Vaccines and Related Biological Products Advisory Committee Meeting. November 14, FDA Briefing Document

Transcription

1 Vaccines and Related Biological Products Advisory Committee Meeting November 14, 2012 FDA Briefing Document Influenza A (H5N1) Virus Monovalent Vaccine, Adjuvanted Applicant: ID Biomedical Corporation of Quebec (dba GlaxoSmithKline Biologicals)

2 Table of Contents Influenza A (H5N1) Virus Monovalent Vaccine, Adjuvanted (Q-Pan H5N1) 1.0 General Information 1.1 Influenza A (H5N1) Virus Monovalent Vaccine, Adjuvanted 1.2 GSK s Influenza Virus Vaccines 2.0 Executive Summary 3.0 Introduction and Background 3.1 Influenza Pandemics 3.2 Currently Available Vaccines 3.3 February 2012 Vaccines and Related Biological Products Advisory Committee (VRBPAC) Meeting on Licensure of Pandemic Influenza Vaccines: Demonstration of Effectiveness 4.0 Overview of Clinical Studies 5.0 Pivotal Clinical Immunogenicity and Safety Studies Conducted with Q-Pan- H5N1 Vaccine 5.1 Study Q-Pan-H5N Study Design Study Objectives and Endpoints Populations Analyzed Subject Demographics and Disposition Immunogenicity Results Safety Results 5.2 Study Q-Pan-H5N Study Design Study Objectives and Endpoints Demographics and Disposition Immunogenicity Results Safety Results 6.0 Pivotal Effectiveness Study Conducted with Arepanrix TM (Q-Pan H1N1) Vaccine (study identifier: FLU Q-PAN H1N1-AS DB) 6.1 Study: A Test-negative Case-Control Study to Evaluate the Effectiveness of GSK Biologicals Adjuvanted Monovalent Inactivated H1N1 Influenza Vaccine (Arepanrix TM ) in Young Children (6 months to < 10 years of age) Study Design Study Objectives Populations Analyzed Vaccine Effectiveness (VE) Results Conclusions 7.0 Supportive Safety Studies 7.1 Integrated Summaries of Safety (ISS-1 and ISS-2) 8.0 Publications on Arepanrix (Q-Pan H1N1) Vaccine Effectiveness Studies - 2 -

3 9.0 Post-marketing Experience: AS03-adjuvanted H1N1 Pandemic Influenza Virus Vaccines (Pandemrix H1N1 and Arepanrix H1N1) 10.0 Proposed Pharmacovigilance Plan 10.1 Passive Surveillance Prior to an H5N1 Pandemic During an H5N1 Pandemic 10.2 Active Surveillance: Pandemic Cohort Objectives Study Design Analysis Status 10.3 U.S Surveillance 10.4 Pregnancy Registry 11.0 Post-marketing Studies to Verify and Describe the Clinical Benefit of the Q- Pan H5N1 Vaccine 12.0 Focus of Questions to the Committee 13.0 References - 3 -

4 1.0 General Information 1.1 Influenza A (H5N1) Virus Monovalent Vaccine, Adjuvanted (Q-Pan or Q-Pan H5N1) Product name: Proper name: Trade name: Description: Influenza A (H5N1) Virus Monovalent Vaccine, Adjuvanted (This vaccine is also referred to as Q-Pan, Q-Pan H5N1, Q- Pan H5N1 + AS03 A or Q-Pan AS03 A in this document) None Inactivated, split A/H5N1 influenza virus antigen and oilin-water adjuvant system (AS03). The antigen is manufactured in Quebec according to the FluLaval seasonal influenza vaccine manufacturing process licensed in the U.S. (hence the name Q-Pan for pandemic antigen manufactured in Quebec). The Adjuvant System 03 (AS03) contains two biodegradable oils, squalene and D,Lalpha-tocopherol, mixed with an aqueous phase consisting of phosphate buffered saline. Polysorbate 80 is used as surfactant to stabilize the oil/water interfaces. AS03 is manufactured and filled at GSK s facility in Rixensart, Belgium. Product composition: Applicant: Proposed Indication: Each 0.5 ml dose contains: 3.75 micrograms (µg) hemagglutinin (HA) of the influenza virus strain A/Indonesia/05/ mg squalene mg (D,L)-alpha-tocopherol 4.86 mg polysorbate 80 5 µg thimerosal, a mercury derivative, as a preservative (< 2.5 µg mercury) Each 0.5 ml dose may also contain residual amounts of ovalbumin ( µg), formaldehyde ( 12.5 µg), and sodium deoxycholate ( 3.75 µg) from the antigen manufacturing process. ID Biomedical Corporation of Quebec (dba GlaxoSmithKline Biologicals) Influenza A (H5N1) Virus Monovalent Vaccine, Adjuvanted is a vaccine indicated for active immunization for the prevention of influenza disease in persons 18 years of age and older at increased risk of exposure to the influenza A virus H5N1 subtype contained in the vaccine

5 Dosage Form and Route of Administration: 1.2 GSK s Influenza Virus Vaccines Emulsion for intramuscular injection supplied as 2 separate vials: multi-dose vials of the antigen suspension and the AS03 adjuvant emulsion that must be mixed before use, resulting in ten 0.5 ml doses. Table 1 provides a brief description of GSK s egg-based influenza virus vaccines that are relevant for the discussion and understanding of the clinical development of the Q-Pan H5N1 vaccine. Table 1: Description of GSK s Influenza Virus Vaccines Vaccine Description Haemagglutinin (HA) Adjuvant Influenza A (H5N1) Virus Monovalent Vaccine, Adjuvanted* FluLaval Arepanrix (Arepanrix H1N1) Fluarix Pandemrix GSK s H5N1 AS03- adjuvanted pandemic vaccine, manufactured in Québec, Canada ( Q-Pan H5N1 ) GSK s US-licensed seasonal Influenza vaccine (trivalent, inactivated, split virion), manufactured in Québec, Canada GSK s H1N1 AS03- adjuvanted pandemic vaccine, manufactured in Québec, Canada ( Q-Pan H1N1 ) GSK s US-licensed seasonal Influenza vaccine (trivalent, inactivated, split virion), manufactured in Dresden, Germany GSK s H1N1 AS03- adjuvanted pandemic vaccine, manufactured in Dresden, Germany by the Fluarix process( D-Pan ) per 0.5mL dose 3.75 µg HA of the influenza virus strain A/Indonesia/05/ µg (HA) of each of three influenza viruses from subtypes H1N1, H3N2 and type B 3.75 µg HA of A/California/7/2009 (H1N1)v-like strain (X- 179A) 15 µg (HA) of each of three influenza viruses from subtypes H1N1, H3N2 and type B 3.75 µg HA of A/California/7/2009 (H1N1)v-like strain (X- 179A) AS03 A None AS03 A None AS03 A *Authorized as Pumarix by the EMA AS03A: fulldose AS03:10.69 mg squalene, mg D,L-alpha-tocopherol and 4.86 mg polysorbate 80 per 0.5mL dose 2.0 Executive Summary The United States (U.S.) Department of Health and Human Services (DHHS) contracted GlaxoSmithKline (GSK) Biologicals to develop and submit for licensure a candidate H5N1 influenza virus vaccine with antigen-sparing potential for inclusion in the U.S

6 Strategic National Stockpile. A Biologics License Application (BLA) was submitted by GSK to the Food and Drug Administration (FDA) for an adjuvanted H5N1 influenza virus monovalent vaccine (designated Q-Pan H5N1). The candidate vaccine is dosesparing and includes a proprietary adjuvant, the AS03 Adjuvant System. The H5N1 influenza antigen is produced in eggs using the U.S.-licensed FluLaval manufacturing process. The BLA includes immunogenicity and safety data from two pivotal clinical studies, Q- Pan-H5N1-001 (conducted in adults years of age) and Q-Pan-H5N1-002 (conducted in adults 18 years of age and older), and several supportive studies (conducted in adults). In addition, GSK submitted to the BLA data from an observational casecontrol test negative, retrospective, vaccine effectiveness study using GSK s AS03 adjuvanted monovalent H1N1 influenza virus vaccine as the basis to confirm vaccine effectiveness (See 6.0). After reviewing this observational study, CBER concluded that the data were not adequate to provide a reliable vaccine effectiveness estimate to support traditional approval of Q-Pan H5N1. However, Q-Pan H5N1 fulfilled the immunogenicity criteria to support licensure under accelerated approval, as outlined in the May 2007 Guidance for Industry: Clinical Data Needed to Support the Licensure of Pandemic Influenza Vaccines [1]. Under the accelerated approval regulations (21 CFR ), licensure is based on a surrogate marker that is reasonably likely to predict clinical benefit. For Q-Pan H5N1, the surrogate marker is an antibody response as measured by a hemagglutination-inhibition (HI) assay. Q-Pan H5N1 was associated with increased solicited local and systemic reactogenicity compared to unadjuvanted H5N1 antigen and saline placebo. In a study in which 3342 subjects received Q-Pan H5N1 vaccine and 1139 received saline placebo 0.4% of Q-Pan H5N1 recipients and 0.2% of placebo recipients reported at least one SAE through day 42 post-vaccination. This Vaccines and Related Biological Products Advisory Committee (VRBPAC) meeting is being convened to review and discuss presentations of safety and immunogenicity data derived from studies conducted with Q-Pan H5N1 and submitted in the BLA. The committee will be asked to vote on whether the available data are adequate to support the safety and effectiveness of Q-Pan H5N1 for the proposed indication via the accelerated approval regulations. Committee members will be asked to discuss potential approaches to verify and describe the clinical benefit of Q-Pan H5N1 as required under the Accelerated Approval regulations (21 CFR )

7 3.0 Introduction and Background 3.1 Influenza Pandemics Three influenza pandemics occurred in the twentieth century (1918, 1957, and 1968). Influenza viruses, especially highly pathogenic influenza strains like H5N1, which are endemic in wild birds in many parts of the world and continue to infect people sporadically, remain a pandemic threat. Between 2003 and August 2012, the World Health Organization (WHO) has reported 608 confirmed human cases of H5N1 infection, with a mortality rate of about 60% [2]. In April 2009, a novel influenza A virus, a reassortant of avian, swine, and human influenza viruses, was detected in California. This novel 2009 H1N1 influenza virus was subsequently found to be the same one causing an expanding outbreak of late season respiratory illness in Mexico. On June 11, 2009, the WHO declared a global pandemic, making the 2009 H1N1 influenza virus the first influenza pandemic in over 40 years. In the U.S., five Influenza A H1N1 (2009) Monovalent Virus Vaccines were licensed in the fall of These vaccines were considered strain changes to licensed seasonal influenza vaccines. The strain change approach to licensure of the Influenza A (H1N1) 2009 Monovalent Virus Vaccines was feasible because the seasonal influenza virus vaccine included the influenza A H1N1 subtype and there was clinical experience showing that the dose and dosing regimen for Influenza A subtypes included in the seasonal vaccine were safe and effective. This approach is not applicable to vaccines containing Influenza A subtypes such as H5N1 or adjuvants that are not included in the manufacturer s seasonal influenza vaccine. In these cases, licensure of a prototype pandemic vaccine, such as this Q-Pan H5N1 made with A/Indonesia/05/2005, provides a regulatory mechanism for timely approval of a vaccine made with the actual pandemic strain as a strain change supplement. 3.2 Currently Available Vaccines In April 2007, CBER licensed a monovalent H5N1 vaccine (Influenza Virus Vaccine, H5N1) manufactured by Sanofi Pasteur Inc. for use in persons 18 through 64 years of age who are at increased risk of exposure to the H5N1 influenza virus subtype included in the vaccine. This vaccine is included in the U.S. Strategic National Stockpile. This vaccine was discussed by the VRBPAC in February 2007, and the committee recommended that the available data were sufficient to support safety and effectiveness of the product (refer to transcript: Safety and immunogenicity data supported the dose of antigen (90 µg/1 ml dose) and dosing regimen (2 doses approximately 28 days apart) in persons 18 through 64 years of age. The vaccine is manufactured by the same process as the seasonal influenza vaccine manufactured by Sanofi Pasteur, Inc. (Fluzone), which is approved for use in persons 6 months of age and older. Implicit in the approval of the H5N1 vaccine was that effectiveness data for Fluzone supported that of the H5N1 vaccine

8 3.3 February 2012 Vaccines and Related Biological Products Advisory Committee (VRBPAC) Meeting on Licensure of Pandemic Influenza Vaccines: Demonstration of Effectiveness In February 2012, FDA presented to VRBPAC approaches for potential licensure pathways for adjuvanted and unadjuvanted pandemic influenza virus vaccines. Q-Pan is an adjuvanted pandemic influenza vaccine, therefore, the approaches for licensure pertaining to adjuvanted pandemic influenza vaccines and discussed at the February 2012 VRBPAC are restated here: For an adjuvanted pandemic influenza vaccine against an influenza A subtype not included in the seasonal influenza vaccines, but manufactured according to the same process as a licensed seasonal influenza vaccine (with the exception of the adjuvant), safety and immunogenicity data to support the dose and dosing regimen are required prior to licensure. If the corresponding licensed unadjuvanted seasonal influenza vaccine was approved under the Traditional Approval pathway on the basis of a clinical endpoint efficacy trial(s), then effectiveness of the new subtype (e.g., H5) pandemic vaccine is inferred from the effectiveness of the manufacturer s licensed unadjuvanted seasonal influenza vaccine and traditional approval can be granted for the adjuvanted pandemic influenza vaccine. If the corresponding licensed unadjuvanted seasonal influenza vaccine was approved under the Accelerated Approval Regulations, based on immunogenicity data, with verification of clinical benefit pending, there are two potential pathways to licensure of the adjuvanted pandemic influenza vaccine: (a) The adjuvanted influenza A subtype vaccine (e.g., H5) is approved under the accelerated approval regulations, with a requirement to verify clinical benefit using the licensed unadjuvanted seasonal influenza virus vaccine. (b) Effectiveness of the adjuvanted influenza A subtype vaccine (e.g., H5) is inferred from effectiveness data (such as from a case-control study during the 2009 pandemic) using a non-u.s. licensed adjuvanted pandemic influenza A H1N1 subtype vaccine made by the same manufacturing process to support full (traditional) approval. Regarding potential pathway (a) above, while some committee members acknowledged that our understanding of the immune response to influenza antigens in the presence of adjuvants is incomplete, most members agreed that it is reasonable to infer effectiveness of an adjuvanted pandemic influenza virus vaccine from the efficacy of an unadjuvanted licensed seasonal influenza virus vaccine made by the same manufacturer and process. Regarding potential pathway (b) above, while some committee members acknowledged limitations inherent to observational studies, they also noted the potential relevance of observational data from vaccine effectiveness studies of a non-us licensed adjuvanted pandemic influenza A H1N1 vaccine made by the same manufacturer and process. (Refer to transcript: - 8 -

9 dvaccinesandotherbiologics/vaccinesandrelatedbiologicalproductsadvisorycommittee /UCM pdf). As described in Section 1, Influenza A (H5N1) Virus Monovalent Vaccine, Adjuvanted (Q-Pan H5N1) is an investigational vaccine which consists of inactivated, split virion, A/H5N1 influenza antigen with AS03 adjuvant. The H5N1 influenza antigen (A/Indonesia/05/2005) is produced in eggs using the U.S. licensed FluLaval manufacturing process. However, the corresponding licensed unadjuvanted seasonal influenza virus vaccine, FluLaval, was approved based on immunogenicity data under the accelerated approval regulations, and verification of clinical benefit by an adequate and well-controlled disease endpoint efficacy study is still pending. Thus, in keeping with the approaches discussed at the February 2012, VRBPAC there are two potential pathways to licensure for Q-Pan H5N1 vaccine: a) Q-Pan is approved under the accelerated approval regulations, with a requirement to verify clinical benefit using the licensed unadjuvanted seasonal influenza vaccine. b) Effectiveness of Q-Pan is inferred from effectiveness data (such as from a casecontrol study conducted during the 2009 pandemic) using a non-u.s. licensed adjuvanted pandemic influenza A H1N1 subtype vaccine made by the same manufacturing process to support full (traditional) approval 4.0 Overview of Clinical Studies GSK conducted 2 pivotal clinical trials, Q-Pan-H5N1-001 and Q-Pan-H5N1-002, and submitted immunogenicity, safety and reactogenicity data from the 5,241 subjects in these studies in support of the Q-Pan H5N1 BLA. GSK also submitted H1N1 effectiveness data from an observational test-negative case control study to verify the clinical benefit of the Q-Pan H5N1 vaccine and to obtain traditional approval. These effectiveness data came from a study conducted by the New Brunswick Department of Public Health with GSK s H1N1 vaccine (Arepanrix H1N1 (Van Buynder et al. [3] (See Section 6.0)). Arepanrix H1N1 was manufactured using the same FluLaval process as Q- Pan H5N1 and the study used the pediatric dose, which contains half-dose of AS03 (relative to the Q-Pan H5N1 candidate vaccine adult dose of AS03 (See Table 1)). An overview of these three pivotal studies is presented in Table

10 Table 2: Overview of pivotal clinical studies Study Study Type Study Design Subjects Treatment Groups Total # of Subjects Entered (Completed) Q-Pan-H5N1-001 Pivotal immunogenicity and safety evaluating: Adjuvant effect of two adjuvant doses AS03 A (full strength) and AS03 B (half strength). The equivalence of Q-Pan to the GSK Dresdenmanufactured H5N1 vaccine, D-Pan Randomized, observerblind, parallel group, active control Study duration 6 months Adults years old 2 IM doses, 21-day interval Core groups: A. Q-Pan H5N1: 3.8µg HA alone B. Q-Pan H5N1: 3.8µg HA; AS03A C. Q-Pan H5N1: 3.8µg HA;AS03B D. D-Pan H5N1: 3.8µg HA; AS03A E. D-Pan H5N1: 3.8µg HA;AS03B Contingency groups: F. Q-Pan H5N1 1.9µg HA; AS03A G. Q-Pan H5N1 1.9µg HA;AS03B Total 680 (673) 78 (76) 152 (150) 151 (151) 151 (151) 148 (145) 100 (99) 50 (50) 50 (49) Q-Pan-H5N1-002 VanBuynder et al., 2010 Pivotal immunogenicity, safety and lot-to-lot consistency Pivotal effectiveness study Randomized, placebocontrolled, observer-blind, parallel group Study duration originally 6 months, amended to 12 months. Retrospective cohort, community-based, case-control, testnegative Adults > 18 years old Children 6 months to 9 years old with medically attended ILI for whom pandemic H1N1 influenza testing was sought in New Brunswick, Canada 2 IM doses, 21-day interval Q-Pan H5N1: 3.8µg HA; AS03A Saline placebo 1 or 2 IM doses of Q-Pan H1N1 pandemic 1.9µg HA, AS03B (Arepanrix H1N1 pediatric dose) Total 4561 (4457) 3422 (3343) 1139 (1114) 28 cases 63 controls

11 5.0 Pivotal Clinical Immunogenicity and Safety Studies Conducted with Q-Pan- H5N1 Vaccine 5.1 Study Q-Pan-H5N Study Design Q-Pan-H5N1-001 (Q-Pan-001) is a randomized, observer-blind, multi-center, activecontrolled trial to evaluate the immunogenicity and safety of two doses of Q-Pan H5N1 with AS03 A or AS03 B adjuvant administered on Days 0 and 21 intramuscularly (IM) to healthy adults 18 to 64 years old. The study was designed to evaluate the immunogenicity, safety and reactogenicity of Q-Pan H5N1 with two different adjuvant strengths [AS03 A (full strength) and AS03 B (half strength)] as compared to Q-Pan H5N1 antigen with no adjuvant to determine the effect of adjuvant on both immunogenicity and safety. The study was also intended to provide a comparison of the antigen manufactured in GSK s Quebec facility (Q-Pan) to the antigen manufactured in GSK s Dresden facility (D-Pan). The study was conducted in 7 sites in the U.S. and 3 sites in Canada. A total of 680 subjects were randomized in a 1:2:2:2:2 ratio to 1 of 5 treatment arms: Group A: Q-Pan H5N1 3.8 μg HA, IM on Day 0 and 21 (N 78) or Group B: Q-Pan H5N1 3.8 μg HA + AS03 A, IM on Day 0 and 21 (N 152) or Group C: Q-Pan H5N1 3.8 μg HA + AS03 B, IM on Day 0 and 21 (N 151) or Group D: D-Pan H5N1 3.8 μg HA + AS03 A, IM on Day 0 and 21 (N 151) or Group E: D-Pan H5N1 3.8 μg HA + AS03 B, IM on Day 0 and 21 (N 148) Study Objectives and Endpoints The primary immunogenicity objective was to demonstrate immunologic superiority of Q-Pan H5N1 (3.8 μg HA + AS03 A ) compared to an unadjuvanted formulation of Q-Pan H5N1 (3.8 μg HA) when analyzed for HI antibody seroconversion rates (SCR, defined as: percentage of subjects with either pre-vaccination HI <1:10 and post vaccination HI >1:40 or pre-vaccination >1:10 and post-vaccination 4-fold rise) and HI GMTs at Day 42. The study protocol specified that HI antibody responses to Q-Pan H5N1 (3.8 μg HA) with adjuvant would be considered superior to Q-Pan H5N1 (3.8 μg HA) without adjuvant, if the GMT ratio >2, and the difference in SCR > 15%. The primary safety objective was to describe the safety of Q-Pan H5N1 at the 3.8 μg dose level with AS03 A and AS03 B in terms of solicited local and general reactogenicity events (Days 0 6 post each vaccination), unsolicited adverse events (AEs) (Study Days 0 84), and serious adverse events (SAEs) (Study Days 0-182) in comparison to antigen alone Populations Analyzed For both Study Q-Pan-001 and Study Q-Pan-002 (see below) the following populations were used for the primary analyses: Analysis of immunogenicity was performed on the according to protocol immunogenicity (ATP-I) cohort, which included all evaluable subjects (i.e., those that meet all eligibility criteria, complying with the procedures defined in the protocol, with no elimination criteria during the study) for whom a complete set of data concerning immunogenicity

12 endpoint measures required for the primary endpoints was available (i.e., Day 0 and Day 42 HI titer results). Subjects in this cohort had to have received the correct vaccine. Total Vaccinated Cohort (TVC) was to include all subjects who received at least one dose of vaccine for whom any post-vaccination data were available. The TVC analysis for immunogenicity and safety was to be performed based on the treatment actually received. The primary analysis of safety was to be performed on the TVC Subject Demographics and Disposition The mean age of study subjects was 38.6 years, with a minimum age of 18 years and a maximum age of 64 years. A total of 371 (54.6%) subjects were between the ages of 18 and 40, and the remaining 309 (45.4%) subjects were between the ages of 41 and 65. A total of 393 (57.8%) subjects were female and 287 (42.2%) subjects were male. The majority, (86.8%) of subjects were Caucasian; of the remaining subjects, 5.6% were African American, 4% were unspecified race, 1.3% were Southeast Asian, and all other races were less than 1%. All 680 subjects enrolled and randomized received at least one dose of vaccine and made up the TVC. The majority of subjects (n=665, 97.8%) received 2 doses of vaccine, and 99% (n=673) completed the study through Day 42. A total of 7 subjects withdrew from the study through Day 42 (6 withdrawn consents, 1 migration from the study area). An additional 11 subjects withdrew from the study through Day 182 (9 lost to follow-up and 3 migration from study area). No subjects withdrew from the study due to an AE. The ATP-I cohort consisted of 648 (95%) subjects Immunogenicity Results At Day 42, the HI antibody responses to Q-Pan (3.8 µg HA) combined with either AS03 A or AS03 B were superior to those of unadjuvanted Q-Pan based on pre-specified criteria. The criteria set forth by CBER in the Guidance for Industry [1] for accelerated approval of pandemic vaccines were met (i.e., the lower bound of the 2-sided 95% CI for HI antibody SCR was > 40% and the lower bound of the 2-sided 95% CI for the percent of subjects achieving HI antibody levels >1:40 was > 70%). Also, Q-Pan and D-Pan elicited similar HI antibody GMTs, supporting the chosen 3.8 µg antigen dose. By Day 182, all immune parameters decreased significantly from Day 42, but HI titers in all the adjuvanted Q-Pan and D-Pan groups remained higher than to those in the non-adjuvanted Q-Pan group

13 Table 3: Post-vaccination HI antibody immune responses to Q-Pan and D-Pan H5N1 (A/Indonesia/5/2005) formulated with and without AS03 at Day 42 and Day 182, Study Q-Pan-001 (ATP-I) Treatment Group Day 42 A (Q-Pan unadjuvanted) N n HI Antibody SCR 1 (95% CI) % of Subjects with HI titer > 1:40 (95% CI) GMT (95% CI) (9.6, 27.8) 17.3 (9.6, 27.8) 10.5 (8.2, 13.5) B (Q-AS03A) (93, 99.2) 97.2 (93, 99.2) (383.4, 563.4) C (Q-AS03B) (83.6, 94.1) 89.7 (83.6, 94.1) (246.9, 416.6) D (D-AS03A) (91.9, 98.8) 96.4 (91.9, 98.8) (390.5, 590.7) E (D-AS03B) (86.6, 96.1) 92.3 (86.6, 96.1) (272, 444.5) Day 182 A (Q-Pan unadjuvanted) (0.3, 9.4) 2.7 (0.3, 9.4) 5.6 (5.1, 6.2) B (Q-AS03A) (46, 63) 54.6 (46, 63) 27.8 (22.8, 33.8) C (Q-AS03B) (37.2, 54) 45.5 (37.2, 54) 22.6 (18.4, 27.9) D (D-AS03A) (40, 57.2) 49.3 (40.7, 57.9) 26.1 (20.7, 32.8) E (D-AS03B) (36.5, 53.6) 45.7 (37.2, 54.3) 22.6 (18.3, 28) N = number of subjects with available data n = number of responders 1 percentage of subjects with either pre-vaccination HI <1:10 and post vaccination HI >1:40 or pre-vaccination >1:10 and post-vaccination 4-fold rise > 95% CI LB of 40% at Day 42 was the primary endpoint. Source: BLA , Module , CSR (FLU Q-PAN-001 PRI) Day 42 Tables 22, 23 and 24 and (FLU Q-PAN-001 PRI) Day 182 Annex Tables 9, 10 and 11 Based on these immunogenicity data and the results of a post-hoc, age-stratified analysis showing that subjects > 40 years of age had higher GMTs in response to H5N1 formulated with AS03 A than with AS03 B, GSK decided to move forward with the clinical development of Q-Pan 3.8 μg + AS03 A as the preferred adult pandemic influenza vaccine formulation Safety Results Pain and all general systemic AEs, except temperature (Table 4), were graded on a 4- point scale, with Grade 0 being no AE up to a Grade 3 which prevented normal activity. Redness and swelling were recorded in millimeters with Grade 3 reactions measuring > 100 mm. Table 4: Temperature grading scale Grade 0 Grade 1 Grade 2 Grade 3 Grade 4 < 38 º C > º C > º C > º C > 40 º C (report also as an SAE) Source: BLA , Module , CSR (FLU Q-PAN-001 PRI), Table 8. Solicited AE rates did not differ significantly from Dose 1 to Dose 2; therefore, safety results are presented overall by subject, which includes either Dose 1 or Dose 2 related events

14 Pain was the most commonly reported solicited local symptom in all treatment groups, but it was reported at a higher rate in the Q-Pan H5N1 + AS03 A group (88%) as compared to the unadjuvanted Q-Pan H5N1 vaccine group (23%). Severe pain at the injection site was also reported more frequently in the Q-Pan H5N1 + AS03 A group (6%) as compared to the unadjuvanted Q-Pan H5N1 group (1%). Table 5: Incidence of solicited local reactions overall by subject, Days 0-6, Study Q- Pan-001 (TVC) Local Symptom, n (%) Group A Q-Pan unadjuvanted N=78 Group B Q-AS03A N=152 Group C Q-AS03B N=151 All Pain 18 (23.1) 133 (87.5) 130 (86.1) Gr 3 Pain 1 (1.3) 9 (5.9) 2 (1.3) All Redness, 0 7 (4.6) 2 (1.3) Gr 3 Redness All Swelling, 0 12 (7.9) 10 (6.6) Gr 3 Swelling N = number of subjects with at least one documented dose n = number of subjects reporting AE at least once Gr 3 - Grade 3, severe Source: BLA , Module , CSR (FLU Q-PAN-001 Table 30 The incidence of solicited general reactions is presented in Table 6. Overall, the adjuvanted groups reported general symptoms at a higher rate than the unadjuvanted group

15 Table 6: Incidence of solicited general reactions overall by subject, study Q-Pan- 001 (TVC) General Symptom, n (%) Group A Q-Pan unadjuvanted N=78 Group B Q-AS03A N=152 Group C Q-AS03B N=151 All Fatigue 16 (20.5) 64 (42.1) 50 (33.1) Gr 3 Fatigue 2 (2.6) 6 (3.9) 3 (2.0) All Headache 25 (32.1) 71 (46.7) 61 (40.4) Gr 3 1 (1.3) 10 (6.6) 2 (1.3) Headache All Arthralgias 12 (15.4) 49 (32.2) 36 (23.8) Gr 3 1 (1.3) 7 (4.6) 2 (1.3) Arthralgias All Myalgias 15 (19.2) 74 (48.7) 64 (42.4) Gr 3 Myalgias 1 (1.3) 9 (5.9) 4 (2.6) All Shivering 4 (5.1) 18 (11.8) 21 (13.9) Gr 3 Shivering 0 5 (3.3) 0 All Sweating 6 (7.7) 23 (15.1) 12 (7.9) Gr 3 Sweating 0 3 (2.0) 0 All Temperature > 38 º C 0 4 (2.6) 4 (2.6) Temperature > 39 º C N = number of subjects with at least one documented dose n = number of subjects reporting AE at least once Gr 3 - Grade 3, severe Source: BLA , Module , CSR (FLU Q-PAN-001 Table 31 Unsolicited AEs were reported by a slightly higher proportion of subjects in the Q-Pan H5N1 + AS03 A treatment group (51%) as compared to the unadjuvanted Q-Pan H5N1 group (45%). No adverse event identified by MeDRA preferred term was reported by more than 10% of subjects in a treatment group. Overall, the most commonly reported adverse events were headache, nausea, pharygolaryngeal pain, nasopharyngitis, upper respiratory infection, and back pain, which occurred in both the adjuvanted and unadjuvanted groups. When comparing the unadjuvanted group to the Q-Pan H5N1 + AS03 A adjuvanted group, diarrhea, anemia, lymphadenopathy, dizziness, muscle spasms, and sinusitis occurred exclusively in the adjuvanted group in 2 3% of subjects; and nausea occurred at a higher rate in the adjuvanted group than in the unadjuvanted group (6.6% vs. 3.8% respectively). Rates of unsolicited AEs in the Q-Pan H5N1 + AS03 B adjuvanted group were similar to the Q-Pan H5N1 + AS03 A adjuvanted group, with the exception of anemia, which was reported at a higher rate (2.6% ) in the Q-Pan H5N1 + AS03 A as compared to the AS03 B adjuvanted group. Grade 3 unsolicited AEs were reported by 5% of subjects overall with no clinically relevant differences in reporting rates or AE types between the treatment groups. No subjects died during the study period through Day

16 A total of 15 SAEs were reported in 6 subjects across all study groups including 2 subjects in the Q-Pan H5N1 + AS03 A group (1 subject reporting 2 AEs (cholelithiasis and pancreatitis) and 1 subject reporting chest pain) and none in the unadjuvanted group. None of the SAEs were deemed by the investigator to be vaccine-related. Potential immune-mediated diseases are a subset of AEs that include both autoimmune diseases and other inflammatory and/or neurological disorders which may or have autoimmune etiologies. In this and all studies that were initiated prior to December 2008, potential immune mediated diseases (pimds) were sought by querying the database for a broad range of MedDRA preferred terms that included symptoms as well as diagnoses. Overall, in Study Q-Pan-001 < 3% of subjects reported AEs with pimd preferred terms. These AE reports included common or non-specific ailments such as back pain (n=8), allergies (n=1), asthma (n=2), elevated serum creatinine (n=1), localized allergic reaction (n=1), fire ant sting reaction (n=1), and sensation of generalized hyperesthesia (n=1). None of these events suggested a new or exacerbated autoimmune event. Adjuvanted Q-Pan vaccine resulted in a higher proportion and severity of solicited AEs as compared to unadjuvanted Q-Pan. However, no clinically relevant differences were observed in the proportion or types of unsolicited or serious AEs. 5.2 Study Q-Pan-H5N Study Design Q-Pan H5N1-002 (Q-Pan-002) is a Phase III, observer-blind, saline-placebo controlled, multicenter study designed to evaluate the safety and immunogenicity of a two-dose series of Q-Pan-H5N1 3.8μg + AS03 A administered IM in adults > 18 years of age. This study also evaluated three lots of antigen and three lots of adjuvant for consistency. The study was conducted at 30 sites in the U.S. and 10 sites in Canada. Subjects eligible for the study were males or females 18 years of age or older at the time of vaccination. Subjects 18 to 49 years of age were in good general health as established by pre-enrollment medical history and physical examination. Subjects older than 49 years of age were in stable health, as defined by absence of a serious health event or change to ongoing medication necessitated by therapeutic failure or drug toxicity within a month prior to enrollment. The target enrollment was 4,440. Subjects were stratified and randomized 3:1 by age group to treatment with Q-Pan H5N1 3.8 μg HA + AS03 A or saline placebo as shown in Table

17 Table 7: Q-Pan-002 study groups, by age strata Age in Years 1 Study Arms Total Subject N Lot Consistency N Immunogenicity Cohort N Q-Pan H5N1 + AS03A Placebo Q-Pan H5N1 + AS03A Placebo > 64 Q-Pan H5N1 + AS03A Placebo Subjects in age cohort were further stratified by age years and years. Subjects in age cohort > 64 were further stratified by age years and >75 years. Source: BLA , Module , CSR (FLU Q-PAN-002 PRI) Table 2 Subjects were seen at the clinical site on Screening day and Days 0, 21, 42 and 182. An additional contact for safety evaluation was conducted via telephone on Day 84. The Day 364 safety evaluation contact was either on site or via telephone per the investigator s discretion. Blood samples were collected from a subset of subjects for immunogenicity analyses on Days 0, 42, and 182. Diary cards were provided to collect solicited and unsolicited AEs on Days 0 6 post each vaccination. Unsolicited AEs were collected through Day 84. All SAEs and MAEs were collected through Day 364. Day 364 safety assessments were an amendment to the original protocol and required additional consent from subjects. Overall, 76% of subjects in each treatment group consented to the longer follow-up Study Objectives and Endpoints The primary objectives were: To demonstrate that Q-Pan-H5N1 + AS03 A elicits an immune response measured by post-immunization vaccine-homologous virus hemagglutination inhibition (HI) titers that meets or exceeds the 95% CI lower bounds set forth in CBER s Guidance for Industry [1] for HI antibody SCR and proportions of subjects with reciprocal titers > 40 based on post-immunization reciprocal HI antibody titers. This was to be tested separately for the 2 age strata: 18 to 64 years of age and > 64 years of age. To demonstrate the immunogenic equivalence of 3 consecutive lots of Q-Pan H5N1 vaccine antigen combined with 3 consecutive lots of AS03. To describe the safety of Q-Pan H5N1 + AS03 A in terms of solicited local and systemic reactogenicity events, unsolicited adverse events (AEs), and serious adverse events (SAEs) in comparison to placebo in adult subjects > 18 years of age. The primary safety endpoints were: The occurrence of specifically-solicited local and general signs and symptoms during a 7-day follow-up period (i.e., day of vaccination and 6 subsequent days) after each vaccination, and overall per subject considering both postimmunization periods

18 The occurrence of all unsolicited adverse events during a 21-day follow-up period for each vaccination, as well as overall per subject (Day 0 through Day 84). The occurrence of serious adverse events and medically-attended events Day 0 through Day 182. The primary immunogenicity endpoint was: Vaccine-homologous virus antibody response in subjects receiving 2 doses of study vaccine, as demonstrated by the HI antibody titer at 21 days after the second dose of Q-Pan H5N1 vaccine (Study day 42) for younger adults age 18 to 64 years and older adults age > 64 years. The criteria used to determine whether the antibody responses were sufficient are taken from CBER s Guidance for Industry [1] and are outlined below (Table 8) by age group. Table 8: CBER recommended criteria for supporting accelerated approval of a new pandemic influenza vaccine Age Group HI seroconversion rate (SCR)* years Lower bound of 95% CI > 40% > 64 years Lower bound of 95% CI > 30% Percent subjects with HI antibody titer > 1:40 Lower bound of 95% CI > 70% Lower bound of 95% CI > 60% *SCR: the percentage of subjects with either a pre-vaccination HI titer <1:10 and a post-vacciantion titer >1:40 or a prevaccination titer>1:10 and a minimum four-fold rise in post-vaccination HI antibody titer Demographics and Disposition The mean age of study subjects for the 18 to 64 years age group was 39 years (range 18 to 64 years) and the mean age of subjects in the > 64 years age group was 72 years (range 65 to 91 years). In the 18 to 64 years age stratum for both the Q-Pan and placebo groups, 74% of subjects were between 18 to 49 years of age and 26% of subjects were between 50 to 64 years of age. A total of 2,569 (56%) subjects were female and 1,992 (44%) subjects were male, with no notable difference in gender distribution between age strata and treatment groups. The majority (86% of the 18 to 64 years age group and 94% of the > 64 years age group) of subjects were Caucasian. A total of 4,457 subjects (98%) completed the study through Day 42, and 104 subjects (2%) withdrew from the study prematurely. Withdrawals occurred at similar rates across treatment groups. Through Day 42, a total of 2 subjects in the age stratum [1 Q-Pan H5N1 (fatal MI) and 1 placebo (pneumococcal pneumonia)] and 2 subjects in the > 64 age stratum [both received Q-Pan H5N1 (1 with septic arthritis, pulmonary embolism, spinal abscess and compression and 1 with fatal carcinoma )], withdrew due to SAEs (see section for details on SAEs). A total of 6 subjects in the age stratum and 3 subjects in the > 64 age stratum withdrew because of non-serious adverse events

19 5.2.4 Immunogenicity Results Please refer to Section for definitions of the analysis populations. Lot consistency criteria were met based on the pre-specified analysis in subjects years of age. Therefore, for subsequent immunogenicity analyses data from the three lots were pooled. The primary immunogenicity analysis was performed on the ATP-I cohort. A total of 1,647 subjects in the 18 to 64 years of age stratum were included in the ATP-I cohort (1,571 Q-Pan subjects and 76 placebo subjects). A total of 436 subjects in the > 64 years of age stratum were included in the ATP-I cohort (396 Q-Pan subjects and 40 placebo subjects). Overall, greater than 5% of the subset of subjects randomly selected for immunologic testing were excluded from the ATP-I cohort due to various protocol deviations and, therefore, a confirmatory secondary immunogenicity analysis was also performed on all members of the TVC with immunogenicity data. This secondary immunogenicity analysis resulted in findings similar to the ATP-I analysis. The primary outcomes for immunogenicity were met; Q-Pan H5N1 3.8 μg + AS03 A fulfilled the prespecified immune criteria at Day 42 for both age strata as set forth in CBER guidance [1]. Although the Day 182 immune parameters (a secondary endpoint) declined relative to the Day 42 parameters, the Q-Pan group maintained immune responses at levels well above those in the placebo group (Table 9)

20 Table 9: Post-vaccination HI antibody immune responses according to age at Day 42 and Day 182, Study Q-Pan-002 (ATP-I) Treatment Group N n Day 42 Q-Pan yrs Placebo yrs Q-Pan >64 years Placebo >64 years Day 182 Q-Pan yrs Placebo yrs Q-Pan >64 years Placebo HI Antibody SCR (95% CI) 1 % of Subjects with HI titer > 1:40 (95% CI) GMT (95% CI) (89.3, 92.2) 90.8 (89.3, 92.2) 249 (231.8, 267.5) (0, 7.1) 1.3 (0, 7.1) 5.1 (4.9, 5.4) (69.4, 78.2) 74 (69.4, 78.2) 81.9 (69.7, 96.2) (0.1, 13.2) 2.5 (0.1, 13.2) 5.5 (4.5, 6.8) (56.3, 66.5) 61.5 (56.3, 66.5) 36.2 (31, 42.2) (0.1, 17.8) 2.7 (0.1, (4.8, 6.5) 91 59* 64.8 (54.1, 74.6) 65.9 (55.3, 75.5) 44.8 (33.3, 60.4) 60** (0, 17.6) 0 (0, 17.6) 5.4 (4.6, 6.3) >64 years N Number of subjects with both pre- and post- vaccination results available n number of responders * number of subjects with a 4-fold or greater HI titer rise ** number of subjects with a > 1:40 HI titer 1 percentage of subjects with either pre-vaccination HI <1:10 and post vaccination HI >1:40 or pre-vaccination >1:10 and post-vaccination 4-fold rise > 95% CI LB of 40% at Day 42 was the primary endpoint. Source: BLA , Module , CSR (FLU Q-PAN-002 PRI) Day 42 Tables 24, 25, and Supplement 29; and Day 182 Annex Tables 10, 11 and Safety Results The same reactogenicity grading scale used in Q-Pan-001 was used in Q-Pan-002. AE rates did not differ significantly from Dose 1 to Dose 2; therefore, safety results are presented overall by subject, which includes either Dose 1 or Dose 2 related events. Pain was the most commonly reported solicited local symptom in both the Q-Pan group and the placebo group. However, any grade pain was reported at greater than a 4-fold higher rate in the Q-Pan group (83%) than in the placebo group (20%) (Table 10). At least moderate (Grade 2) and severe pain (Grade 3) were also reported at significantly higher rates in the Q-Pan group, nearly 10-fold and 5-fold higher than the rates reported in the placebo group. Overall, pain lasted a median of three days in the Q-Pan group compared to one day in the placebo group. Similarly, any grade redness and swelling were reported at rates 10-fold and 8-fold higher in the Q-Pan group than in the placebo group. No subjects reported seeking medical attention for any local reaction

21 Table 10: Incidence of solicited local reactions overall by subject, Days 0 6, Study Q-Pan-002 (TVC) Local Symptom, Q-Pan Overall N = 3376 n (%) Saline Placebo Overall N = 1122 n (%) Q-Pan years N = 2267 n (%) Saline Placebo years N = 754 n (%) Q-Pan > 64 years N = 1109 n (%) Saline Placebo > 64 years N = 368 n (%) All Pain 2808 (83.2) 224 (20.0) 2024 (89.3) 171 (22.7) 784 (70.7) 53 (14.4) Gr > 2 Pain 1244 (36.8) 43 (3.8) 1059 (46.7) 33 (4.4) 185 (16.7) 10 (2.7) Gr 3 Pain 156 (4.6) 8 (0.7) 141 (6.2) 6 (0.8) 15 (1.4) 2 (0.5) All Redness, 287 (8.5) 8 (0.7) 181 (8) 7 (0.9) 106 (9.6) 1 (0.3) Gr 3 Redness 4 (0.1) 0 4 (0.2) All Swelling, Gr 3 Swelling 351 (10.4) 8 (0.7) 241 (10.6) 7 (0.9) 110 (9.9) 1 (0.3) 4 (0.1) 0 3 (0.1) 0 1 (0.1) 0 N = number of subjects with at least one documented dose n = number of subjects reporting reaction at least once Gr > 2 at least moderate, Grade 2 Gr 3 - Grade 3, severe Source: BLA , Module , CSR (FLU Q-PAN-002 PRI) Table 42 Myalgias were the most commonly reported solicited general symptom overall, and were reported at a 2-fold higher rate for the Q-Pan group (45%) than the placebo group (21%). The incidence of severe myalgias (Grade 3) was relatively low, with a rate of 3% in the Q-Pan group and a rate of 2% in the placebo group. Table 11 presents the incidence of solicited general AEs including myalgia

22 Table 11: Incidence of solicited general reactions overall by subject, study Q-Pan- 002 (TVC) General Symptom, Q-Pan Overall N = 3375 n (%) Saline Placebo Overall N = 1123 n (%) Q-Pan years N = 2266 n (%) Saline Placebo years N = 755 n (%) Q-Pan > 64 years N = 1109 n (%) Saline Placebo > 64 years N = 368 n (%) All Fatigue 1148 (34) 253 (22.5) 890 (39.3) 189 (25) 258 (23.3) 64 (17.4) Gr 3 Fatigue 107 (3.2) 26 (2.3) 89 (3.9) 21 (2.8) 18 (1.6) 5 (1.4) All Headache 1179 (34.9) 312 (27.8) 932 (41.1) 249 (33) 247 (22.3) 63 (17.1) Gr 3 97 (2.9) 27 (2.4) 89 (3.9) 24 (3.2) 8 (0.7) 3 (0.8) Headache All 853 (25.3) 136 (12.1) 645 (28.5) 97 (12.8) 208 (18.8) 39 (10.6) Arthralgias Gr 3 63 (1.9) 10 (0.9) 55 (2.4) 8 (1.1) 8 (0.7) 2 (0.5) Arthralgias All Myalgias 1526 (45.2) 231 (20.6) 1188 (52.4) 175 (23.2) 338 (30.5) 56 (15.2) Gr (3.2) 21 (1.9) 95 (4.2) 17 (2.3) 14 (1.3) 4 (1.1) Myalgias All Shivering 563 (16.7) 109 (9.7) 456 (20.1) 87 (11.5) 107 (9.6) 22 (6.0) Gr 3 66 (2.0) 12 (1.1) 58 (2.6) 9 (1.2) 8 (0.7) 3 (0.8) Shivering All Sweating 362 (10.7) 82 (7.3) 314 (13.9) 67 (8.9) 48 (4.3) 15 (4.1) Gr 3 28 (0.8) 13 (1.2) 26 (1.1) 11 (1.5) 2 (0.2) 2 (0.5) Sweating All 156 (4.6) 38 (3.4) 121 (5.3) 32 (4.2) 35 (3.2) 6 (1.6) Temperature > 38 ºC Temperature 31 (0.9) 10 (0.9) 28 (1.2) 10 (1.3) 3 (0.3) 0 > 39 ºC Temperature > 40 ºC 4 (0.1) 4 (0.4) 4 (0.2) 4 (0.5) 0 0 N = number of subjects with at least one documented dose n = number of subjects reporting reaction at least once Gr 3 - Grade 3, severe Source: BLA , Module , CSR (FLU Q-PAN-002 PRI) Table 45 Similar to myalgias, joint pains (arthralgias) were reported by twice as many subjects (25% ) in the Q-Pan group as subjects in the placebo group (12%), and was severe in intensity (Grade 3) in 2% and 1% of subjects, respectively. The remaining solicited general symptoms (shivering, sweating, and temperature) were reported by < 20% of subjects overall across both treatment groups. The incidence of severe shivering, sweating, and elevated temperature were low with 2% of subjects in the Q-Pan group reporting severe shivering and all other symptoms reported by < 1% of subjects. Overall, more Q-Pan subjects (23%) took concomitant antipyretic medication during the 7-day solicited AE period as compared to placebo subjects (13%). At least one unsolicited AE was reported by 38% of subjects in the Q-Pan group and 35% in the placebo group. No MedDRA preferred term was reported by more than 4.1% of subjects in either treatment group. The most frequently reported events (in > 2% of subjects) in the Q-Pan group (nasopharyngitis, pharygolaryngeal pain, headache, nausea,

23 diarrhea, cough, upper respiratory infection, nasal congestion) occurred at a similar rate in the placebo group. Events reported in the Q-Pan group, at a rate at least twice that of the placebo group, include injection site pruritus (1.6% vs. 0.4%), injection site warmth (1.3% vs. 0.2%) and dizziness (1.4% vs. 0.7%). Grade 3 unsolicited AEs were reported by 5% of subjects in the Q-Pan group and 6% of subjects in the placebo group. The most commonly reported Grade 3 unsolicited AEs were influenza-like illness, upper respiratory tract infection, nasopharyngitis, and headache. All Grade 3 MedDRA preferred terms were reported at similar rates across treatment groups and all at < 1%. Each treatment group reported seeking medical attention for unsolicited adverse events at similar rates (9% of subjects in the Q-Pan group and 10% of subjects in the placebo group). A total of 11 subjects died during the study through the end of the Day 364 visit, 4 (0.1%) in the Q-Pan group and 7 (0.6%) in the placebo group (see below for list of fatal events). None of the deaths were considered vaccine related by the investigator. Through Day 42, 24 SAEs were reported by 18 subjects [15 (0.4%) in the Q-Pan group and 3 (0.2%) in the placebo group). Overall, 149 subjects [109 (3.2%) Q-Pan subjects and 40 (3.5%) placebo subjects] reported at least one SAE through Day 364 visit. The most commonly reported SAEs for Q-Pan subjects through day 364 were chest pain, pneumonia, myocardial infarction, atrial fibrillation, osteoarthritis, thyroid cancer, cerebrovascular accident, convulsion and pulmonary embolism. The most commonly reported SAEs for placebo subjects were myocardial infarction, coronary artery disease, acute coronary syndrome, coronary artery stenosis, pneumonia and osteoarthritis. Of the most commonly reported SAEs in the Q-Pan group, chest pain (n=5), thyroid cancer (n=3), convulsion (n=3) and pulmonary embolism (n=3) occurred exclusively in Q-Pan subjects. All SAEs were deemed unrelated to vaccine by the investigator. Adverse events of special interest/potentially immune-mediated disorders (AESI/pIMDs) were evaluated by querying the safety database for MedDRA Preferred Terms corresponding to the diagnoses among the following categories: Neuroinflammatory disorders (optic neuritis, multiple sclerosis, demyelinating disease, transverse myelitis, Guillain-Barre syndrome, myasthenia gravis, encephalitis, neuritis, Bell s palsy) Musculoskeletal disorders (systemic lupus erythematosus, cutaneous lupus, Sjogren s syndrome, scleroderma, dermatomyositis, polymyositis, rheumatoid arthritis, juvenile rheumatoid arthritis, polymyalgia rheumatica, reactive arthritis, psoriatic arthropathy, ankylosing spondylitis, spondylarthropathy) Gastrointestinal disorders (Crohn s disease, ulcerative colitis, celiac disease) Metabolic diseases (autoimmune thyroiditis, Grave's or Basedow s disease, Hashimoto thyroiditis, insulin-dependent diabetes mellitus [IDDM], Addison s disease) Skin disorders (psoriasis, vitiligo, Raynaud s phenomenon, erythema nodosum, autoimmune bullous skin diseases) Others (autoimmune hemolytic anemia, idiopathic thrombocytopenic purpura,

24 antiphospholipid syndrome, vasculitis, temporal arteritis, Behcet's syndrome, pernicious anemia, autoimmune hepatitis, primary biliary cirrhosis, primary sclerosing cholangitis, autoimmune glomerulonephritis, autoimmune uveitis, autoimmune cardiomyopathy, sarcoidosis, Stevens-Johnson syndrome). A total of 13 subjects (12 Q-Pan subjects, 1 placebo group) reported 14 AESI/pIMDS (Table 11a)

25 Table 11a: AESI/pIMDs by treatment group Q-Pan-002* Treatment Diagnosis Subject Age, Gender and Significant Past Medical History Q-Pan + AS03A Q-Pan + AS03A Q-Pan + AS03A Q-Pan + AS03A Q-Pan + AS03A Q-Pan + AS03A Q-Pan + AS03A Q-Pan + AS03A Q-Pan + AS03A Q-Pan + AS03A Q-Pan + 4 th Nerve Palsy Lumbar radiculitis 71 y.o. M w/ hypertension 74 y.o F w/ vertebral disc disease 72 y.o. F w/ long standing hip pain PMR 84 y.o. F w/ leg pain. PMR Temporal arteritis RA 44 y.o. F w/ Hashimoto s Crohn s 36 y.o. F w/ abdominal pain and diarrhea Autoimmune hepatitis (AIH) AESI Onset Days Post Last Vaccine Dose Dose Alternate Plausible Cause per GSK Additional Comments 22 2 Y GSK considered hypertension a likely cause? 2 Y 81 2 Y Symptoms unchanged post vaccination. New diagnoses applied 82 2 _ The investigator considered fibromyalgia plausible though patient had new onset back pain 2 days after initial vaccine and new onset neck and shoulder pain and elevated ESR with diagnosis of PMR Y RA is a common secondary autoimmune disease with Hashimoto s Y Symptoms antedated vaccine by 6 months 27 y.o. M 40 2 Y Baseline sera tested after diagnosis made was consistent with positivie serology for AIH. Psoriasis 49 y.o. F 5 2 N Psoriasis 37 y.o. F 31 2 Y Guttate psoriasis 8 days post streptococcal pharyngitis Rheumatoid 72 y.o. F w/? 2 Y Lung rheumatoid arthritis Erythema 35 y.o. F 40 2 N Nodosum Facial Palsy 62 y.o. F 78 2 N AS03A Q-Pan + AS03A Q-Pan + Celiac??? N Only a diagnosis was provided AS03A disease Placebo Psoriasis 68 y.o. M N *Subjects were stratified and randomized 3:1 to treatment with Q-Pan H5N1 3.8ug HA + AS03A or saline placebo PMR: polymyalgia rheumatica; RA: rheumatoid arthritis None of the AESI/pIMD events were considered vaccine related by the investigator and GSK, upon review of the events, considered that most of them had an alternate, plausible cause