness of the estimates of VE with respect to the primary efficacy endpoints will be assessed in the PPE population. In particular, robustness of VE will be assessed by imputing efficacy case status for subjects who have not yet become a case of the primary efficacy endpoints and are lost to follow-up prior to the end of the study, and then BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

281 V503 Statistical Analysis Plan 70 recomputing the estimates of VE based on the observed and imputed cases. A 95% CI around the recomputed estimate of VE will be computed using exact methods [2]. Subjects Lost-to-Follow-up In sensitivity analyses of VE with respect to the primary efficacy endpoint, where cases are imputed among subjects lost-to-follow-up (as defined in Table 10), case imputation will be conducted as follows. 1. For both the 9-valent HPV L1 VLP vaccine group and the GARDASIL group, cases will be imputed under the assumption of no vaccine efficacy. 2. The cumulative incidence distribution of HPV 31/33/45/52/58-related CIN2/3, VIN2/3, VaIN2/3, AIS, invasive cervical carcinoma, vulvar cancer, or vaginal cancer in the GARDASIL group, excluding subjects who were lost to follow-up, will be estimated using life-table methods. The life-table will contain 6-months time intervals, representing time elapsed from the start date of case counting (i.e., starting after Month 7 for the PPE analysis). 3. For each of the 9-valent HPV L1 VLP vaccine group and the GARDASIL group, cases will be imputed among subjects lost-to-follow-up as follows: a. Among n 1 subjects lost-to-follow-up during the time interval 0 to 6 months, a total of m 1 = n 1 p 1, cases will be imputed, where p 1 is the life-table derived probability (obtained from step 2 above) of becoming an endpoint case in the GARDASIL group during the 0 to 6 months time interval. b. Among n 2 subjects lost-to-follow-up during the time interval 6 to 12 months, a total of m 2 = n 2 p 2, cases will be imputed, where p 2 is the life-table derived probability (obtained from step 2 above) of becoming an endpoint case in the GARDASIL group during the 6 to 12 months time interval. c. Cases will be imputed in similar fashion among subjects lost-to-follow-up during the succeeding time intervals. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

282 V503 Statistical Analysis Plan 71 Table 10 Disposition of Handling of Missing Data in Efficacy Populations Subject Included/ Excluded in Analysis PPE Subject Eligible to be Case based on Missing Result(s) Subject Included/ Excluded in Analysis Type of Missing Data Subject lost-to-follow-up before becoming an endpoint case and before the end of the study. Included NO Included NO Incomplete data with respect to persistent infection endpoint. Subject not previously classified as an endpoint case, and detected with vaccine HPV Included NO Included NO type-related infection at the subject s last study visit. Missing Day 1 serology result for relevant HPV type. Excluded N/A Excluded N/A Missing 1 or both Day 1 PCR swab results for relevant HPV type. Excluded N/A Excluded N/A Missing Month 7 EEC or LVPP swab PCR results for relevant HPV type. Missing post-month 7 EEC or LVPP swab PCR result for relevant HPV type Missing post-month 7 Thinsection PCR result for relevant HPV type for a biopsy specimen. Missing post-month 7 Pathology Panel diagnosis for a biopsy specimen. HN-TS, All-HN, FAS Subject Eligible to be Case based on Missing Result(s) Excluded N/A Included Only if the other swab is positive. Included Only if the other swab is Included Only if the other swab is positive. positive. Included NO Included NO Included YES - infection endpoints NO - disease endpoints Included YES - infection endpoints NO - disease endpoints Subject is censored at the date of loss to follow-up/last study visit in the analysis of vaccine efficacy. Included/excluded for the relevant HPV type. Not applicable because samples are used to define the analysis population. EEC = endo/ectocervical; HN-TS = HPV-naïve to the relevant type; HPV = Human papillomavirus; LVPP = LVPP = Labial, vulvar, perineal, and perianal; N/A = Not applicable; PCR = Polymerase chain reaction; PPE = Per protocol efficacy. BA736.doc VERSION 2.0 APPROVED--22-Aug-2011 Restricted Confidential Limited Access

283 V503 Statistical Analysis Plan Analyses of Immune Responses Four ANOVA models (1 per HPV type) will be used to test the primary hypothesis in Part A and Part B of noninferiority of anti-hpv 6, 11, 16 and 18 GMTs in subjects receiving the selected 9-valent HPV L1 VLP vaccine compared to subjects receiving GARDASIL. The response variable will be natural log individual titer values and a fixed effect for treatment group will be used. The difference in log titers between the selected 9-valent HPV L1 VLP vaccine group and the GARDASIL group will be estimated from the ANOVA model, and a CI constructed, using the mean square error from the model as the estimate of variance. The point estimate and interval limits will be exponentiated to express the interval on the GMT ratio scale. For the hypothesis test in Part A, a minor multiplicity adjustment for the interim summary will be applied (see Section 7.5). The secondary hypothesis of noninferiority of seroconversion percentages for each of HPV types 6, 11, 16, and 18 will be addressed by 4 one-sided tests of noninferiority (one corresponding to each HPV type) conducted at the α=0.025 level (1-sided). The tests will be conducted based on methods developed by Miettinen and Nurminen [4] for testing the equivalence of 2 proportions. For the secondary hypothesis of acceptability of anti-hpv 31, 33, 45, 52, and 58 seroconversion rates, exact 95% CIs for a binomial proportion will be calculated. 7.5 Multiplicity Part A For the Month 7 hypothesis test of noninferiority of GMTs of the 9-valent HPV L1 VLP vaccine formulation selected for use in Part B compared with GARDASIL, multiplicity with respect to the multiple hypotheses for HPV types 6, 11, 16 and 18 is addressed by requiring success for all 4 HPV types. As described in section 5, there will be an interim summary of immunogenicity in the study when ~100% of all serology data are available through Week 4 Postdose 2 for subjects in Part A. Although there will be no formal hypothesis testing done at the time of the interim analysis and although the interim analysis will be based on postdose 2 data and the final Part A immunogenicity analysis will be based on postdose 3 data, a nominal 1-sided alpha level of will be used to test the hypothesis at Month 7 for each of HPV types 6, 11, 16, and 18 as if an extreme alpha-spending rule had been used for the interim summary, such as the Haybittle-Peto rule [9]. This would theoretically apply a nominal significance level of to any interim summary (even though no hypothesis testing will be conducted). The use of the slightly reduced nominal alpha level at the final analysis in Part A for hypothesis testing will protect the overall study type 1 error at not more than 5% 2-sided. Part B Efficacy Because no efficacy data will be summarized for the interim analysis in Part A, no multiplicity adjustment will be made to account for the subjects from Part A that will be included in the efficacy evaluation for Part B. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

284 V503 Statistical Analysis Plan 73 Immunogenicity The primary immunogenicity analysis in Part B will be conducted independently from the analysis in Part A. Thus, no multiplicity adjustment is required. Multiplicity with respect to the multiple hypotheses for HPV types 6, 11, 16 and 18 is addressed by requiring success for all 4 HPV types thereby controlling the overall alpha level. Thus, no multiplicity adjustment is needed for the overall immunogenicity hypotheses. Success in this study will be declared if the primary efficacy hypothesis and the primary immunogenicity hypothesis in Part B are achieved. Thus, no multiplicity adjustment is needed to control the overall type I error rate for the study with respect to the 2 hypotheses. 7.6 Subgroup Analyses Point estimates of vaccine efficacy along with 95% confidence intervals for the primary endpoint of HPV 31/33/45/52/58-related persistent infection of 6 months (±1 month) or clinical disease, and the secondary endpoint of HPV 31/33/45/52/58-related persistent infection of 12 months or clinical disease, will be provided by: geographical region (i.e., Asia-Pacific, Europe, Latin America, and North America) lifetime number of male and/or female sex partners at baseline (grouped as 0 partners, 1 to 2 partners, and 3 partners) Pap status at baseline (grouped as Negative for SIL and ASC-US or worse). 8. Ground Rules and Data Handling Conventions 8.1 Baseline Definitions or Conventions Baseline serology status for the 9 vaccine HPV types will be determined by the clia results from the Day 1 serum sample. As detailed in Table 1, subjects will provide 2 swabs at each anogenital examination, for HPV DNA detection by PCR assay. For the per-protocol populations, swabs collected at both Day 1 and Month 7 are used to determine baseline PCR status for the vaccine HPV types. For the other HPV naïve populations, only the Day 1 swabs are used for this purpose. With respect to baseline PCR status, a single positive swab is sufficient to define a subject as PCR positive at a given visit. Otherwise, to be confirmed as PCR negative, both swabs must be negative. If less than 2 swab results are available for a subject at a given visit without a positive result, the PCR status for that visit will be unknown. 8.2 Time Points, Day Ranges, and Phasing of Study Periods Day Ranges of Scheduled Injection of Study Vaccines Table 11 presents the protocol specified visit windows guiding when injections are to be given, and the windows that will be used for inclusion of data for Part A and Part B statistical analysis purposes (in days relative to the date of the injection of dose 1 of 9- valent HPV L1 VLP vaccine or GARDASIL ). BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

285 V503 Statistical Analysis Plan 74 Table 11 Acceptable Day Ranges for Vaccination Visits Dose of 9-Valent HPV L1 VLP Vaccine or GARDASIL Scheduled for Injection Day Range for Inclusion in Statistical Analysis (Relative to Day 1 ) Protocol Specified Visit Window Dose 1 Day 1 0 Dose 2 Month 2 ± 3 weeks 36 to 84 Dose 3 Month 6 ± 4 weeks 148 to 218 Day 1 refers to the date when dose 1 of 9-valent HPV L1 VLP vaccine or GARDASIL is injected. For post-day 1 vaccinations, the day ranges for inclusion in the statistical analysis are wider than the protocol specified visit windows primarily to account for differences at the sites in counting months (e.g., 1 calendar month versus 30 days versus 4 weeks). Acceptable Day Ranges for Collection of Swab and Serum Samples Table 12 presents the acceptable day ranges for collection of swab and serum samples for PCR and anti-hpv testing, respectively. Table 12 Acceptable Day Ranges for Collection of Swab and Serum Samples Study Visit Day 1 Sample Type Target Collection Day (Relative to Day 1 ) Day Range for Inclusion in Statistical Analysis (Relative to Day 1 ) Swabs 0 14 to 10 Serum 0 14 to 0 Month 3 Serum 30 days post dose 2 21 to 49 post dose 2 Month 7 Swabs 30 days post dose 3 14 to 72 post dose 3 Serum 30 days post dose 3 21 to 49 post dose 3 Month 12 Serum to 547 Month 24 Serum to 914 Month 36 Serum to 1189 Month 42 Serum to 1464 Day 1 refers to the date when dose 1 of 9-valent HPV L1 VLP vaccine or GARDASIL is injected. For Month 7, indicated target collection/day range is relative to date of injection of dose 3 of 9-valent HPV L1 VLP vaccine or GARDASIL. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

286 V503 Statistical Analysis Plan 75 The rationale for differences in day ranges between swab and serum samples is as follows. At Day 1, vaccination could impact serum titer values/seroconversion very quickly, so requiring the Day 1 serology measurement to be obtained no later than Day 1 is the only way to ensure a pre-vaccination measurement. PCR is not affected by vaccination, so a more liberal window is acceptable. With respect to Month 7, the serum sample has narrow window because we are targeting a response close to the expected peak immune response at nominal Month 7. The Month 7 swab window is less critical within this study and is only used to define the HPV-naïve population. Multiple Serum Samples Collected within Acceptable Day Range If there are multiple values within the day ranges for Day 1 samples, the value from the sample collected closest to the Day 1 vaccination will be used. In the event that a given subject has multiple serum samples collected within the acceptable day range for post Day 1 serology assessment, and thus has more than one data point available for immunogenicity analysis, the results from the serum sample collected closest, in absolute number of days, to the target collection day will be used. Phasing of Study Periods All data collected in Part A and Part B of the study will be allocated to a phase. The study phases, and periods within each phase, are as follows in Table 13, with the definition of the beginning of each period. Table 13 Study Phase Definitions Phase Period Start date Pre-vaccination Pre-vaccination All data before Day 1 Vaccination Vaccination 1 Date of vaccination 1 (Day 1) Vaccination Vaccination 2 Date of vaccination 2 Vaccination Vaccination 3 Date of vaccination 3 Follow-Up Month 07 Follow-Up Date of Month 7 visit 8.3 Description of Data Handling Procedures Prior to Unblinding Prior to providing the database to the unblinded statistician for the interim Week 4 Postdose 2 (Month 3) immunogenicity analysis, all data related to the interim analysis will be reviewed, screened and cleaned following the SPONSOR S GDP for Data Management. The database will not be frozen for the Week 4 Postdose 2 immunogenicity analysis. The database will be frozen for the primary Part A Month 7 immunogenicity analysis, which will also be conducted by the Merck unblinded statistician. All critical database fields will be identified, and the data will be cleaned and screened in a blinded manner prior to database lock. After the data through Month 7 are declared clean and complete, BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

287 V503 Statistical Analysis Plan 76 the database will be audited, and a copy of the database will be frozen for use by the unblinded statistician. The official clinical database for the primary Part B efficacy and immunogenicity analyses will be unblinded only after medical/scientific review has been completed and the database has been declared complete. The data will be reviewed and errors corrected, and identification of how subjects are handled in the PPE, PPI, and ASaT analysis populations will be made prior to unblinding the clinical database for efficacy, immunogenicity, and safety analyses. The clinical database will be unblinded as described in Section 7.3. The in-house unblinded database(s) will be frozen, and all analyses pertaining to regulatory submissions and responses to regulatory queries that involve V will be performed on the relevant frozen clinical database, as described in Section Description of Protocol Violations In general, the per-protocol analyses of efficacy and immunogenicity will not exclude subjects due to protocol violations with the following exceptions noted below: 1. Violations With Respect to the Vaccination Regimen or Sample Collection Regimen: Subjects who do not meet the PPE and PPI population inclusion criteria relating to receipt of vaccinations within acceptable day ranges will be excluded from the relevant analyses as detailed in Section 3.2; Subjects who do not meet the PPE and PPI population inclusion criteria relating to collection of serum and swab samples within acceptable day ranges will be excluded from the relevant analyses as detailed in Section Violations With Respect to Inclusion/Exclusion Criteria: Subjects who have engaged in sexual intercourse within 48 hours prior to a scheduled visit post-month 7 that includes a pelvic examination will have that visit treated as missing in the evaluation of persistent infection; Analyses of Safety No subject will be excluded from safety analyses due to a protocol violation. For safety analyses purposes, subjects will be assigned to the vaccination group corresponding to the clinical material actually received ( as-treated ). As stated in Section 4.1, a safety profile listing will be created for cross-treated subjects separate from the safety reports that will be provided for the protocol-defined vaccination groups. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

288 V503 Statistical Analysis Plan Datasets/Programs/Variables All statistical programs will be developed and validated using good programming practice guidelines. Standard names and formats for all efficacy, immunogenicity, and safety variables will be used. Key statistical analysis datasets will be electronically provided to the FDA. Key statistical programs may also be provided. Neither datasets nor programs will be provided to other regulatory agencies except upon request. A document will be written to accompany the electronic submissions, which will cover the following topics: Directory structure to be used to store files for the electronic submissions; Organization of data sets and (if provided) statistical programs; Process of running the analysis programs (if provided); Naming conventions for the output files and analysis results. BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011

289 V503 Statistical Analysis Plan References 1. Procedure for the Handling and Adjudication of Biopsy Specimens in the Merck HPV Program. Merck data on file. 2. Chan ISF, Bohidar NR. Exact power and sample size for vaccine efficacy studies. Commun Stat Theory Meth 1998;27: Hasselblad V, Kong DF. Statistical methods for comparison to placebo in activecontrolled clinical trials. Drug Inf J 2001;35(2): Miettinen O, Nurminen M. Comparative analysis of two rates. Stat Med 1985;4: DSMB Charter: 9-valent HPV L1 VLP (V503) Vaccine Program, Merck data on file. 6. sidmc Charter: HPV Next Generation Vaccine Program, Merck data on file. 7. MRL Statistical Report: Cross-Protection of Quadrivalent HPV L1 VLP Vaccine, 28- Mar Harper DM, Franco EL, Wheeler CM, Moscicki A-B, Romanowski B, Roteri-Martins CM, et al. Sustained efficacy up to 4.5 years of a bivalent L1 virus-like particle vaccine against human papillomavirus types 16 and 18: follow-up from a randomized control trial. The Lancet 2006;367: Haybittle JL. Repeated assessment of results in clinical trials of cancer treatment. Brit J Radiol 1971; 44: BA736.doc VERSION 2.0 APPROVED Restricted Confidential Limited Access 22-Aug-2011