Not applicable ENCEPP/SDPP/6658 NCT EU/1/08/497/ EMEA/H/C/ Amgen Europe B.V.

Transcription

1 Date: 04 April 2016 Page 1 of 270 PASS INFORMATION Title Version Identifier of the Final Study Report Date of Last Version of the Study Report EU PAS Register No: Clinicaltrials.gov Identifier No: Active Substance A Cross-sectional Study of Patients With Immune Thrombocytopenic Purpura and Caregivers to Estimate the Proportion Who Administer Romiplostim Correctly After Receipt of Home Administration Training Materials Version 1.0 Not applicable ENCEPP/SDPP/6658 NCT Medicinal Product Nplate Product Reference: Procedure Number: Marketing Authorization Holder(s) Joint PASS Research Question and Objectives Countries of Study Author ATC code B02BX04, romiplostim EU/1/08/497/ EMEA/H/C/ Amgen Europe B.V. No The primary objective of this study is to estimate the proportion of subjects and caregivers who administer romiplostim correctly after being trained with the home administration training (HAT) pack. Austria, Belgium, France, Germany, Greece, The Netherlands, Spain, and The United Kingdom Anouchka Seesaghur Manager, Observational Research Amgen Limited 1 Uxbridge Business Park Sanderson Road Uxbridge UB8 1DH The United Kingdom Tel: +44 (0) Fax: +44 (0)

2 Date: 04 April 2016 Page 2 of 270 Marketing Authorization Holder(s) Marketing Authorization Holder(s) MAH Contact Person Amgen Europe B.V. Minervum ZK Breda The Netherlands Scott Stryker Medical Director, Observational Research Amgen, Inc. One Amgen Center Mail Stop 24-2-A Thousand Oaks, CA United States Tel: +1 (650)

3 Date: 04 April 2016 Page 3 of 270 TABLE OF CONTENTS PASS INFORMATION... TABLE OF CONTENTS 1. ABSTRACT LIST OF ABBREVIATIONS INVESTIGATORS OTHER RESPONSIBLE PARTIES MILESTONES RATIONALE AND BACKGROUND RESEARCH QUESTION AND OBJECTIVES AMENDMENTS AND UPDATES RESEARCH METHODS Study Design Setting Subjects Variables Data Sources and Measurement Bias Study Size Data Transformation Statistical Methods Main Summary Measures Main Statistical Methods Missing Values Sensitivity Analyses Amendments to the Statistical Analysis Plan Quality Control RESULTS Participants Descriptive Data Outcome Data Main Results Primary and Secondary Endpoints Dose Administration Variables Supporting the Primary, Secondary, and Additional Endpoints

4 Date: 04 April 2016 Page 4 of Self-administration Diary Other Analyses Adverse Events/Adverse Reactions DISCUSSION Key Results Limitations Interpretation Generalizability OTHER INFORMATION CONCLUSION REFERENCES SUMMARY TABLES, FIGURES, AND LISTINGS ANNEXES

5 Date: 04 April 2016 Page 5 of 270 List of Tables Table 5 1. Study Milestones Table Summary of Demographics and Baseline Characteristics (Full Analysis Set)... Table Summary of Primary and Secondary Endpoints at the First 4-Week Visit (Full Analysis Set)... Table Instances of Healthcare Professional Intervention to Correct an Error During Administration at the First Standard of-care 4-week Visit (Study ) and Related HAT Pack Training Materials... Table Summary of Subject Participation - Overall - All Screened Subjects... Table Summary of Subject Participation - by Country - All Screened Subjects... Table Summary of Demographic and Baseline Characteristics - Full Analysis Set... Table Primary Endpoint: Subject or Caregiver Administers Romiplostim Correctly at the First Standard of Care 4 Week Visit - Full Analysis Set... Table Secondary Endpoints at the First Standard of Care 4 Week Visit - Full Analysis Set... Table Secondary Endpoint: Summary of the Difference (%) Between the Prescribed and Administered Dose (Micrograms) of Romiplostim at the First Standard of Care 4 Week Visit - Full Analysis Set... Table Additional Endpoints at the First Standard of Care 4 Week Visit - Full Analysis Set... Table Summary of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care 4 Week Visit - Full Analysis Set

6 Date: 04 April 2016 Page 6 of 270 List of Annexes Annex 1. List of Stand-alone Documents... Annex 2. Study Protocol and Amendments... Annex 3. Signature of Coordinating Investigator... Annex 4. Statistical Analysis Plan... Annex 5. HAT Pack Materials

7 Date: 04 April 2016 Page 7 of 270 List of Listings Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set... Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set... Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set... Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set

8 Date: 04 April 2016 Page 8 of ABSTRACT Title A Cross-sectional Study of Patients With Immune Thrombocytopenic Purpura and Caregivers to Estimate the Proportion Who Administer Romiplostim Correctly After Receipt of Home Administration Training Materials Abstract date: 04 April 2016 Name and affiliation of main author: Anouchka Seesaghur (Amgen Limited) Keywords Nplate, immune thrombocytopenic purpura (ITP), Home Administration Training (HAT), non-interventional, self-administration Rationale and Background A variation to the romiplostim (Nplate) Marketing Authorisation and Product Information was approved in the European Union (EU) in December 2012 to allow for the possibility of administering romiplostim by patients or their caregivers. To mitigate risks of medication errors from self-administration, a HAT pack was designed to support the training of physicians and patients. This study was conducted to measure the effectiveness of this additional risk minimization activity. The Nplate EU Summary of Product Characteristics (SmPC), Section 4.2 Method of Administration, provided the following approved instructions for self-administration: Patients who have a stable platelet count 50 x 10 9 /L for at least 4 weeks without dose adjustment may, at the discretion of the supervising physician, self-administer Nplate solution for injection. Patients eligible for self-administration of Nplate should be trained in these procedures. After the first 4 weeks of self-administration, the patient should again be supervised while reconstituting and administering Nplate. Only patients who demonstrate the ability to reconstitute and self-administer Nplate are allowed to continue doing so.

9 Date: 04 April 2016 Page 9 of 270 Research Question and Objectives The primary objective of this study was to estimate the proportion of adult subjects and caregivers who administered romiplostim correctly after being trained with the HAT pack. The secondary objectives of the study were to: estimate the proportion of subjects and caregivers who reconstituted romiplostim correctly; estimate the subjects and caregivers accuracy in administering the prescribed dose of romiplostim; and estimate the proportion of subjects and caregivers who injected romiplostim successfully. Study Design This cross-sectional study involved direct observation by a healthcare professional (HCP) of a series of subjects and caregivers in the act of administering romiplostim at their first standard-of-care visit 4 weeks after training with the HAT pack. Further observations, if requested by the HCP, were also recorded in the study (if made within 16 weeks of enrollment). Additionally, data was collected from the subjects self-administration diary at the first standard-of-care 4-week visit to ensure there were no problems with administration while not at the clinic. Setting The study was conducted at 12 study centers in Austria, Belgium, France, Germany, Greece, The Netherlands, Spain, and The United Kingdom. The recruitment period began on 07 July The last subject last visit was on 20 November 2015 and the database lock was on 20 January Subjects and Study Size, Including Dropouts Subjects were eligible if they met the following criteria: (1) adult ITP patient, treated per EU SmPC, or caregiver new (or at least a 3-month gap) to romiplostim administration; (2) received HAT pack training; (3) available at standard-of-care medical visit 4 weeks (range 2 to 8 weeks) after HAT pack training; and (4) provided informed consent. A total of 41 subjects or caregivers who were provided with HAT pack training were enrolled into the study. One subject failed study eligibility criteria as the first standard-of-care visit occurred 1 week post-hat pack training; therefore, 40 subjects or caregivers were included in the Full Analysis Set. No subjects were lost to follow-up.

10 Date: 04 April 2016 Page 10 of 270 Variables and Data Sources Variables for data collection included subject demographics, expected and observed injection volume per syringe, appropriate use of alcohol wipe at injection site, clinically appropriate handling of syringe to avoid contamination, and clinically appropriate technique of subcutaneous injection. In this non-interventional study the data collected were derived from routine clinical care. HCPs completed a standardized data collection form to record their direct observation of subject or caregiver administration. The primary endpoint in this study was defined as a yes/no indicator for whether the subject or caregiver administered romiplostim correctly at the standard-of-care 4-week visit. The secondary endpoints in this study were defined as: a yes/no indicator for whether the subject or caregiver reconstituted romiplostim correctly at the standard-of-care 4-week visit / at follow-up visits for accuracy in administering the prescribed dose of romiplostim, the difference between the prescribed and self-administered dose of romiplostim, expressed as a % a yes/no indicator for whether the subject or caregiver injected romiplostim successfully at the standard-of-care 4-week visit / at follow-up visits a yes/no indicator for whether the subject or caregiver administered romiplostim correctly at follow-up visits. Results For the primary endpoint, at the first standard-of-care visit 4 weeks after training with the HAT pack, 35 subjects or caregivers (87.5%) administered romiplostim correctly (ie, reconstituted romiplostim correctly, administered the prescribed dose accurately, and successfully injected romiplostim without any HCP intervention). A HCP intervened in 5 instances to correct the administration of romiplostim by 3 subjects and 2 caregivers. Two of these 5 subjects or caregivers had a follow-up visit and did administer romiplostim correctly at that visit. For the secondary endpoints, 1 subject (2.5%) did not reconstitute romiplostim correctly because the subject did not ensure all of the romiplostim was dissolved. All 40 subjects or caregivers injected romiplostim successfully and the difference between the prescribed and administered dose of romiplostim was within the 10% margin error for dose. Subsequent voluntary visits (if requested by the HCP) were recorded for 6 subjects or caregivers (15%) and all administered romiplostim correctly at these visits.

11 Date: 04 April 2016 Page 11 of 270 No adverse drug reactions (ADRs) were reported. Discussion This study was conducted to measure the effectiveness of the HAT pack training in mitigating the risk of medication errors from self-administration. Most subjects or caregivers (87.5%) were able to administer romiplostim correctly without any HCP intervention after HAT pack training. The types of issues observed in this study were associated with addressable factors which can be managed by training with the current HAT pack materials. The physician guide in selecting and treating patients for home administration lists the eligibility criteria for the patient to be an appropriate candidate for self-administration. The patient training materials include specific instructions (paper and video formats) indicating steps where care needs to be taken to avoid the risk of medication error, the importance of checking that romiplostim has been reconstituted correctly, the importance of receiving the correct dose, and the consequences of dosing errors. The results of this study highlight the importance of supervising the patient or caregiver again after the first 4 weeks of self-administration (as required by the EU SmPC and the HAT pack training materials), patient selection by the physician, and the HAT pack training in helping to mitigate the risk of medication errors during self-administration. Marketing Authorization Holder(s) Amgen Europe B.V. Minervum ZK Breda The Netherlands Names and Affiliations of Principal Investigators Martin Schipperus HagaZiekenhuis, Leyenburg Hematology Leyweg 275 Den Haag 2545 CH The Netherlands

12 Date: 04 April 2016 Page 12 of LIST OF ABBREVIATIONS Abbreviation or Term ADR CI EMA EU HAT HCP ITP MAH PASS PT SAP SmPC SOC Definition/Explanation adverse drug reaction(s) confidence interval European Medicines Agency European Union home administration training healthcare professional immune thrombocytopenic purpura Marketing Authorization Holder post-authorization safety study preferred term statistical analysis plan Summary of Product Characteristics system organ class

13 Date: 04 April 2016 Page 13 of INVESTIGATORS The coordinating investigator details are as follows: Martin Schipperus HagaZiekenhuis, Leyenburg Hematology Leyweg 275 Den Haag 2545 CH The Netherlands A list of all collaborating institutions and investigators will be made available upon request.

14 Date: 04 April 2016 Page 14 of OTHER RESPONSIBLE PARTIES The Amgen and Marketing Authorization Holder (MAH) contact person for this study is: Scott Stryker Medical Director, Observational Research Amgen Inc. One Amgen Center Mail Stop 24-2-A Thousand Oaks, CA United States

15 Date: 04 April 2016 Page 15 of MILESTONES The milestones for this study are provided in Table 5-1. Table 5-1. Study Milestones Milestone Planned Date Actual Date Comments Start of data collection August July 2014 Earlier identification of the first subject than originally anticipated End of data collection June November 2015 Study enrolled at a faster pace than originally anticipated Registration in the EU PASS register Registration on clintrials.gov NA 08 July 2014 Date of acceptance NA 11 March 2014 NA Final report of study results December April 2016 Study enrolled at a faster pace than originally anticipated EU=European Union; NA=not applicable; PASS=post-authorization safety study

16 Date: 04 April 2016 Page 16 of RATIONALE AND BACKGROUND Typically, romiplostim is administered by a healthcare professional (HCP). Administration by a HCP is a multistep process at weekly intervals, including: dose calculation based on the patient's weight and platelet count response, reconstitution of drug, measurement of volume to be injected, and subcutaneous injection. Detailed instructions of each step are specified in the European Union Summary of Product Characteristics (EU SmPC) (2016). A variation to the Nplate Marketing Authorization and Product Information was approved in the EU in December 2012 allowing for the possibility of self-administration of romiplostim by patients or their caregivers. While the dose calculation is performed by the prescribing physician, self-administration steps completed by the patient or caregiver include reconstitution of drug, measurement of volume to be injected as per the physician s prescription, and subcutaneous injection. Medication error is an important identified risk in the Risk Management Plan for romiplostim. A total of 62 medication error reports were received by Amgen through 31 January 2012 during an estimated patient-years occurring through commercial distribution of romiplostim (Safety Assessment Report, 2012). Three of the reports had notable clinical consequences. Although none of these medication error reports involved self-administration of romiplostim, an effort to mitigate the risk of medication errors is considered important in the self-administration setting. To mitigate risks of medication errors from self-administration, a home administration training (HAT) pack was designed to support the appropriate selection and training of patients by HCPs and to provide support to patients during self-administration at home. For HCPs, the HAT pack includes a guide and checklist for patient selection and training (Annex 5). The HCP may consider a patient for self-administration who has a stable platelet count 50 x 10 9 /L for at least 4 weeks without dose adjustment. For patients, the HAT pack training materials include a placemat to lay out the administration components, a checklist, instructions for use, a self-administration diary, and an instructional video (Annex 5). Patients are required to undergo a period of training using the materials in the HAT pack and must demonstrate their ability to self-administer to a HCP. In addition, after the first 4 weeks of self-administration patients should again be supervised while reconstituting and administering romiplostim. Only patients who demonstrate the ability to reconstitute and self-administer romiplostim are allowed to continue doing so (EU SmPC, 2016).

17 Date: 04 April 2016 Page 17 of 270 The HAT pack for the patients or caregivers was evaluated in a human factors engineering study (2011). In the study, a total of 30 subjects (29 immune thrombocytopenic purpura [ITP] patients and 1 caregiver) were asked to perform the process of self-administration over a 4-week period utilizing a 100-mcg kit or a 500 mcg-kit. Of the 120 attempts (30 subjects x 4 weeks), the success rate of proper administration was 95%. In the 6 attempts with error, the subjects either misjudged the reference point of the plunger head (eg, assuming the middle or bottom of the plunger head represented the fill level) or counted in the wrong direction on the syringe (counting up from 0.10 milliliter to get to 0.15 milliliter instead of counting down). This observational research study was conducted to measure the effectiveness of the HAT pack as an additional risk minimization activity in the routine clinical care setting.

18 Date: 04 April 2016 Page 18 of RESEARCH QUESTION AND OBJECTIVES This study was descriptive in nature and there was no a priori hypothesis. Data collected in this study provided information to identify possible refinement of the HAT pack and served as an early indicator of the potential for medication errors associated with self-administration. The primary objective was to estimate the proportion of subjects and caregivers who administered romiplostim correctly. The secondary objectives were to: estimate the proportion of subjects and caregivers who reconstituted romiplostim correctly; estimate the subjects and caregivers accuracy in administering the prescribed dose of romiplostim; and estimate the proportion of subjects and caregivers who injected romiplostim successfully.

19 Date: 04 April 2016 Page 19 of AMENDMENTS AND UPDATES None

20 Date: 04 April 2016 Page 20 of RESEARCH METHODS 9.1 Study Design This multicenter, non-interventional, cross-sectional study involved direct observation by a HCP of a series of subjects and caregivers in the act of administering romiplostim at their first standard-of-care visit which occurred 4 weeks after training with the HAT pack. Further observations, if they occurred, were also recorded within 16 weeks of enrollment. Additional observations were voluntary and were not required for study participation; they occurred only if the HCP requested them. Additionally, data were collected from the self-administration diary at the first standard-of-care visit to ensure there were no problems with administration while not at the clinic. The strength of this study design is that it describes the absolute risk of medication error with romiplostim self-administration by an approach that was reliable (direct observation by trained HCP) (National Research Council, 2007) and of little burden to the subject and HCP. The primary endpoint in this study was defined as a yes/no indicator for whether the subject or caregiver administered romiplostim correctly at the standard-of-care 4-week visit. The indicator was "yes" if: the difference between the prescribed and administered dose of romiplostim was within ± 10% (as specified in the protocol [Annex 2]) ; and the HCP observed that the subject or caregiver reconstituted romiplostim correctly; and successfully injected romiplostim as described in the secondary endpoint; and if the HCP did not have to intervene to correct an error during self-administration by subject or caregiver (eg, volume in syringe). The 10% margin of error for dose was consistent with a published definition of medication error for an injectable product (National Research Council, 2007; Barker et al, 2002). The indicator was "no" if any of the criteria above were not met. The secondary endpoints in this study were defined as: a yes/no indicator for whether the subject or caregiver reconstituted romiplostim correctly at the standard-of-care 4-week visit/at follow-up visits. The indicator was yes if the HCP observed that aseptic techniques were used in preparing the vial; all the water from the sterile water syringe was injected gently into the vial; and all romiplostim was dissolved. The indicator was "no" if any of these criteria were not met. for accuracy in administering the prescribed dose of romiplostim, the difference between the prescribed and self-administered dose of romiplostim, expressed as a %.

21 Date: 04 April 2016 Page 21 of 270 a yes/no indicator for whether the subject or caregiver injected romiplostim successfully at the standard-of-care 4-week visit/at follow-up visits. The indicator was yes if the HCP observed that all air bubbles were removed from syringe; the patient made clinically appropriate use of alcohol wipe at injection site; clinically appropriate handling of syringe to avoid contamination; and clinically appropriate technique of subcutaneous injection. The indicator was "no" if any of these criteria were not met. a yes/no indicator for whether the subject or caregiver administered romiplostim correctly at follow-up visits. The indicator was yes or no as described in the primary endpoint. Additional details are provided in Section 9.1 of Annex Setting The study was conducted at 12 study centers in Austria, Belgium, France, Germany, Greece, The Netherlands, Spain, and The United Kingdom. Study centers with clinicians known to treat ITP patients with romiplostim were approached for possible participation in this study. Study centers within each country were eligible for participation after launch of the HAT pack for self-administration at the country level. The availability of the HAT pack was staggered at the country level from March 2013 onwards. Study centers and countries were selected based on availability and willingness to participate; no minimum number of patients treated with romiplostim, use of self-administration, or other criteria were applied. Subject enrollment began on 07 July 2014 and the last subject last visit was on 20 November Subjects received HAT pack training and were observed by a HCP at their first standard-of-care visit which occurred 4 weeks (range 2 to 8 weeks) after training with the HAT pack. Further observations, if they occurred, were also recorded within 16 weeks of enrollment. 9.3 Subjects At each participating site, patients who met the following criteria were enrolled as a study subject: adult ( 18 years of age) ITP patient, treated per EU SmPC, or caregiver new (or at least a 3-month gap) to administering romiplostim received HAT pack training available at first standard-of-care medical visit 4 weeks (range 2 to 8 weeks) after HAT pack training provided informed consent The enrolling HCP asked the patient at start of the HAT pack training of their willingness to participate in this observational study, and obtained informed consent. Consecutive

22 Date: 04 April 2016 Page 22 of 270 eligible subjects at each site were enrolled into the study (ie, non-random sampling from source population). HCPs directly observed the subject or caregiver administering romiplostim at the first standard-of-care 4-week visit after HAT pack training. To obtain the proportion of patient participation, each enrolling site also provided the number of eligible subjects and caregivers who were offered participation in the study but refused. 9.4 Variables Data for the following variables were collected using a standardized data collection form that was filled out by the observing HCP. subject demographics - sex (male or female), age (years) who administered the drug - (subject or caregiver) dates - date of initial prescription for self-administration, date of completion of HAT pack training, dates of observation reason for follow-up visit (if applicable) (text) HCP observing - clinical site, provider type (clinician, nurse, other) HCP observing was same person as trainer (yes/no) number of vials of romiplostim needed for single administration injection volume per syringe prescribed by physician - milliliter vial size of romiplostim - (250 or 500 microgram kit) aseptic techniques were used in preparing the vial (yes/no) gently injected all water from sterile water syringe into the vial (yes/no) ensured all romiplostim dissolved (yes/no) ensured all air bubbles removed from injection syringe (yes/no) injection volume per syringe, observed - milliliter clinically appropriate use of alcohol wipe at injection site - (yes/no) clinically appropriate handling of syringe to avoid contamination - (yes/no) clinically appropriate technique of subcutaneous injection - (yes/no) HCP intervened to correct an error during administration by subject or caregiver - (yes/no) description of error - (narrative data of observer) self-administration diary brought in (yes/no) date of administration at home correct dose administered at home (yes/no) any other problems with administration at home (yes/no) description of problem, if yes (narrative data)

23 Date: 04 April 2016 Page 23 of 270 The following variables were calculated from some of the data variables above: romiplostim dose administered by subject or caregiver - micrograms; calculated by observed injection volume (milliliter) x 500 (micrograms/milliliter) romiplostim dose prescribed by physician - micrograms; calculated by prescribed injection volume (milliliter) x 500 (micrograms/milliliter) 9.5 Data Sources and Measurement This was a non-interventional study in which data were collected as a part of routine clinical care. The HCP completed a standardized data collection form to record their direct observation of a subject or caregiver administering romiplostim. 9.6 Bias A new-user design was chosen for this study to help avoid any biases identified from historical use and self-administration exposure with romiplostim. Enrollment began within 1 year of HAT pack availability in the EU countries and was restricted to subjects treated per the most recent EU SmPC or caregivers new to administering romiplostim (or at least a 3-month gap). Consent was obtained following the initial prescription for self-administration. An important component when assessing the effectiveness of the HAT pack training is the exposure window with respect to the HAT pack training. Subjects or caregivers were assessed 2 to 8 weeks after completion of the HAT pack training. One subject (Subject ) was excluded from the Full Analysis Set as the first standard-of-care visit occurred only 1 week after HAT pack training. A second visit occurred 2 weeks after HAT pack training. Of note, the subject administered romiplostim correctly at both of these visits. A longer exposure window of 16 weeks post HAT pack training was voluntary and occurred if the HCP requested additional observations. Selection bias of subjects may have occurred if there was a tendency to include patients who had performed well during the initial HAT pack training in the study. To reduce potential selection bias, consent from patients was solicited before the HAT pack training and they were enrolled consecutively. Inaccuracies in the recording of clinical information (ie, data abstraction and data entry procedures) have the potential to produce information bias. An electronic case report form was developed to standardize data entry across the study centers to minimize information bias. Use of data from the self-administration diary was collected in order to assess ability to administer when the subject or caregiver was not observed as being

24 Date: 04 April 2016 Page 24 of 270 under observation may cause bias. Quality control procedures were implemented to minimize potential data errors (Section 9.8 of Annex 2). 9.7 Study Size The sample size for this study was 40 subjects or caregivers, which allowed an adequately precise point estimate for the primary endpoint and was feasible for the study enrollment target given the rarity of ITP. Further details are provided in the Statistical Analysis Plan (SAP) in Annex Data Transformation Not applicable. 9.9 Statistical Methods Details of the statistical analyses for this study are provided in the SAP in Annex Main Summary Measures To obtain the proportion of subject participation (ie, cooperation proportion), the number of eligible subjects who were offered participation but refused were provided by each enrolling site. Background demographics and baseline characteristics including sex, age, and administerer of romiplostim (subject or caregiver) were summarized and displayed in listings. The analyses of the primary and secondary endpoints are provided in Section Variables associated with the primary and secondary endpoints achievement status were displayed in listings. Assessment of the secondary endpoints at the follow-up visits were also presented in the listings. Safety data were to be summarized overall and by country. The Medical Dictionary for Regulatory Activities version 16.0 or later was to be used to code all adverse drug reactions (ADRs) to a system organ class (SOC) and a preferred term (PT). All ADRs (serious, nonserious, and overall) were to be tabulated by SOC and PT in descending order of frequency. The tabulations were to be presented by pre- and post-completion of the HAT pack training (day 1). ADRs with a start date on or after the date of consent, but before the date of completion of the HAT pack training were to be classified as pre-completion of the HAT pack training and ADRs with a start date on or after the date of completion of the HAT pack training were classified as post-completion of the HAT pack training.

25 Date: 04 April 2016 Page 25 of 270 Dosing administration variables used to derive the composite primary and secondary endpoints were summarized and listed. All available data transcribed from the self-administration diary at the first standard-of-care 4-week visit were listed for each subject Main Statistical Methods The analyses were descriptive in nature and no formal hypothesis was tested. Endpoints of a categorical binary nature (yes/no) were summarized as the frequencies and proportions (percentages) based on the appropriate analysis set denominator. For the primary and secondary endpoints, a 95% confidence interval (CI) (Wilson s score method with no continuity correction) was calculated around the proportions responding yes/no, as a measure of precision. Endpoints of continuous nature were described by the mean, standard deviation, range (minimum and maximum), median, and lower and upper quartiles. Detailed statistical methods used in this study are provided in the SAP Section 9 (Annex 4) Missing Values Missing data was expected to be minimal for the background and study drug administration variables collected in the standardized data collection form. If no response was entered for the composite categorical binary (yes/no) endpoints, then a conservative approach was to be assumed (the criteria were not met) and a no was to be imputed for any of the questions or items that comprise the endpoint. This would have enabled the sum of the numerators across the categories to be equal to the denominator for the appropriate analysis set. If no response was entered for the continuous endpoints, then no imputation was to be performed and summary statistics were to be calculated based on a reduced denominator. If the endpoint was comprised of 2 continuous variables (ie, the difference between the prescribed and administered dose of romiplostim), no imputation was to be performed and the endpoint was missing. The actual responses from the individual variables in the standardized data collection form were displayed in the Listings, including missing if applicable Sensitivity Analyses No sensitivity analyses were performed on the data in this study.

26 Date: 04 April 2016 Page 26 of Amendments to the Statistical Analysis Plan There were no changes to the statistical analysis plan (Annex 4); however, there were no recorded ADRs. Therefore, the safety analyses of ADRs as described in Section were not completed Quality Control To ensure the quality of clinical data across all subjects and study centers, a clinical data management review was performed on subject data received at Amgen. During this review, data were checked for consistency, omissions, and any apparent discrepancies. In addition, the data were reviewed for adherence to the protocol and Good Clinical Practice guidelines (as applicable by local law). To resolve any questions arising from the clinical data management review process, data queries and/or study center notifications were created in the electronic data capture system database for study center resolution and closed by an Amgen reviewer.

27 Date: 04 April 2016 Page 27 of RESULTS 10.1 Participants The recruitment period began on 07 July The last subject last visit was on 20 November 2015 and the database lock was on 20 January A total of 41 patients or caregivers were provided with HAT pack training and were considered eligible for this study. A total of 40 subjects or caregivers were enrolled in this study from 12 study centers across 8 countries and comprised the Full Analysis Set. They were from Austria (3 subjects), Belgium (1 subject), France (1 subject), Germany (4 subjects), Greece (4 subjects and 5 caregivers), The Netherlands (8 subjects and 4 caregivers), Spain (4 subjects and 2 caregivers), and The United Kingdom (3 subjects and 1 caregiver) (Table and Table and Listing and Listing ). One subject (Subject ) in The United Kingdom became ineligible and was not included in the Full Analysis Set because the first standard-of-care visit occurred only 1 week after training with the HAT pack (Listing ) Descriptive Data The study population consisted of more autonomous ITP patients than caregiver-assisted patients (28 subjects [70%] versus 12 caregivers [30%]). A summary of the demographic characteristics is provided in Table 10-1.

28 Date: 04 April 2016 Page 28 of 270 Table Summary of Demographics and Baseline Characteristics (Full Analysis Set) Total (N = 40) Sex - n (%) Male 17 (42.5) Female 23 (57.5) Age (years) n 40 Mean 58.7 SD 18.2 Median 60.0 Q1, Q3 46.0, 73.5 Min, Max 25, 91 Age group - n (%) years 23 (57.5) >= 65 years 17 (42.5) Administerer of drug - n (%) Patient 28 (70.0) Caregiver 12 (30.0) SD = Standard Deviation. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-dm-sum.sas Output: t dm-sum-fas-p.rtf (Date Generated: 22FEB16:04:24:31 ) Source: sdf.adsl 10.3 Outcome Data The Full Analysis Set included 40 subjects or caregivers who received HAT pack training and attended their first standard-of-care visit 2 to 8 weeks afterwards. This analysis set was used to analyze all of the outcomes in this study Main Results Primary and Secondary Endpoints The primary and secondary endpoints were described in Section 9.1 of this report. The results are summarized below.

29 Date: 04 April 2016 Page 29 of 270 Table Summary of Primary and Secondary Endpoints at the First 4-Week Visit (Full Analysis Set) Total (N = 40) n (%) (95% CI) Primary Endpoint Subject or caregiver administers romiplostim correctly Yes 35 (87.5) (73.9, 94.5) No 5 (12.5) (5.5, 26.1) Secondary Endpoints: Subject or caregiver reconstitutes romiplostim correctly Yes 39 (97.5) (87.1, 99.6) No 1 (2.5) (0.4, 12.9) Subject or caregiver injects romiplostim successfully Yes 40 (100) (91.2, 100) No 0 (0) (0, 8.8) Difference (%) between the prescribed and administered dose (micrograms) of romiplostim N 40 Mean 0.16 SD 1.01 Median 0.00 Q1, Q3 0.00, 0.00 Min, Max 0.0, 6.4 First standard-of-care 4-week visit was 2 to 8 weeks after completion of training with the HAT pack. CI = confidence interval, HAT = home administration training, SD = standard deviation. Source: Table , Table , and Table For the primary endpoint, at the first standard-of-care 4-week visit (2 to 8 weeks after completing the HAT pack training), 35 subjects or caregivers (87.5%) administered romiplostim correctly (95% CI: 73.9%, 94.5%) (ie, reconstituted romiplostim correctly, administered the prescribed dose accurately, and successfully injected romiplostim without any HCP intervention) and 5 subjects or caregivers (12.5%) did not administer romiplostim correctly (95% CI: 5.5%, 26.1%) (Table 10-2). A HCP intervened in 5 instances to correct the administration of romiplostim for 3 subjects and 2 caregivers (Table and Listing ). Of the 3 subjects who did not administer romiplostim correctly, 1 subject (Subject ) did not ensure all romiplostim was dissolved (Table and Listing ), 1 subject (Subject ) was insecure and needed verbal encouragement, and 1 subject (Subject ) needed nursing intervention to read the correct dose from the vial because she didn t see well with her glasses worn. Of the 2 caregivers who did not administer romiplostim correctly, 1 caregiver (for Subject ) needed verbal support and the other caregiver (for Subject ) needed guidance with regards to the syringe and vial connection. The caregiver connected the syringe with the

30 Date: 04 April 2016 Page 30 of 270 needle instead of the vial, and broke the needle. A new needle was provided and romiplostim was able to be administered correctly (Table 10-2 and Listing ). For the secondary endpoints, 39 subjects or caregivers (97.5%) reconstituted romiplostim correctly and 1 subject (2.5%) (Subject , see above) did not reconstitute romiplostim correctly; the subject did not ensure all romiplostim was dissolved (Table ). All 40 subjects or caregivers injected romiplostim successfully and the difference between the prescribed and administered dose of romiplostim was within the 10% margin error for dose despite 1 case (2.5%) of dose preparation inaccuracy (1 caregiver [for Subject ] prepared a dose of 250 mcg instead of 235 mcg) (Table and Listing ). Follow-up visits occurred only if the HCP requested them. A total of 6 subjects (15%) in the Full Analysis Set had follow-up visits. Three subjects (Subjects , , ) were assessed during their first standard-of-care visit and at 3 subsequent monthly routine visits. One subject (Subject ) was assessed during the first standard-of-care visit and at 1 subsequent monthly routine visit. One subject (Subject ) and 1 caregiver (for Subject ) were assessed at the first standard-of-care visit and during an additional monitoring visit because they previously required verbal encouragement from the HCP (Listing and Listing ) Dose Administration Variables Supporting the Primary, Secondary, and Additional Endpoints The subjects or caregivers performed the process of self-administration utilizing 250 or 500 mcg kits. At the first standard-of-care 4-week visit, a total of 36 subjects (90%) used 1 vial; 3 subjects (7.5%) used 2 vials and 1 subject (2.5%) used 3 vials of romiplostim for single administration, the dose ranged from 75 to 840 mcg (Table and Listing ). A summary of dose administration variables supporting the primary, secondary, and additional endpoints are provided in Table Self-administration Diary Data transcribed from the self-administration diary at the first standard-of-care visit is provided in Listing A total 18 subjects (45%) brought their self-administration diary to the first standard-of-care 4-week visit. Of these, 2 subjects recorded that a correct dose of

31 Date: 04 April 2016 Page 31 of 270 romiplostim was not self-administered on the correct date; Subject decided to administer the week 3 injection of romiplostim later than was prescribed and Subject administered the week 4 injection of romiplostim earlier than was prescribed. Of note, these 2 subjects administered romiplostim correctly at the first standard-of-care 4-week visit Other Analyses No additional analyses were conducted Adverse Events/Adverse Reactions No ADRs were reported.

32 Date: 04 April 2016 Page 32 of DISCUSSION 11.1 Key Results This study was conducted to help determine whether the HAT pack needed to be refined and to provide an early indication of the potential for medication errors associated with self-administration. The primary objective was to estimate the proportion of subjects and caregivers who administered romiplostim correctly. At the first standard-of-care visit, 4 weeks (range: 2 to 8 weeks) after training with the HAT pack, 35 subjects or caregivers (87.5%) administered romiplostim correctly. A total of 5 subjects or caregivers (12.5%) did not administer romiplostim correctly (3 subjects and 2 caregivers). Of the 3 subjects who did not administer romiplostim correctly, 1 subject did not ensure all romiplostim was dissolved, 1 subject was insecure and needed verbal encouragement, and 1 subject needed nursing intervention to read the correct dose from the vial because she didn t see well with her glasses worn. Of the 2 caregivers who did not administer romiplostim correctly, 1 caregiver needed verbal support and the other caregiver did not reconstitute romiplostim correctly and needed guidance with regards to the syringe and vial connection. The caregiver connected the syringe with the needle instead of the vial, and broke the needle. In these 5 instances the HCP intervened during the administration by the subject or caregiver. For the secondary objectives, at the first standard-of-care 4-week visit, 1 subject (2.5%) did not reconstitute romiplostim correctly; the subject did not ensure all of the romiplostim was dissolved. All 40 subjects or caregivers injected romiplostim successfully. The difference between the prescribed and administered dose of romiplostim was within the margin of error despite 1 case (2.5%) of dose preparation inaccuracy (1 caregiver prepared a dose of 250 mcg instead of 235 mcg). No ADRs were reported Limitations Given that this study was conducted on a convenience versus random sample of patients, generalizability of the results may be limited. ITP is a rare disease, only a subgroup of patients receive romiplostim, and a smaller subset of these patients are evaluated and approved by their prescribers for self-administration, leading to a relatively small source population of patients that may be widely dispersed or concentrated in specific institutions. Moreover, the availability of self-administration kits is limited to select countries due to reimbursement restrictions. Given that the source

33 Date: 04 April 2016 Page 33 of 270 population is likely in the range of hundreds, the sample size of 40 reflects a sizable portion of the source population. The study results may have limited generalizability if only a subset of eligible subjects actually participated in the study. To assess for this bias, a 'cooperation proportion' for clinical study centers and subjects was tallied by the number of clinical study centers and subjects participating in the study divided by the number offered inclusion. A total of 46 subjects were offered inclusion into the study and 5 subjects did not enroll into the study. Therefore, the proportion of subject participation (ie, cooperation proportion) was 89.1% (data on file). To identify study subjects within this source population, clinical study centers with clinicians known to treat ITP patients with romiplostim were approached for possible participation in the study. Study centers and countries were selected based on availability and willingness to participate. Selection bias may have occurred if investigators selectively included some ITP patients who are different from the overall source population of ITP patients or caregivers who are different from the general caregiver population. To reduce the influence of this potential bias, no minimum number of subjects treated with romiplostim, preference to use of self-administration, or other criteria were applied when selecting sites. Another possible limitation is that the direct observation at the first standard-of-care visit may not reflect the subject s or caregiver s ability to administer romiplostim between clinical visits when the act of administration is not observed. To assess this issue, data from the self-administration diary were collected. Because it was not compulsory to bring the diary to the first standard-of-care 4-week visit, 18 subjects (45%) brought their self-administration diary to the first standard-of-care visit and the quality of completion varied among subjects. The observers could also be biased by preconceived expectations of what they thought they should observe or the subjects could act differently when being watched (ie, Hawthorne effect [McCarney et al, 2007]). Direct observation, as performed by different observers, may have added a level in inter-observer variability. In this study, the impact of inter-observer variability was minimal as 95% of the subjects (or caregivers) who performed self-administration at the standard-of-care 4-week visit were observed by the same person who trained them with the HAT pack (Table ).

34 Date: 04 April 2016 Page 34 of Interpretation This study was conducted to measure the effectiveness of the HAT pack training in mitigating the risk of medication errors from self-administration. The study has helped to highlight potential sources of medication errors associated with self-administration. Since subsequent observations were voluntary and that data was limited, it is not possible to determine whether all study participants continued to successfully administer romiplostim after the first standard-of-care 4-week visit and whether any dose adjustment or re-evaluation of the self-administration prescription was required for these subjects. Of note, 3 subjects were repeatedly assessed at 3 subsequent monthly routine visits and continued to administer romiplostim correctly within this extended follow-up period. At the first standard-of-care 4-week visit, the HCP intervened during 5 instances of romiplostim self-administration for 3 subjects and 2 caregivers. For each of these instances there are instructions and specific wording in the current HAT pack training materials which explain the corresponding risk minimization steps to be taken (Table 11-1). In accordance with the European Medicines Agency s Good Practice Guide on Risk Minimisation and Prevention of Medication Errors guidance (2015), Amgen provides HAT pack training to physicians or HCPs and patients (Annex 5). The current HAT pack materials for the HCPs include a guide for selecting and training eligible patients for home administration. Patient selection is an important factor to consider when prescribing the self-administration kit and the HAT pack as it is essential that the patients or their caregivers have the ability to follow instructions, to properly handle all of the required elements of self-administration, and have no physical condition that prevents them from accurately, reconstituting and injecting romiplostim. In addition, the HAT pack materials for the patients provide instructions and specific wording indicating steps where care needs to be taken to avoid the risk of medication error, the importance of checking that romiplostim has been reconstituted correctly, the importance of receiving the correct dose, and the consequences of dosing errors. This is further emphasized by the step-by-step self-administration video within the HAT pack, where the steps during which medication errors may occur are highlighted. Table 11-1 summarizes the instances of HCP intervention to correct an error during administration and the corresponding risk minimization details provided in the training materials.

35 Date: 04 April 2016 Page 35 of 270 As per standard-of-care, the completion of the self-administration diary is not mandatory for subjects. It is also not compulsory for subjects to bring the diary to the first standard-of-care visit. Hence, the data collected from the self-administration diary was limited and it is not possible to get a good understanding of subjects behaviors and the encountered issues within their home settings. Two instances of incorrect administration date were reported by 2 subjects in their self-administration diaries (Listing ). The self-administration diary (Annex 5) highlights the importance of taking the right dose at the right time and has fields to capture the day and date the romiplostim dose is due, amount of administered dose, and a yes or no indicator for whether the right dose is taken on the right date.

36 Date: 04 April 2016 Page 36 of 270 Table Instances of Healthcare Professional Intervention to Correct an Error During Administration at the First Standard-of-care 4-week Visit (Study ) and Related HAT Pack Training Materials Instances of Healthcare Professional Intervention Training Materials Risk Minimization Details in Training Materials Failure to check that all romiplostim was dissolved Step-by-step Guide to Preparation and Step 4 - Dissolve Nplate by injecting water into vial: Administration of the Nplate Injection (Subject forgot to check if all romiplostim was dissolved) Holding the area where the vial and vial adapter connect between your fingers, gently swirl the vial by rotating your wrist until all of the powder has dissolved and the liquid in the vial is clear and colorless (page 11). Before continuing: Do visually inspect the dissolved liquid for particles and/or discolouration. It must be clear and colourless and fully dissolved. Note: If there is any colour or particles in the liquid, contact your healthcare professional. Do make sure liquid is fully dissolved before removing syringe (page 12). When Nplate is completely dissolved, remove the empty syringe by twisting it anti-clockwise off of the vial adapter (page 12). Incorrect attachment of the syringe and vial, and incorrect preparation of dose (Caregiver [Subject ], connected the syringe with the needle instead of the vial and the needle broke, and the caregiver prepared a dose of 250 mcg instead of 235 mcg) Source: Listing and Annex 5 Quick Guide to Home Administration of Nplate Quick guide checklist Firm connections: Make sure you follow the instructions and your training when attaching parts of the kit (page 5). Taking the right dose at the right time: Make sure that you administer the exact dose your healthcare professional has prescribed (page 3). Quick guide checklist The right dose of Nplate: Even in small doses Nplate can have an influence on platelet production, which is why you need to take the correct dose so remember to check that you are administering the exact dose that your healthcare professional has prescribed. Your healthcare professional will tell you the correct dose to administer. This will be recorded in your Self-administration diary (page 5). Page 1 of 2

37 Date: 04 April 2016 Page 37 of 270 Table Instances of Healthcare Professional Intervention to Correct an Error During Administration at the First Standard-of-care 4-week Visit (Study ) and Related HAT Pack Training Materials Instances of Healthcare Professional Intervention Training Materials Risk Minimization Details in Training Materials Incorrect attachment of the syringe and vial, and incorrect preparation of dose (continued) Poor eyesight (Subject : Nursing intervention was required to get the correct dose from the vial because patient says she does not see well with glasses worn ) Verbal encouragement (Caregiver [Subject ] was given some verbal support) (Subject is insecure and needs some verbal stimulation) Source: Listing and Annex 5 Step-by-step Guide to Preparation and Administration of the Nplate Injection Selecting and Training Patients for Home Administration of Nplate Quick Guide to Home Administration of Nplate Selecting and Training Patients for Home Administration of Nplate Step 4 Dissolve Nplate by injecting water into vial : Keeping the vial on the table, attach water-filled syringe to vial adapter by holding the side of the vial adapter with one hand and twisting the syringe tip clockwise onto the adapter with the other hand until you feel a slight resistance (page 9). Step 6 Prepare injection needle: Attach safety needle to filled syringe. Apply strong force while twisting to attach the safety needle onto syringe (page 17). Step 5 Prepare new syringe for injection: Check your dose by looking at your dose card, make sure the top of the plunger head lines up with the syringe marking that matches your prescribed dose. If necessary push liquid back into the vial to achieve the desired dose. Do a final check to ensure the correct amount of liquid for your dose is in the syringe and all air bubbles have been removed (page 15). Criteria for selecting patients: They have no physical condition that prevents them accurately reconstituting and injecting Nplate (page 4) Training for home administration: The time required to learn how to prepare and administer Nplate will vary across patients. The most important thing is that you feel confident to prepare and administer the injection without direct supervision of a healthcare professional (page 6) Performing training steps : After the first 4 weeks of self-administration, the patient should again be supervised while reconstituting and administering Nplate (page 7). Page 2 of 2

38 Date: 04 April 2016 Page 38 of Generalizability The overall goal of this study was to identify possible refinement of the HAT pack and serve as an early indicator of the potential for medication errors associated with self-administration. By including data from 8 countries across Europe and a sizable proportion of the source population reflected by the sample, the results of this study can provide a broad insight to the use of the HAT pack by patients and caregivers. It is however limited to select ITP patients treated per the EU SmPC, identified as being eligible for self-administration, and who had access to the self-administration kit and the HAT pack in their countries. Exposure and familiarity to drug administration devices may vary among the caregivers included in the study as they may be either family/friends naïve to such devices, or HCPs used to such devices but new to the romiplostim kit. In addition, participants from countries where the drug administrators are mainly limited to HCPs (such as hospital pharmacists and clinicians) may not continue with self-administration. Given the heterogeneous nature of the study population, the variations in health literacy levels, and the differing needs and motivations to adopt self-administration, the study may not be fully representative in terms of the autonomous and caregiver-assisted patients who would self-administer romiplostim and use the HAT pack.

39 Date: 04 April 2016 Page 39 of OTHER INFORMATION Not applicable

40 Date: 04 April 2016 Page 40 of CONCLUSION The results of this study highlight the importance of supervising the patient or caregiver again after the first 4 weeks of self-administration (as required by the EU SmPC and the HAT pack training materials), patient selection by the physician, and the HAT pack training in helping to mitigate the risk of medication errors during self-administration. The main conclusions from the study are: Most subjects or caregivers who received HAT pack training administered romiplostim correctly at the first standard-of-care 4-week visit (35 subjects or caregivers [87.5%]). The HAT pack training helped to mitigate the risk of medication errors during self-administration in this study population. The current HAT pack contains sufficient information and is an effective risk minimization tool for medication errors during self-administration of romiplostim, but patient selection by the HCP remains an important factor.

41 Date: 04 April 2016 Page 41 of REFERENCES Amgen Ltd. (2016). Nplate (romiplostim) for subcutaneous injection. Summary of Product Characteristics. Retrieved from _Product_Information/human/000942/WC pdf accessed on 03 March Barker KN, Flynn EA, Pepper GA, Bates DW, Mikeal RL. Medication errors observed in 36 health care facilities. Arch Intern Med. 2002;162: European Medicine s Agency (2015). Good practice guide on risk minimisation and prevention of medication errors guidance. Retrieved from uideline/2015/11/wc pdf accessed on 14 March Human Factor Engineering Study. Nplate self-administration kit validation study. 18 April On file at Amgen, Thousand Oaks, CA. McCarney R, Warner J, Iliffe S, can Haselen R, Giffin M, Fisher P. The Hawthorne Effect: a randomized, controlled trial. BMC Med Res Methodol. 2007;7:30. National Research Council. Preventing Medication Errors: Quality Chasm Series. Washington, DC: The National Academies Press, Safety Assessment Report. Medication errors in association with the use of romiplostim. Romiplostim Global Safety Team. 24 July On file at Amgen, Thousand Oaks, CA.

42 Date: 04 April 2016 Page 42 of SUMMARY TABLES, FIGURES, AND LISTINGS

43 Date: 04 April 2016 Page 43 of 270 Table Summary of Subject Participation - Overall - All Screened Subjects Total Screened subjects, n 41 Eligible subjects, n 41 Participating subjects (FAS), n 40 Proportion of participating subjects / screened subjects, % 97.6 Proportion of participating subjects / eligible subjects, % 97.6 Page 1 of 1 Screened subjects: all ITP patients who have been provided with HAT pack training and assessed for inclusion in the study. Eligible subjects: all screened subjects who satisfy the Protocol eligibility criteria. Participating subjects (FAS): all eligible subjects who attend the first standard-of-care visit 2-8 weeks after training with the HAT pack. FAS = Full Analysis Set. HAT = Home Administration Training. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-ds-sum.sas Output: t ds-sum-p.rtf (Date Generated: 21JAN2016:08:41) Source Data: sdf.adsl

44 Date: 04 April 2016 Page 44 of 270 Table Summary of Subject Participation - by Country - All Screened Subjects Austria Screened subjects, n 3 Eligible subjects, n 3 Participating subjects (FAS), n 3 Total Proportion of participating subjects / screened subjects, % Proportion of participating subjects / eligible subjects, % Belgium Screened subjects, n 1 Eligible subjects, n 1 Participating subjects (FAS), n 1 Proportion of participating subjects / screened subjects, % Proportion of participating subjects / eligible subjects, % France Screened subjects, n 1 Eligible subjects, n 1 Participating subjects (FAS), n 1 Proportion of participating subjects / screened subjects, % Proportion of participating subjects / eligible subjects, % Page 1 of 3 Screened subjects: all ITP patients who have been provided with HAT pack training and assessed for inclusion in the study. Eligible subjects: all screened subjects who satisfy the Protocol eligibility criteria. Participating subjects (FAS): all eligible subjects who attend the first standard-of-care visit 2-8 weeks after training with the HAT pack. FAS = Full Analysis Set. HAT = Home Administration Training. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-ds-sum.sas Output: t ds-sum-ctry-p.rtf (Date Generated: 21JAN2016:08:41) Source Data: sdf.adsl

45 Date: 04 April 2016 Page 45 of 270 Table Summary of Subject Participation - by Country - All Screened Subjects Germany Screened subjects, n 4 Eligible subjects, n 4 Participating subjects (FAS), n 4 Total Proportion of participating subjects / screened subjects, % Proportion of participating subjects / eligible subjects, % Greece Screened subjects, n 9 Eligible subjects, n 9 Participating subjects (FAS), n 9 Proportion of participating subjects / screened subjects, % Proportion of participating subjects / eligible subjects, % The Netherlands Screened subjects, n 12 Eligible subjects, n 12 Participating subjects (FAS), n 12 Proportion of participating subjects / screened subjects, % Proportion of participating subjects / eligible subjects, % Page 2 of 3 Screened subjects: all ITP patients who have been provided with HAT pack training and assessed for inclusion in the study. Eligible subjects: all screened subjects who satisfy the Protocol eligibility criteria. Participating subjects (FAS): all eligible subjects who attend the first standard-of-care visit 2-8 weeks after training with the HAT pack. FAS = Full Analysis Set. HAT = Home Administration Training. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-ds-sum.sas Output: t ds-sum-ctry-p.rtf (Date Generated: 21JAN2016:08:41) Source Data: sdf.adsl

46 Date: 04 April 2016 Page 46 of 270 Table Summary of Subject Participation - by Country - All Screened Subjects Spain Screened subjects, n 6 Eligible subjects, n 6 Participating subjects (FAS), n 6 Total Proportion of participating subjects / screened subjects, % Proportion of participating subjects / eligible subjects, % United Kingdom Screened subjects, n 5 Eligible subjects, n 5 Participating subjects (FAS), n 4 Proportion of participating subjects / screened subjects, % 80.0 Proportion of participating subjects / eligible subjects, % 80.0 Page 3 of 3 Screened subjects: all ITP patients who have been provided with HAT pack training and assessed for inclusion in the study. Eligible subjects: all screened subjects who satisfy the Protocol eligibility criteria. Participating subjects (FAS): all eligible subjects who attend the first standard-of-care visit 2-8 weeks after training with the HAT pack. FAS = Full Analysis Set. HAT = Home Administration Training. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-ds-sum.sas Output: t ds-sum-ctry-p.rtf (Date Generated: 21JAN2016:08:41) Source Data: sdf.adsl

47 Date: 04 April 2016 Page 47 of 270 Table Summary of Demographic and Baseline Characteristics - Full Analysis Set Total (N = 40) Sex - n (%) Male 17 (42.5) Female 23 (57.5) Age (years) n 40 Mean 58.7 SD 18.2 Median 60.0 Q1, Q3 46.0, 73.5 Min, Max 25, 91 Age group - n (%) years 23 (57.5) >= 65 years 17 (42.5) Administerer of drug - n (%) Patient 28 (70.0) Caregiver 12 (30.0) Page 1 of 1 SD = Standard Deviation. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-dm-sum.sas Output: t dm-sum-fas-p.rtf (Date Generated: 22FEB16:04:24:31 ) Source: sdf.adsl

48 Date: 04 April 2016 Page 48 of 270 Table Primary Endpoint: Subject or Caregiver Administers Romiplostim Correctly at the First Standard of Care 4 Week Visit - Full Analysis Set Total (N = 40) n (%) (95% CI) Subject or caregiver administers romiplostim correctly Yes 35 (87.5) (73.9, 94.5) No 5 (12.5) (5.5, 26.1) Page 1 of 1 First standard of care 4 week visit is 2-8 weeks after completion of training with the Home Administration Training (HAT) pack. CI = Confidence Interval. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-qs-primary-fas.sas Output: t qs-primary-fas-p.rtf (Date Generated: 22JAN2016:01:52) Source Data: sdf.adqs, sdf.adsl

49 Date: 04 April 2016 Page 49 of 270 Table Secondary Endpoints at the First Standard of Care 4 Week Visit - Full Analysis Set Total (N = 40) n (%) (95% CI) Subject or caregiver reconstitutes romiplostim correctly Yes 39 (97.5) (87.1, 99.6) No 1 (2.5) (0.4, 12.9) Subject or caregiver injects romiplostim successfully Yes 40 (100) (91.2, 100) No 0 (0) (0, 8.8) Page 1 of 1 First standard of care 4 week visit is 2-8 weeks after completion of training with the Home Administration Training (HAT) pack. CI = Confidence Interval. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-qs-secondary-fas.sas Output: t qs-secondary-fas-p.rtf (Date Generated: 22JAN2016:01:53) Source Data: sdf.adqs, sdf.adsl

50 Date: 04 April 2016 Page 50 of 270 Table Secondary Endpoint: Summary of the Difference (%) Between the Prescribed and Administered Dose (Micrograms) of Romiplostim at the First Standard of Care 4 Week Visit - Full Analysis Set Total (N = 40) Difference (%) n 40 Mean 0.16 SD 1.01 Median 0.00 Q1, Q3 0.00, 0.00 Min, Max 0.0, 6.4 Page 1 of 1 First standard of care 4 week visit is 2-8 weeks after completion of training with the Home Administration Training (HAT) pack. SD = Standard Deviation. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-qs-diffdose-fas.sas Output: t qs-diffdose-fas-p.rtf (Date Generated: 19FEB16:00:58:43 ) Source: sdf.adqs, sdf.adsl

51 Date: 04 April 2016 Page 51 of 270 Table Additional Endpoints at the First Standard of Care 4 Week Visit - Full Analysis Set Total (N = 40) n (%) (95% CI) Difference between the prescribed and administered dose of romiplostim is within +/- 10% Yes 40 (100) (91.2, 100) No 0 (0) (0, 8.8) HCP intervened to correct an error during administration by subject or caregiver Yes 5 (12.5) (5.5, 26.1) No 35 (87.5) (73.9, 94.5) Page 1 of 1 First standard of care 4 week visit is 2-8 weeks after completion of training with the Home Administration Training (HAT) pack. CI = Confidence Interval. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-qs-additional-fas.sas Output: t qs-additional-fas-p.rtf (Date Generated: 22JAN2016:01:54) Source Data: sdf.adqs, sdf.adsl

52 Date: 04 April 2016 Page 52 of 270 Table Summary of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care 4 Week Visit - Full Analysis Set Total (N = 40) n (%) HCP observing - clinical site, provider type Clinician 15 (37.5) Nurse 25 (62.5) Other 0 (0.0) HCP observing was same person as trainer Yes 38 (95.0) No 2 (5.0) Number of vials of romiplostim needed for single administration 1 36 (90.0) 2 3 (7.5) 3 1 (2.5) Aseptic techniques were used in preparing the vial/s Yes 40 (100.0) No 0 (0.0) Gently injected all water from sterile water syringe into the vial/s Yes 40 (100.0) No 0 (0.0) Ensured all romiplostim dissolved Yes 39 (97.5) No 1 (2.5) Ensured all air bubbles removed from injection syringe Yes 40 (100.0) No 0 (0.0) Clinically appropriate use of alcohol wipe at injection site Yes 40 (100.0) No 0 (0.0) Clinically appropriate handling of syringe to avoid contamination Yes 40 (100.0) No 0 (0.0) Page 1 of 2 First standard of care 4 week visit is 2-8 weeks after completion of training with the Home Administration Training (HAT) pack. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-qs-sum-endpoints-fas.sas Output: t qs-sum-endpoints-fas-p.rtf (Date Generated: 19FEB16:00:59:47 ) Source: sdf.adqs, sdf.adsl

53 Date: 04 April 2016 Page 53 of 270 Table Summary of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care 4 Week Visit - Full Analysis Set Total (N = 40) n (%) Clinically appropriate technique of subcutaneous injection Yes 40 (100.0) No 0 (0.0) HCP intervened to correct an error during administration by subject or caregiver Yes 5 (12.5) No 35 (87.5) Page 2 of 2 First standard of care 4 week visit is 2-8 weeks after completion of training with the Home Administration Training (HAT) pack. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/tables/t-qs-sum-endpoints-fas.sas Output: t qs-sum-endpoints-fas-p.rtf (Date Generated: 19FEB16:00:59:47 ) Source: sdf.adqs, sdf.adsl

54 Date: 04 April 2016 Page 54 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit Austria Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 04SEP OCT SEP OCT2014 Follow-up visit (observation 2) 06NOV2014 ROUTINE CLINICAL VISIT Follow-up visit (observation 3) 05DEC2014 ROUTINE CLINICAL VISIT Follow-up visit (observation 4) 08JAN2015 ROUTINE VISIT Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 20JAN JAN JAN FEB2015 Follow-up visit (observation 2) 15FEB2015 CLINICAL ROUTINE FOR NON-ITP REASONS Follow-up visit (observation 3) 23FEB2015 ROUTINE CONTROL Follow-up visit (observation 4) 02MAR2015 ROUTINE CONTROL Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 1 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

55 Date: 04 April 2016 Page 55 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit Austria Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 23SEP SEP SEP OCT2015 Follow-up visit (observation 2) 06NOV2015 ROUTINE VISIT Follow-up visit (observation 3) 13NOV2015 ROUTINE VISIT Follow-up visit (observation 4) 20NOV2015 ROUTINE VISIT Belgium Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) France Initial prescription for self-administration Consent Completion of HAT pack training 13JUL JUL JUL AUG JAN2015 N/A 19FEB2015 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 2 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

56 Date: 04 April 2016 Page 56 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit France Standard of care visit (observation 1) 19MAR2015 Germany Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 01DEC DEC DEC DEC DEC DEC DEC JAN JAN JAN JAN FEB2015 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 3 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

57 Date: 04 April 2016 Page 57 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit Germany Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 08JAN JAN JAN FEB2015 Greece Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent 11MAR MAR MAR APR MAY MAY JUN JUN SEP SEP2015 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 4 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

58 Date: 04 April 2016 Page 58 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit Greece Completion of HAT pack training 09SEP2015 Standard of care visit (observation 1) 21OCT Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 05OCT OCT OCT NOV APR APR APR MAY JUN JUN JUN JUL2015 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 5 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

59 Date: 04 April 2016 Page 59 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit Greece Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 17JUN JUN JUN JUL Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) The Netherlands Initial prescription for self-administration Consent 17JUL JUL JUL AUG JUL JUL JUL AUG SEP SEP2014 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 6 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

60 Date: 04 April 2016 Page 60 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit The Netherlands Completion of HAT pack training 30SEP2014 Standard of care visit (observation 1) 28OCT2014 Follow-up visit (observation 2) 25NOV2014 CONTROL MOMENT Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 07OCT OCT OCT NOV2014 Follow-up visit (observation 2) 02DEC2014 MOMENT OF CONTROL Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent 19NOV NOV NOV DEC DEC DEC2014 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 7 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

61 Date: 04 April 2016 Page 61 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit The Netherlands Completion of HAT pack training 23DEC2014 Standard of care visit (observation 1) 27JAN Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 30DEC DEC DEC JAN FEB FEB FEB FEB MAY MAY MAY JUN2015 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 8 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

62 Date: 04 April 2016 Page 62 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit The Netherlands Follow-up visit (observation 2) 25JUN2015 PATIENT WAS INSECURE ABOUT SELFADMINISTRATION AND NEEDS VERBAL STIMULATION Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training 02JUN JUN JUN JUL JUN JUN JUN JUN SEP SEP SEP2015 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 9 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

63 Date: 04 April 2016 Page 63 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit The Netherlands Standard of care visit (observation 1) 13OCT Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Spain Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 22SEP SEP SEP OCT OCT OCT OCT NOV AUG AUG AUG SEP2014 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 10 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

64 Date: 04 April 2016 Page 64 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit Spain Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 06AUG AUG AUG SEP Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent 09JUN JUN JUN JUL JUN JUN JUN JUL JUN JUN2014 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 11 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

65 Date: 04 April 2016 Page 65 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit Spain Completion of HAT pack training 30JUN2014 Standard of care visit (observation 1) 28JUL Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) United Kingdom Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 14AUG AUG AUG SEP DEC DEC DEC JAN JAN FEB FEB FEB2015 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 12 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

66 Date: 04 April 2016 Page 66 of 270 Listing Listing of Relevant Dates by Country, Site and Subject - Eligible Subject Set Country Site Subject Date Type Date Reason for Follow-up Visit United Kingdom Follow-up visit (observation 2) 23FEB2015 STANDARD OF CARE Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) Initial prescription for self-administration Consent Completion of HAT pack training Standard of care visit (observation 1) 03FEB FEB FEB MAR MAR MAR APR APR SEP SEP SEP OCT2015 Subject not included in the Full Analysis Set. HAT = Home Administration Training. Page 13 of 13 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-ds-dates-ess.sas Output: l ds-dates-ess-l.rtf (Date generated: 02FEB2016:09:51) Source data: sdf.adqs

67 Date: 04 April 2016 Page 67 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Austria M 32 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 09OCT2014 Yes Yes Yes Yes No 06NOV2014 Yes Yes Yes Yes No 05DEC2014 Yes Yes Yes Yes No 08JAN2015 Yes Yes Yes Yes No Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs Page 1 of 41

68 Date: 04 April 2016 Page 68 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Austria F 67 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 09FEB2015 Yes Yes Yes Yes No 15FEB2015 Yes Yes Yes Yes No 23FEB2015 Yes Yes Yes Yes No 02MAR2015 Yes Yes Yes Yes No Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs Page 2 of 41

69 Date: 04 April 2016 Page 69 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Austria M 49 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 29OCT2015 Yes Yes Yes Yes No 06NOV2015 Yes Yes Yes Yes No 13NOV2015 Yes Yes Yes Yes No 20NOV2015 Yes Yes Yes Yes No Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs Page 3 of 41

70 Date: 04 April 2016 Page 70 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Belgium F 61 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 03AUG2015 No No Yes Yes Yes Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs Page 4 of 41

71 Date: 04 April 2016 Page 71 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug France M 61 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 19MAR2015 Yes Yes Yes Yes No Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs Page 5 of 41

72 Date: 04 April 2016 Page 72 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Germany M 74 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 22DEC2014 Yes Yes Yes Yes No Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs Page 6 of 41

73 Date: 04 April 2016 Page 73 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Germany M 73 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 20JAN2015 Yes Yes Yes Yes No Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs Page 7 of 41

74 Date: 04 April 2016 Page 74 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Germany M 81 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 04FEB2015 Yes Yes Yes Yes No Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs Page 8 of 41

75 Date: 04 April 2016 Page 75 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Germany F 77 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 10FEB2015 Yes Yes Yes Yes No Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs Page 9 of 41

76 Date: 04 April 2016 Page 76 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Greece F 54 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 22APR2015 Yes Yes Yes Yes No Page 10 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

77 Date: 04 April 2016 Page 77 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Greece F 72 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 29JUN2015 Yes Yes Yes Yes No Page 11 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

78 Date: 04 April 2016 Page 78 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Greece M 82 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 21OCT2015 Yes Yes Yes Yes No Page 12 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

79 Date: 04 April 2016 Page 79 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Greece F 25 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 06NOV2015 Yes Yes Yes Yes No Page 13 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

80 Date: 04 April 2016 Page 80 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Greece F 46 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 28MAY2015 Yes Yes Yes Yes No Page 14 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

81 Date: 04 April 2016 Page 81 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Greece M 85 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 16JUL2015 Yes Yes Yes Yes No Page 15 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

82 Date: 04 April 2016 Page 82 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Greece F 46 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 17JUL2015 Yes Yes Yes Yes No Page 16 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

83 Date: 04 April 2016 Page 83 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Greece F 74 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 17AUG2015 Yes Yes Yes Yes No Page 17 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

84 Date: 04 April 2016 Page 84 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Greece F 79 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 28AUG2015 Yes Yes Yes Yes No Page 18 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

85 Date: 04 April 2016 Page 85 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands M 26 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 28OCT2014 Yes Yes Yes Yes No 25NOV2014 Yes Yes Yes Yes No Page 19 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

86 Date: 04 April 2016 Page 86 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands M 66 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 04NOV2014 No Yes Yes Yes Yes 02DEC2014 Yes Yes Yes Yes No Page 20 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

87 Date: 04 April 2016 Page 87 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands M 49 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 17DEC2014 Yes Yes Yes Yes No Page 21 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

88 Date: 04 April 2016 Page 88 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands F 33 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 27JAN2015 Yes Yes Yes Yes No Page 22 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

89 Date: 04 April 2016 Page 89 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands M 64 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 27JAN2015 Yes Yes Yes Yes No Page 23 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

90 Date: 04 April 2016 Page 90 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands F 53 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 26FEB2015 Yes Yes Yes Yes No Page 24 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

91 Date: 04 April 2016 Page 91 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands F 73 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 18JUN2015 No Yes Yes Yes Yes 25JUN2015 Yes Yes Yes Yes No Page 25 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

92 Date: 04 April 2016 Page 92 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands F 28 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 07JUL2015 Yes Yes Yes Yes No Page 26 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

93 Date: 04 April 2016 Page 93 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands F 59 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 30JUN2015 Yes Yes Yes Yes No Page 27 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

94 Date: 04 April 2016 Page 94 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands M 46 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 13OCT2015 Yes Yes Yes Yes No Page 28 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

95 Date: 04 April 2016 Page 95 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands M 79 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 20OCT2015 Yes Yes Yes Yes No Page 29 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

96 Date: 04 April 2016 Page 96 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug The Netherlands F 46 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 17NOV2015 Yes Yes Yes Yes No Page 30 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

97 Date: 04 April 2016 Page 97 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Spain F 46 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 04SEP2014 Yes Yes Yes Yes No Page 31 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

98 Date: 04 April 2016 Page 98 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Spain F 47 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 04SEP2014 Yes Yes Yes Yes No Page 32 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

99 Date: 04 April 2016 Page 99 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Spain F 52 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 07JUL2014 Yes Yes Yes Yes No Page 33 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

100 Date: 04 April 2016 Page 100 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Spain F 91 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 07JUL2014 Yes Yes Yes Yes No Page 34 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

101 Date: 04 April 2016 Page 101 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Spain F 72 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 28JUL2014 No Yes Yes Yes Yes Page 35 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

102 Date: 04 April 2016 Page 102 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug Spain F 56 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 18SEP2015 No Yes Yes Yes Yes Page 36 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

103 Date: 04 April 2016 Page 103 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug United Kingdom F 48 Caregiver Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 08JAN2015 Yes Yes Yes Yes No Page 37 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

104 Date: 04 April 2016 Page 104 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug United Kingdom F 57 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 16FEB2015 Yes Yes Yes Yes No 23FEB2015 Yes Yes Yes Yes No Page 38 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

105 Date: 04 April 2016 Page 105 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug United Kingdom M 71 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 09MAR2015 Yes Yes Yes Yes No Page 39 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

106 Date: 04 April 2016 Page 106 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug United Kingdom M 81 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 29APR2015 Yes Yes Yes Yes No Page 40 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

107 Date: 04 April 2016 Page 107 of 270 Listing Listing of Primary, Secondary and Additional Endpoints Status at the First Standard of Care Visit and Follow-up Visits by Country, Site and Subject - Eligible Subject Set Visit Date Subject or Caregiver Administers Romiplostim Correctly? Country Site Subject Sex Age Administerer of Drug United Kingdom M 25 Patient Subject or Caregiver Reconstitutes Romiplostim Correctly? Subject or Caregiver Injects Romiplostim Successfully? Difference Between the Prescribed and Administered Dose of Romiplostim is Within +/- 10%? HCP intervened to correct an error during administration by subject or caregiver? 15OCT2015 Yes Yes Yes Yes No Page 41 of 41 Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-endpoints.sas Output: l qs-endpoints-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

108 Date: 04 April 2016 Page 108 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Austria M 32 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 09OCT2014 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No NOV2014 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No DEC2014 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No JAN2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 1 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

109 Date: 04 April 2016 Page 109 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Austria F 67 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 09FEB2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No FEB2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No FEB2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No MAR2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 2 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

110 Date: 04 April 2016 Page 110 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Austria M 49 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 29OCT2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No NOV2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No NOV2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No NOV2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 3 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

111 Date: 04 April 2016 Page 111 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Belgium F 61 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 03AUG2015 Nurse No Yes Yes No Yes Yes Yes Yes Yes SUBJECT FORGOT TO CHECK IF ALL ROMIPLOSTIM WAS DISSOLVED Page 4 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

112 Date: 04 April 2016 Page 112 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug France M 61 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 19MAR2015 Nurse No Yes Yes Yes Yes Yes Yes Yes No Page 5 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

113 Date: 04 April 2016 Page 113 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Germany M 74 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 22DEC2014 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 6 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

114 Date: 04 April 2016 Page 114 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Germany M 73 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 20JAN2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 7 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

115 Date: 04 April 2016 Page 115 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Germany M 81 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 04FEB2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 8 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

116 Date: 04 April 2016 Page 116 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Germany F 77 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 10FEB2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 9 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

117 Date: 04 April 2016 Page 117 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Greece F 54 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 22APR2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 10 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

118 Date: 04 April 2016 Page 118 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Greece F 72 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 29JUN2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 11 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

119 Date: 04 April 2016 Page 119 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Greece M 82 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 21OCT2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 12 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

120 Date: 04 April 2016 Page 120 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Greece F 25 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 06NOV2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 13 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

121 Date: 04 April 2016 Page 121 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Greece F 46 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 28MAY2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 14 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

122 Date: 04 April 2016 Page 122 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Greece M 85 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 16JUL2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 15 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

123 Date: 04 April 2016 Page 123 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Greece F 46 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 17JUL2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 16 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

124 Date: 04 April 2016 Page 124 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Greece F 74 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 17AUG2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 17 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

125 Date: 04 April 2016 Page 125 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Greece F 79 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 28AUG2015 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 18 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

126 Date: 04 April 2016 Page 126 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands M 26 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 28OCT2014 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No NOV2014 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 19 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

127 Date: 04 April 2016 Page 127 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands M 66 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 04NOV2014 Nurse Yes Yes Yes Yes Yes Yes Yes Yes Yes CAREGIVER WAS GIVEN SOME VERBAL SUPPORT 02DEC2014 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 20 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

128 Date: 04 April 2016 Page 128 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands M 49 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 17DEC2014 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 21 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

129 Date: 04 April 2016 Page 129 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands F 33 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 27JAN2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 22 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

130 Date: 04 April 2016 Page 130 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands M 64 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 27JAN2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 23 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

131 Date: 04 April 2016 Page 131 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands F 53 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 26FEB2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 24 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

132 Date: 04 April 2016 Page 132 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands F 73 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 18JUN2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes Yes SUBJECT IS INSECURE AND NEEDS SOME VERBAL STIMULATION 25JUN2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 25 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

133 Date: 04 April 2016 Page 133 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands F 28 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 07JUL2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 26 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

134 Date: 04 April 2016 Page 134 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands F 59 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 30JUN2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 27 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

135 Date: 04 April 2016 Page 135 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands M 46 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 13OCT2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 28 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

136 Date: 04 April 2016 Page 136 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands M 79 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 20OCT2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 29 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

137 Date: 04 April 2016 Page 137 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug The Netherlands F 46 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 17NOV2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 30 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

138 Date: 04 April 2016 Page 138 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Spain F 46 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 04SEP2014 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 31 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

139 Date: 04 April 2016 Page 139 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Spain F 47 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 04SEP2014 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 32 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

140 Date: 04 April 2016 Page 140 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Spain F 52 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 07JUL2014 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 33 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

141 Date: 04 April 2016 Page 141 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Spain F 91 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 07JUL2014 Clinician Yes Yes Yes Yes Yes Yes Yes Yes No Page 34 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

142 Date: 04 April 2016 Page 142 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Spain F 72 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 28JUL2014 Nurse Yes Yes Yes Yes Yes Yes Yes Yes Yes THE CAREGIVER CONNECTS THE NEEDLE TO DISSOLVE THE LYOPHILIZED POWDER WRONGLY: CONNECTS THE SYRINGE WITH THE NEEDLE INSTEAD OF THE VIAL AND THE NEEDLE BREAKS. A NEW NEEDLE IS PROVIDED AND THEN IS OK Page 35 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

143 Date: 04 April 2016 Page 143 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug Spain F 56 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 18SEP2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes Yes PREPARATION AND ADMINISTRATION OF THE DRUG WAS SUCCESSFUL, BUT NURSING INTERVENTION WAS REQUIRED TO GET THE CORRECT DOSE FROM THE VIAL BECAUSE PATIENT SAYS "SHE DOES NOT SEE WELL WITH GLASSES WORN" Page 36 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

144 Date: 04 April 2016 Page 144 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug United Kingdom F 48 Caregiver Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 08JAN2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 37 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

145 Date: 04 April 2016 Page 145 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug United Kingdom F 57 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 16FEB2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No FEB2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 38 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

146 Date: 04 April 2016 Page 146 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug United Kingdom M 71 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 09MAR2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 39 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

147 Date: 04 April 2016 Page 147 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug United Kingdom M 81 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 29APR2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 40 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

148 Date: 04 April 2016 Page 148 of 270 Listing Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the First Standard of Care Visit and Follow-Up Visits by Country, Site and Subject - Eligible Subject Set Administerer Country Site Subject Sex Age of Drug United Kingdom M 25 Patient Visit Date [A] [B] [C] [D] [E] [F] [G] [H] [I] [J] [K] [L] [M] [N] [O] 15OCT2015 Nurse Yes Yes Yes Yes Yes Yes Yes Yes No Page 41 of 41 [A] HCP observing - clinical site, provider type; [B] HCP observing was same person as trainer; [C] Number of vials of romiplostim needed for single administration; [D] Vial sizes of romiplostim (micrograms); [E] Aseptic techniques were used in preparing the vial/s; [F] Gently injected all water from sterile water syringe into the vial/s; [G] Ensured all romiplostim dissolved; [H] Ensured all air bubbles removed from injection syringe; [I] Clinically appropriate use of alcohol wipe at injection site; [J] Clinically appropriate handling of syringe to avoid contamination; [K] Clinically appropriate technique of subcutaneous injection; [L] HCP intervened to correct an error during administration by subject or caregiver; [M] If HCP did intervene, description of the error; [N] Total administered dose of romiplostim (micrograms); [O] Total prescribed dose of romiplostim (micrograms). Subject not included in the Full Analysis Set. HCP = Healthcare Professional. Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-doseadmin.sas Output: l qs-doseadmin-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adqs

149 Date: 04 April 2016 Page 149 of 270 Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set Subject Sex Age Administerer of Drug Visit Date Country Site Austria Was Dose Diary Brought in? Date of Administration at Home Was the Correct Dose Taken on the Right Date? Any Other Problems With Administration at Home? M 32 Patient 09OCT2014 Yes 11SEP2014 Yes 18SEP2014 Yes 02OCT2014 No PATIENT INJECTED 1 WEEK TOO LATE, PATIENT'S DECISION F 67 Patient 09FEB2015 No M 49 Patient 29OCT2015 No Subject not included in the Full Analysis Set. Page 1 of 11 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-diary.sas Output: l qs-diary-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adsad

150 Date: 04 April 2016 Page 150 of 270 Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set Country Site Belgium Subject Sex Age Administerer of Drug Visit Date Was Dose Diary Brought in? Date of Administration at Home Was the Correct Dose Taken on the Right Date? F 61 Patient 03AUG2015 Yes 23JUL2015 Yes Subject not included in the Full Analysis Set. Any Other Problems With Administration at Home? Page 2 of 11 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-diary.sas Output: l qs-diary-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adsad

151 Date: 04 April 2016 Page 151 of 270 Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set Country Site France Subject Sex Age Administerer of Drug Visit Date Was Dose Diary Brought in? M 61 Patient 19MAR2015 No Date of Administration at Home Was the Correct Dose Taken on the Right Date? Any Other Problems With Administration at Home? Subject not included in the Full Analysis Set. Page 3 of 11 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-diary.sas Output: l qs-diary-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adsad

152 Date: 04 April 2016 Page 152 of 270 Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set Country Site Germany Subject Sex Age Administerer of Drug Visit Date Was Dose Diary Brought in? M 74 Patient 22DEC2014 No Date of Administration at Home Was the Correct Dose Taken on the Right Date? Any Other Problems With Administration at Home? M 73 Patient 20JAN2015 No M 81 Patient 04FEB2015 Yes 14JAN2015 Yes F 77 Patient 10FEB2015 No Subject not included in the Full Analysis Set. Page 4 of 11 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-diary.sas Output: l qs-diary-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adsad

153 Date: 04 April 2016 Page 153 of 270 Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set Country Site Greece Subject Sex Age Administerer of Drug Visit Date Was Dose Diary Brought in? Date of Administration at Home Was the Correct Dose Taken on the Right Date? F 54 Patient 22APR2015 Yes 17MAR2015 Yes Any Other Problems With Administration at Home? F 72 Patient 29JUN2015 No M 82 Patient 21OCT2015 Yes 16SEP2015 Yes F 25 Patient 06NOV2015 No F 46 Caregiver 28MAY2015 No M 85 Caregiver 16JUL2015 No F 46 Caregiver 17JUL2015 No F 74 Caregiver 17AUG2015 No F 79 Caregiver 28AUG2015 No Subject not included in the Full Analysis Set. Page 5 of 11 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-diary.sas Output: l qs-diary-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adsad

154 Date: 04 April 2016 Page 154 of 270 Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set Country Site The Netherlands Subject Sex Age Administerer of Drug Visit Date Was Dose Diary Brought in? Date of Administration at Home Was the Correct Dose Taken on the Right Date? M 26 Patient 28OCT2014 Yes 07OCT2014 Yes 14OCT2014 Yes 21OCT2014 Yes Any Other Problems With Administration at Home? M 66 Caregiver 04NOV2014 Yes 14OCT2014 Yes 21OCT2014 Yes 28OCT2014 Yes M 49 Caregiver 17DEC2014 Yes 26NOV2014 Yes 03DEC2014 Yes 10DEC2014 Yes F 33 Patient 27JAN2015 Yes 30DEC2014 Yes 07JAN2015 Yes 14JAN2015 Yes 22JAN2015 Yes Subject not included in the Full Analysis Set. Page 6 of 11 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-diary.sas Output: l qs-diary-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adsad

155 Date: 04 April 2016 Page 155 of 270 Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set Country Site The Netherlands Subject Sex Age Administerer of Drug Visit Date Was Dose Diary Brought in? Date of Administration at Home Was the Correct Dose Taken on the Right Date? M 64 Patient 27JAN2015 Yes 06JAN2015 Yes 13JAN2015 Yes 20JAN2015 Yes Any Other Problems With Administration at Home? F 53 Patient 26FEB2015 Yes 11FEB2015 Yes 19FEB2015 Yes F 73 Patient 18JUN2015 Yes 28MAY2015 Yes 04JUN2015 Yes 11JUN2015 Yes F 28 Patient 07JUL2015 Yes 09JUN2015 Yes 15JUN2015 Yes 21JUN2015 Yes Subject not included in the Full Analysis Set. Page 7 of 11 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-diary.sas Output: l qs-diary-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adsad

156 Date: 04 April 2016 Page 156 of 270 Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set Subject Sex Age Administerer of Drug Visit Date Country Site The Netherlands Was Dose Diary Brought in? Date of Administration at Home Was the Correct Dose Taken on the Right Date? Any Other Problems With Administration at Home? F 28 Patient 07JUL2015 Yes 28JUN2015 No PER ORDER DOCTOR: SHOULD HAVE ADMINISTER 29 JUN NOT DONE BY MISTAKE BY PATIENT F 59 Caregiver 30JUN2015 Yes 16JUN2015 Yes 23JUN2015 Yes M 46 Patient 13OCT2015 Yes 29SEP2015 Yes 06OCT2015 Yes M 79 Caregiver 20OCT2015 Yes 29SEP2015 Yes 06OCT2015 Yes F 46 Patient 17NOV2015 Yes 27OCT2015 Yes 03NOV2015 Yes Subject not included in the Full Analysis Set. Page 8 of 11 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-diary.sas Output: l qs-diary-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adsad

157 Date: 04 April 2016 Page 157 of 270 Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set Country Site The Netherlands Subject Sex Age Administerer of Drug Visit Date Was Dose Diary Brought in? Date of Administration at Home Was the Correct Dose Taken on the Right Date? F 46 Patient 17NOV2015 Yes 11NOV2015 Yes Subject not included in the Full Analysis Set. Any Other Problems With Administration at Home? Page 9 of 11 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-diary.sas Output: l qs-diary-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adsad

158 Date: 04 April 2016 Page 158 of 270 Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set Country Site Spain Subject Sex Age Administerer of Drug Visit Date Was Dose Diary Brought in? F 46 Patient 04SEP2014 No Date of Administration at Home Was the Correct Dose Taken on the Right Date? Any Other Problems With Administration at Home? F 47 Patient 04SEP2014 No F 52 Patient 07JUL2014 No F 91 Caregiver 07JUL2014 No F 72 Caregiver 28JUL2014 No F 56 Patient 18SEP2015 No Subject not included in the Full Analysis Set. Page 10 of 11 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-diary.sas Output: l qs-diary-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adsad

159 Date: 04 April 2016 Page 159 of 270 Listing Listing of Home Administration Diary Data at the First Standard of Care Visit by Country, Site and Subject - Eligible Subject Set Country Site United Kingdom Subject Sex Age Administerer of Drug Visit Date Was Dose Diary Brought in? Date of Administration at Home Was the Correct Dose Taken on the Right Date? F 48 Caregiver 08JAN2015 Yes 25DEC2014 Yes Any Other Problems With Administration at Home? F 57 Patient 16FEB2015 No M 71 Patient 09MAR2015 No M 81 Patient 29APR2015 No M 25 Patient 15OCT2015 No Subject not included in the Full Analysis Set. Page 11 of 11 Program: /userdata/stat/amp2/itp/amp /analysis/final/listings/l-qs-diary.sas Output: l qs-diary-ess-l.rtf (Date generated: 02FEB2016:09:52) Source data: sdf.adsad

160 Date: 04 April 2016 Page 160 of ANNEXES

161 Date: 04 April 2016 Page 161 of 270 Annex 1. List of Stand-alone Documents

162 Date: 04 April 2016 Page 162 of 270 Region Site # Account PI Last Name PI First Name Austria Medizinische Universitaet Steurer Michael Innsbruck Belgium Algemeen Ziekenhuis Sint- Selleslag Dominik Jan Brugge-Oostende France Centre Hospitalier Viallard Jean-François Universitaire de Bordeaux - Hôpital Haut Lévêque Germany Onkologische Stauch Martina Schwerpunktpraxis Germany Charite Universitätsmedizin Nogai Axel Berlin Campus Benjamin Franklin Greece Laiko Hospital Mantzourani Marina Greece Ahepa University Hospital Kaiafa Georgia Netherlands HagaZiekenhuis, Leyenburg Schipperus Martin Spain Hospital Universitario 12 de Grande Garcia Carlos Octubre Spain Hospital Nuestra Señora de Barez Garcia Abelardo Sonsoles Spain Hospital Universitario Infanta Hernandez Rivas Jose Angel Leonor United Kingdom Royal London Hospital Taylor Louise

163 Date: 04 April 2016 Page 163 of 270 Annex 2. Study Protocol and Amendments

164 Date: 04 April 2016 Page 164 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 1 of 33 PASS Information Title Protocol version identifier , v 1.5 Date of last version of the protocol 13 December 2013 EudraCT number EU PAS Register No: Active Substance Medicinal Product Nplate Product Reference: Procedure Number: Marketing Authorisation Holder(s) Joint PASS Research Question and Objectives Countries of Study Authors A cross-sectional study of patients with immune thrombocytopenic purpura and caregivers to estimate the proportion who administer romiplostim correctly after receipt of home administration training materials ATC code B02BX04, romiplostim EU/1/08/497/ Amgen Europe B.V. No The primary objective of this study is to estimate the proportion of subjects and caregivers who administer romiplostim correctly after being trained with the HAT pack. Selection to be determined from European Union countries where romiplostim is approved and reimbursed for self or home administration Jeff Lange - langej@amgen.com Scott Stryker - sstryker@amgen.com Sally Wetten swetten@amgen.com Approved Marketing Authorisation Holder Marketing authorisation holder(s) MAH Contact Amgen Europe B.V. Minervum ZK Breda The Netherlands Scott Stryker Amgen One Amgen Center Mail Stop 24-2-A Thousand Oaks, CA 91320

165 Date: 04 April 2016 Page 165 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 2 of Table of Contents 1. Table of Contents List of Tables List of Annexes List of Abbreviations Responsible Parties Abstract Amendments and Updates Milestones Rationale and Background Research Question and Objectives Research Methods Study Design Setting Variables Data Sources Study Size Data Management Data Analysis Quality Control Source Data Integrity and Validation Validation of Statistical Analyses Limitations of the Research Methods Other Aspects Protocol Amendments and Study Termination Study Documentation and Archive Study Monitoring and Data Collection Language Compensation Protection of Human Subjects Informed Consent Institutional Review Board/Independent Ethics Committee Subject Confidentiality Management and Reporting of Adverse Events/Adverse Reactions Safety Event Definitions Adverse Event Adverse Drug Reaction Definition of Serious Adverse Event Serious Adverse Drug Reaction Other Safety Findings Approved

166 Date: 04 April 2016 Page 166 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 3 of Product Complaints Reportable Events and Reporting Timeframes Plans for Disseminating and Communicating Study Results References Annexes List of Tables Table 1. Precision in Estimate by Subjects Enrolled Table 2. Reporting Timeframes for Reportable Events List of Annexes Annex 1. List of Stand-alone Documents Annex 2. ENCePP Checklist for Study Protocols Annex 3. Sample Safety Reporting Form Annex 4. Pregnancy and Lactation Notification Worksheets Approved

167 Date: 04 April 2016 Page 167 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 4 of List of Abbreviations ADR AE EU ITP ecrf HAT IEC IRB adverse drug reaction adverse event european union immune thrombocytopenic purpura electronic case report form home administration training independent ethics committee institutional review board SADR serious adverse drug reaction SAE serious adverse event SmPC summary of product characteristics 3. Responsible Parties Scott Stryker Amgen One Amgen Center Mail Stop 24-2-A Thousand Oaks, CA Approved 4. Abstract Title: A cross-sectional study of patients with immune thrombocytopenic purpura (ITP) and caregivers to estimate the proportion who administer romiplostim correctly after receipt of home administration training materials Protocol version: , v 1.5 Protocol date: 13 December 2013 Name and affiliation of main authors: Scott Stryker (Amgen), Jeff Lange (Amgen), Sally Wetten (Amgen)

168 Date: 04 April 2016 Page 168 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 5 of 33 Rationale and Background: A variation to the Nplate Marketing Authorisation and Product Information has been approved in the European Union (EU) to allow for the possibility of administering romiplostim by patients or their lay caregivers. To mitigate risks of potential medication errors from self-administration, a Home Administration Training (HAT) pack was designed to support the training of patients. Research Question and Objectives: The primary objective of this study is to estimate the proportion of adult subjects and caregivers who administer romiplostim correctly after being trained with the HAT pack. Study Design: This cross-sectional study will involve direct observation by a healthcare professional of a series of subjects and caregivers in the act of administering romiplostim at their first standard-of-care visit 4 weeks after training with the HAT pack. Further observations, if they occur, will also be recorded in the study if made within 16 weeks of enrolment. (Additional observations are voluntary and are not required for study participation; they occur only if the healthcare professional requests them.) Population: The source population for this study starts with the estimated 3,000 patients on romiplostim therapy within the European Union. Clinical sites with clinicians known to treat ITP patients with romiplostim will be approached for possible study participation. At each participating site, patients that meet the following criteria will be enrolled as a study subject: (1) adult ITP patient, treated per EU Summary of product characteristics (SmPC), or caregiver new (or at least a 3 month gap) to administering romiplostim, (2) has received HAT pack training, (3) available at standard-of-care medical visit 4 weeks (range 2 to 8 weeks) after HAT pack training, (4) patient provides informed consent. Approved Variables: Variables for data collection include subject demographics, expected and observed injection volume per syringe, appropriate use of alcohol wipe at injection site, clinically appropriate handling of syringe to avoid contamination, and clinically appropriate technique of subcutaneous injection. Data Sources: This is a non-interventional study in which data collected derive from routine clinical care. Healthcare professionals will complete a standardized data collection form to record their direct observation of subject or caregiver administration.

169 Date: 04 April 2016 Page 169 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 6 of 33 Study Size: The target sample size is 40 subjects. With this sample size, the 95% confidence interval around an observed proportion of 90% of subjects or caregivers correctly administering romiplostim is 77% to 96%. Data Analysis: The primary endpoint will be summarized as the proportion of subjects and caregivers who administer romiplostim correctly (point estimate and a 95% confidence interval). Milestones: Start of subject enrolment in August 2014, end of data collection by June 2016, and final report by December Amendments and Updates None 6. Milestones Milestone Planned date Start of data collection August 2014 End of data collection June 2016 Final report of study results December Rationale and Background Romiplostim (Nplate ) is a peptibody (attributes of both a peptide and an antibody) designed to bind to the thrombopoietin receptor on megakaryocytes thereby stimulating platelet production. Romiplostim is indicated in Europe for adult chronic ITP in splenectomised patients who are refractory to other treatments and it may be considered as second line treatment for non-splenectomised patients where surgery is contra-indicated. Approved Typically, romiplostim is administered by a healthcare professional. Administration by a healthcare professional is a multistep process at weekly intervals, including: dose calculation based on the patient's weight and response to therapy, reconstitution of drug, measurement of volume to be injected, and subcutaneous injection. Detailed instructions of each step are specified in the Nplate Product Information. 1 A variation to the Nplate Marketing Authorisation and Product Information has been approved in the EU to allow for the possibility of self-administering romiplostim by patients or their lay caregivers. While the dose calculation will remain performed by the prescribing physician, self-administering steps completed by the patient include

170 Date: 04 April 2016 Page 170 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 7 of 33 reconstitution of drug, measurement of volume to be injected, and subcutaneous injection. Medication error is an important identified risk in the Risk Management Plan for romiplostim. A total of 62 medication error reports have been received by Amgen through 31 January 2012 during an estimated 19,270 patient-years occurring through commercial distribution of romiplostim. 2 Three of the reports had notable clinical consequences. Although none of these medication error reports involved self-administration of Nplate, an effort to mitigate the risk of medication errors is considered important in the self-administration setting. To mitigate risks of potential medication errors from self-administration, a HAT pack was designed to support the appropriate selection and training of patients by healthcare professionals. For patients, the HAT pack includes a placemat to lay out the administration components, a checklist, instructions for use, a self-administration diary and an instructional video. For healthcare professionals, the HAT pack includes a guide and checklist for patient selection and training. The healthcare professional may consider a patient for self-administration who has a stable platelet count 50 x 10 9 /L for at least 4 weeks without dose adjustment, and is considered capable for self-administration. Patients are required to undergo a period of training using the materials in the HAT pack and must demonstrate their ability to self-administer to a healthcare professional. In addition, patients must return every 4 weeks for platelet counts to be checked, and at the first such visit, are required to demonstrate that they can still self-administer correctly. Approved The HAT pack for the patients was evaluated in a human factors engineering study. 3 In this study, a total of 30 subjects (29 ITP patients and 1 caregiver) were asked to perform the process of self-administration over a 4 week period utilizing a 100 mcg kit or a 500 mcg kit. Of the 120 attempts (30 subjects x 4 weeks), the success rate of proper administration was 95%. In the 6 attempts with error, the subjects either misjudged the reference point of the plunger head (eg, assuming the middle or bottom of the plunger head represented the fill level) or counted in the wrong direction on the syringe (counting up from 0.10 milliliter to get to 0.15 milliliter instead of counting down).

171 Date: 04 April 2016 Page 171 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 8 of Research Question and Objectives This study is descriptive in nature and there is no a priori hypothesis. Data collected in this study will provide information to identify possible refinement of the HAT pack and serve as an early indicator of the potential for medication errors associated with self-administration. Primary objective: To estimate the proportion of subjects and caregivers who administer romiplostim correctly. Secondary objectives: To estimate the proportion of subjects and caregivers who reconstitute romiplostim correctly. To estimate subjects and caregivers' accuracy in administering the prescribed dose of romiplostim. To estimate the proportion of subjects and caregivers who inject romiplostim successfully. 9. Research Methods 9.1 Study Design This cross-sectional study involves direct observation by a healthcare professional of a series of subjects and caregivers in the act of administering romiplostim at their first standard-of-care visit occurring 4 weeks after training with the HAT pack. Further observations, if they occur, will also be recorded in the study if made within 16 weeks of enrolment. (Additional observations are voluntary and are not required for study participation; they occur only if the healthcare professional requests them.) Additionally, data will be collected from the subjects dose diary at the first standard of care visit to ensure there were no problems with administration while not at the clinic. The strength of this study design is that it will describe the absolute risk of medication error with romiplostim self-administration by an approach that is reliable (direct observation by trained healthcare professional), 4 and of little burden to patient and healthcare professional. Approved Primary Endpoint: A yes/no indicator for whether the subject or caregiver administers romiplostim correctly at the 4 week visit. Indicator will be "yes" if: the difference between the prescribed and administered dose of romiplostim is within ± 10%;

172 Date: 04 April 2016 Page 172 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 9 of 33 and the healthcare professional observes that the subject or caregiver reconstitutes romiplostim correctly; and successfully injects romiplostim as described in the secondary endpoint; and if the healthcare professional does not have to intervene to correct an error during administration by subject or caregiver (eg, volume in syringe). The 10% margin of error for dose is consistent with a published definition of medication error for an injectable product. 4,5 Indicator will be "no" if any of criteria above are not met. Secondary Endpoints: A yes/no indicator for whether the subject or caregiver reconstitutes romiplostim correctly at the 4 week visit / at follow-up visits. The indicator will be yes if: o the healthcare professional observes that aseptic techniques were used in preparing the vial; and all the water from the sterile water syringe was injected gently into the vial; and all romiplostim was dissolved. o Indicator will be "no" if any of these criteria are not met. For accuracy in administering the prescribed dose of romiplostim, the difference between the prescribed and administered dose of romiplostim, expressed as a %. A yes/no indicator for whether the subject or caregiver injects romiplostim successfully at the 4 week visit / at follow-up visits. The indicator will be yes if: o The healthcare professional observes that all air bubbles were removed from syringe; and the patient made clinically appropriate use of alcohol wipe at injection site; and clinically appropriate handling of syringe to avoid contamination; and clinically appropriate technique of subcutaneous injection. o Indicator will be "no" if any of these criteria are not met. A yes/no indicator for whether the subject or caregiver administers romiplostim correctly at follow-up visits. The indicator will be yes or no as described in the primary endpoint. Approved 9.2 Setting The source population for this study starts with the estimated 3,000 patients (based on sales data) currently compliant to romiplostim therapy within the EU. Market research predicts approximately 10% of these patients may be eligible and may select self-administration of romiplostim. However, self-administration may not be available in all countries, depending on the reimbursement status in each country, further reducing the number of total patients exposed to self-administration. Hence for this study, the source population of adult ITP patients who initiate romiplostim self-administration is likely to range in the hundreds. To identify study subjects within this source population, clinical sites with clinicians known to treat ITP patients with romiplostim will be approached for possible participation

173 Date: 04 April 2016 Page 173 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 10 of 33 in the study. Sites within each country will become eligible for participation after launch of the HAT pack for self-administration at the country level. The availability of the HAT pack will be staggered at the country level from March 2013 onwards. Because of a small number of patients in the source population and the staggered launch of the HAT pack, it is anticipated that recruitment of subjects to fulfill the target sample size of 40 subjects may take up to 18 months. Sites and countries will be selected based on availability and willingness to participate; no minimum number of patients treated with romiplostim, use of self-administration, or other criteria will be applied. At each participating site, patients who meet the following criteria will be enrolled as a study subject: adult ( 18 years of age) ITP patient, treated per EU SmPC, or caregiver new (or at least a 3 month gap) to administering romiplostim has received HAT pack training available at standard-of-care medical visit 4 weeks (range 2 to 8 weeks) after HAT pack training patient provides informed consent The enrolling healthcare professional will ask the patient at start of visit (before HAT pack training) of their willingness to participate in this observational study, and obtain informed consent. Enrolment onto the study will be determined after all eligibility criteria have been met. See Section 10.1 for procedure of informed consent. Consecutive eligible subjects at each site will be enrolled into the study (ie, non-random sampling from source population) with direct observation of the subject or caregiver administering at the 4 week visit after HAT pack training. To obtain the proportion of patient participation, each enrolling site will also provide the number of eligible subjects and caregivers offered study participation but who refused. Approved Further observations, if they occur, will also be recorded in the study if made within 16 weeks of enrolment. (Additional observations are voluntary and are not required for study participation; they occur only if the healthcare professional requests them.) 9.3 Variables Data for the following variables will be collected by using a standardized data collection form that will be filled out by the observing healthcare professional. patient demographics - sex (male or female), age (years) administerer of drug - (subject or caregiver)

174 Date: 04 April 2016 Page 174 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 11 of 33 dates - date of initial prescription for self-administration, date of completion of HAT pack training, dates of observation reason for follow-up visit (if applicable) (text) healthcare professional observing - clinical site, provider type (clinician, nurse, other) healthcare professional observing was same person as trainer (yes/no) number of vials of romiplostim needed for single administration (1 or 2) injection volume per syringe prescribed by physician - milliliter vial size of romiplostim - (250 or 500 microgram kit) aseptic techniques were used in preparing the vial (yes/no) gently injected all water from sterile water syringe into the vial (yes/no) ensured all romiplostim dissolved (yes/no) ensured all air bubbles removed from injection syringe (yes/no) injection volume per syringe, observed - milliliter clinically appropriate use of alcohol wipe at injection site - (yes/no) clinically appropriate handling of syringe to avoid contamination - (yes/no) clinically appropriate technique of subcutaneous injection - (yes/no) healthcare professional intervened to correct an error during administration by subject or caregiver - (yes/no) description of error - (narrative data of observer) self-administration diary brought in (yes/no) date of administration at home correct dose administered at home (yes/no) any other problems with administration at home (yes/no) description of problem, if yes (narrative data) Approved The following variables will be calculated from some of the data variables above. romiplostim dose administered by subject or caregiver - micrograms; calculated by observed injection volume (milliliter) * 500 (micrograms / milliliter) romiplostim dose prescribed by physician - micrograms; calculated by prescribed injection volume (milliliter) * 500 (micrograms / milliliter) 9.4 Data Sources This is a non-interventional study in which data are collected as a part of routine clinical care. The healthcare professional will complete a standardized data collection form to record their direct observation of subject or caregiver administering romiplostim.

175 Date: 04 April 2016 Page 175 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 12 of Study Size The sample size should provide an adequate precision of the confidence interval around the central point estimate and be feasible for achieving the study enrolment with real-world patients in an orphan indication in a reasonable time period. For an expected result of at least 80% subjects and caregivers administering romiplostim correctly, the precision of the 95% confidence interval varies little between a sample size of 40 and a sample size of 60 (see Table 1). Hence for this study, the target sample size is 40 subjects. Table 1. Precision in Estimate by Subjects Enrolled. Subjects enrolled Upper confidence bounds on proportion of subjects not selfadministering correctly if 0 errors are observed* Precision in estimate if 90% of subjects selfadminister correctly** Precision in estimate if 80% of subjects selfadminister correctly** 20 15% 70% - 97% 58% - 92% % 77% - 96% 65% - 90% 60 5% 80% - 95% 68% - 88% * The upper confidence bounds in a sample of size N is 3/N 6 ** Precision is based on the 95% confidence interval for a proportion using Wilson s score method (no continuity correction) 7, Data Management Data can be entered onto a standardized data collection form by the healthcare professional and then transferred into the electronic case report form (ecrf) for each visit. All sections of the ecrf should be completed as soon as possible after the observation. Approved 9.7 Data Analysis The data analysis for this study will be descriptive in nature. Endpoints of a binary nature (yes/no) will be summarized as the proportion (percentage). As a measure of precision, a 95% confidence interval (binomial exact) will be calculated around the proportion. Endpoints of a continuous nature will be described by the mean, standard deviation, range and median. Adverse event (AE) information received through this study (see Section 11) will be processed and evaluated as a part of the routine case management and safety signal detection and assessment processes.

176 Date: 04 April 2016 Page 176 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 13 of Quality Control Source Data Integrity and Validation All healthcare professionals who will record their direct observations in this study will be provided with training on the various types of errors that could occur in self-administration. This training should help improve consistency in recognition of errors across study sites. Each variable will be checked for missing data and nonsensical outliers. Missing data or outliers will be checked with observing healthcare professional Validation of Statistical Analyses Programs will be developed and maintained, and output will be verified in accordance with current risk-based quality control procedures. Outputs will be produced with validated standard macro programs where standard macros can produce the specified outputs. The production environment for statistical analyses consists of Amgensupported versions of statistical analysis software (ie, SAS System). 9.9 Limitations of the Research Methods Generalizability of study results may be limited by use of a study population based on a convenience (ie, not random) sample of patients. Because ITP is a rare disease, only a subgroup of patients receive romiplostim, and a smaller subset of those will be evaluated and approved by their prescribers for self-administration; the source population of patients will be small and may perhaps be widely dispersed or concentrated in specific institutions. Given the small source population likely in the range of hundreds, the target sample size of 40 will at least reflect a sizable portion of source population. Another potential limitation in generalizability of study results may occur if only a subset of eligible subjects actually participate in study. To assess for this bias, a 'cooperation proportion' for clinical sites and subjects will be tallied by the number of clinical sites and subjects participating in the study divided by number offered inclusion. Approved Another possible limitation is the direct observation at first standard-of-care visit may not reflect the subject or caregivers s ability to administer romiplostim between clinical visits when the act of administration is not observed. To assess this bias, data from the selfadministration diary is being collected. The observers could be biased by preconceived expectations of what they should observe or the subjects could act differently when being watched (ie, Hawthorne effect 9 ). Direct observation as performed by different observers may add a level in inter-observer variability.

177 Date: 04 April 2016 Page 177 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 14 of Other Aspects Protocol Amendments and Study Termination Amgen may amend the protocol at any time. If Amgen amends the protocol, written agreement from the investigator may need to be obtained where applicable per local governing laws or regulations. The institutional review board/independent ethics committee (IRB/IEC) must be informed of all amendments and give approval. The investigator must send a copy of the approval letter from the IRB/IEC to Amgen. Amgen reserves the right to terminate the study at any time. Both Amgen and the investigator reserve the right to terminate the investigator s participation in the study according to the contractual agreement. The investigator is to notify the IRB/IEC in writing of the study s completion or early termination and send a copy of the notification to Amgen Study Documentation and Archive The investigator should maintain a list of appropriately qualified persons to whom he or she has delegated study duties. All persons authorized to make entries or corrections on the ecrfs will be included on the Amgen Delegation of Authority Form. ecrf entries may be considered source data if the ecrf is the site of the original recording (ie, there is no other written or electronic record of data). In this study, we expect the standardized data collection form will be the source document. The investigator and study staff are responsible for maintaining a comprehensive and centralized filing system of all study-related documentation, suitable for inspection at any time by representatives from Amgen or applicable regulatory authorities. Documentation should include: Approved Informed consent forms and subject identification list Study files containing the protocol with all amendments, and all correspondence to and from the IRB/IEC and Amgen Retention of study documents will be governed by the contractual agreement with Amgen Study Monitoring and Data Collection The Amgen representative and regulatory authority inspectors are responsible for contacting and visiting the investigator for the purpose of inspecting the facilities and, upon request, inspecting the various records of the study provided that subject confidentiality is respected.

178 Date: 04 April 2016 Page 178 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 15 of 33 The Clinical Monitor or designee is responsible for verifying the ecrfs at regular intervals throughout the study to verify adherence to the protocol completeness, accuracy, and consistency of the data; and adherence to local regulations on the conduct of research. The Clinical Monitor, or designee is to have access to subject medical records and other study-related records needed to verify the entries on the ecrfs in accordance with the local laws and regulations. The Investigator agrees to cooperate with the Clinical Monitor, or designee to ensure that any problems detected in the course of these monitoring visits, including delays in completing ecrfs, are resolved. In accordance with the sponsor s audit plans, this study may be selected for audit by representatives from Amgen s Global Compliance Auditing function (or designees). Review of study-related records will occur to evaluate the study conduct and compliance with the protocol, and applicable regulatory requirements. To ensure the quality of clinical data across all subjects and sites, a clinical data management review is performed on subject data received at Amgen. During this review, subject data is checked for consistency, omissions, and any apparent discrepancies. In addition, the data is reviewed for adherence to the protocol. To resolve any questions arising from the clinical data management review process, query notifications will be raised within the ecrf for completion by the site Language All written information and other material to be used by subjects and investigative staff must use vocabulary and language that are clearly understood. Approved Compensation No compensation will be provided to subjects. 10. Protection of Human Subjects 10.1 Informed Consent An informed consent form template will be provided by Amgen for the investigator to prepare the informed consent document to be used at his or her site. Updates to the template are to be communicated formally in writing from the Amgen Clinical Study Manager to the investigator. The written informed consent document is to be prepared in the language of the potential subject population.

179 Date: 04 April 2016 Page 179 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 16 of 33 Before a subject s participation in the study, the investigator is responsible for obtaining written informed consent, where applicable by local regulations, from the subject after adequate explanation of the aims, methods, anticipated benefits, and potential hazards of the study and before any protocol-specific observations are conducted. The acquisition of informed consent is to be documented in the medical records, and the informed consent form is to be signed and personally dated by the subject or a legally acceptable representative and by the person who conducted the informed consent discussion. The original signed informed consent form is to be retained in accordance with institutional policy, and a copy of the signed consent form is to be provided to the subject or a legally acceptable representative. If a potential subject is illiterate or visually impaired and does not have a legally acceptable representative, the investigator must provide an impartial witness to read the informed consent form to the subject and must allow for questions. Thereafter, both the subject and the witness must sign the informed consent form to attest that informed consent was freely given and understood Institutional Review Board/Independent Ethics Committee A copy of the protocol, proposed informed consent form, other written subject information, and any proposed advertising material must be submitted to the IRB/IEC for written approval. A copy of the written approval of the protocol and informed consent form must be received by Amgen before study can be executed. The investigator must submit and, where necessary, obtain approval from the IRB/IEC for all subsequent protocol amendments and changes to the informed consent document. The investigator is to notify the IRB/IEC of deviations from the protocol or serious adverse event(s) (SAE) occurring at the site and other adverse event reports received from Amgen, in accordance with local procedures. The investigator is responsible for obtaining annual IRB/IEC approval/renewal throughout the duration of the study. Copies of the investigator s reports, where applicable by local regulations and the IRB/IEC or other relevant ethical review board continuance of approval must be sent to Amgen. Approved 10.3 Subject Confidentiality The investigator must ensure that the subjects confidentiality is maintained for documents submitted to Amgen. Subjects are to be identified by a unique identification number. Where permitted, date of birth is to be documented and formatted in accordance with local laws and regulations. On the CRFs demographics page, in addition to the unique identification number, include the age in years at time of enrollment. For SAEs reported to Amgen, subjects are to be identified by their unique

180 Date: 04 April 2016 Page 180 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 17 of 33 subject identification number, initials, and date of birth (in accordance with local laws and regulations). Documents that are not for submission to Amgen (eg, signed informed consent forms) are to be kept in confidence by the investigator, except as described below. In compliance with local country regulations, it is required that the Investigator and institution permit authorized representatives of the company, of the regulatory agency(s), and the IRB/IEC direct access to review the subject's original medical records for verification of study-related activities and data. Direct access includes examining, analyzing, verifying, and reproducing any records and reports that are important to the evaluation of the study. The investigator is obligated to inform and obtain the consent of the subject to permit such individuals to have access to his/her study-related records, including personal information. 11. Management and Reporting of Adverse Events/Adverse Reactions General information regarding reporting of adverse drug reactions (ADRs) and/or serious adverse drug reactions (SADRs): Report only ADRs and/or SADRs, other safety findings, or product complaints involving Amgen products Do not report ADRs and/or SADRs, other safety findings, or product complaints that occurred prior to a subject taking an Amgen product 11.1 Safety Event Definitions Adverse Event An AE is any untoward medical occurrence in a subject administered a pharmaceutical product(s) and which does not necessarily have to have a causal relationship with this treatment. Approved An AE can therefore be any unfavorable and unintended sign (including an abnormal laboratory finding, for example), symptom, or disease temporally associated with the use of a product(s), whether or not considered related to the product(s). The definition of an AE includes: Worsening of a pre-existing condition Events occurring from a medication error or overdose of a product(s), whether accidental or intentional Events occurring from abuse of a product(s) Events associated with the discontinuation of the use of a product(s), (eg, appearance of new symptoms) Any lack or loss of intended effect of the product(s)

181 Date: 04 April 2016 Page 181 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 18 of 33 An adverse device effect is any AE related to the use of a medical device. Adverse device effects include AEs resulting from insufficient or inadequate instructions for use, AEs resulting from any malfunction of the device, or adverse events resulting from use error or from intentional misuse of the device Adverse Drug Reaction AEs that are considered related to the Amgen product(s) are classified as ADRs. It is the Investigator s responsibility to evaluate if an event is related to an Amgen product prior to reporting the event to Amgen Definition of Serious Adverse Event A SAE is any AE as defined above that also: is fatal is life threatening (places the subject/patient at immediate risk of death) requires in-patient hospitalization or prolongation of existing hospitalization results in persistent or significant disability/incapacity is a congenital anomaly/birth defect is an other significant medical hazard that does not meet any of the above criteria A hospitalization meeting the regulatory definition for serious is any in-patient hospital admission that includes a minimum of an overnight stay in a healthcare facility. Other significant medical hazards refer to important medical events that may not be immediately life-threatening or result in death or hospitalization, but may jeopardize the subject or may require intervention to prevent one of the other outcomes listed in the definition above. Examples of such events could include allergic bronchospasm, convulsions, and blood dyscrasias, drug-induced liver injury, events that necessitate an emergency room visit, outpatient surgery, or other events that require other urgent intervention. Approved Serious Adverse Drug Reaction SAEs that are considered related to the Amgen product(s) are classified as SADRs. It is the Investigator s responsibility to evaluate if an event is related to an Amgen product prior to reporting the event to Amgen.

182 Date: 04 April 2016 Page 182 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 19 of Other Safety Findings Other Safety Findings include: Medication errors, overdose, misuse, or abuse, whether accidental or intentional, involving an Amgen product, regardless of whether associated with an ADR and/or SADR Pregnancy and lactation exposure regardless of whether associated with an ADR and/or SADR Transmission of infectious agents regardless of whether associated with an ADR and/or SADR Reports of uses outside the terms for authorized use of the product including off label use when associated with an ADR and/or SADR Product Complaints Product Complaints include any written, electronic or oral communication that alleges deficiencies related to the identity, quality, durability, reliability, safety, effectiveness, or performance of a product or device after it is released for distribution. This includes all components distributed with the product(s) such as packaging, product containers, delivery system, labeling, inserts, etc. Product Complaints may include but are not limited to issues related to: Appearance (eg, broken, cracks, color, particles, odor) Labeling (eg, missing, torn, smudged) Durability (eg, stability issues) Open packaging Device damage (eg, pre-filled syringe with bent needle) Inability of customer to understand product labeling Inability of customer to deliver the product successfully, including partial or incomplete delivery (eg, defective delivery system [syringe]) Approved 11.2 Reportable Events and Reporting Timeframes The Investigator is responsible for ensuring that all ADRs, SADRs, other safety findings, and product complaints for Amgen product(s) observed by the Investigator or reported by the subject that occur during the pre-determined data collection period, or enrollment, or after signing of the informed consent through the final study visit are recorded in the subject s medical record and are submitted to Amgen via the supplied Amgen Safety Reporting Form. See Annex 3 for a sample Safety Reporting Form and Annex 4 for sample Pregnancy and Lactation Notification Worksheets. Refer to Table 2 for the reporting timeframes for reportable events.

183 Date: 04 April 2016 Page 183 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 20 of 33 Table 2. Reporting Timeframes for Reportable Events Report Type Description Reporting Timeframe SADR and any associated Other Safety Findings Initial or follow-up reports Within 1 business day of investigator s awareness ADR that are not serious and any associated Other Safety Findings Other Safety Findings NOT associated with ADR or SADR Product complaints Pregnancy and/or Lactation Initial or follow-up reports Within 60 calendar days of investigator s awareness Initial or follow-up reports Within 1 business day of investigator s awareness Initial or follow-up reports Within 1 business day of investigator s awareness Initial or follow-up reports of pregnancies or lactation occurring in females while taking Amgen product(s) and/or Initial or follow-up reports of pregnancies or lactation occurring in female partners of males taking Amgen product(s) Within 1 business day of investigator s awareness The Investigator may be asked to provide additional information for any event submitted, which may include a discharge summary or extracts from the medical record. Information provided about the event must be consistent with information recorded on study ecrfs where safety data may also be recorded (eg, AE Summary CRF). Approved The Investigator is responsible for medical management of subjects who experience AE from the date of awareness to resolution or stabilization. The investigator is to provide an event listing to Amgen for purposes of reconciliation with the safety database per the contractual agreement. Amgen will report ADRs and SADRs as required to regulatory authorities, Investigators/institutions, and IRBs/IECs or other relevant ethical review board in accordance with Pharmacovigilance guidelines and in compliance with local regulations.

184 Date: 04 April 2016 Page 184 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 21 of 33 The Investigator is to notify the appropriate Institutional Review Board/Independent Ethics Committee IRB/IEC or other relevant ethical review board of SADRs occurring at the site and other AE reports received from Amgen, in accordance with local procedures and statutes. 12. Plans for Disseminating and Communicating Study Results The protocol and an abstract of results will be posted as per guidelines for studies meeting the criteria for post authorization safety studies. Authorship of any publications resulting from this study will be determined on the basis of the Uniform Requirement for Manuscripts Submitted to Biomedical Journals, which states: Authorship credit should be based on (1) substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; (2) drafting the article or revising it critically for important intellectual content; (3) final approval of the version to be published. Authors should meet conditions 1, 2, and 3. When a large, multicenter group has conducted the work, the group should identify the individuals who accept direct responsibility for the manuscript. These individuals should fully meet the criteria for authorship defined above. Acquisition of funding, collection of data, or general supervision of the research group, alone, does not justify authorship. All persons designated as authors should qualify for authorship, and all those who qualify should be listed. Each author should have participated sufficiently in the work to take public responsibility for appropriate portions of the content. Approved

185 Date: 04 April 2016 Page 185 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 22 of References 1 Amgen. (2012). Nplate (romiplostim), for subcutaneous injection. Highlights of prescribing information. Retrieved from 2 Safety Assessment Report. Medication errors in association with the use of romiplostim. Romiplostim Global Safety Team. 24 July On file at Amgen, Thousand Oaks, CA. 3 Human Factor Engineering Study. Nplate self-administration kit validation study. 18 April On file at Amgen, Thousand Oaks, CA. 4 National Research Council. Preventing Medication Errors: Quality Chasm Series. Washington, DC: The National Academies Press, Barker KN, Flynn EA, Pepper GA, Bates DW, Mikeal RL. Medication errors observed in 36 health care facilities. Arch Intern Med. 2002;162: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version [updated September 2009]. The Cochrane Collaboration, Available from 7 Newcombe R. Two-Sided Confidence Intervals for the Single Proportion: Comparison of Seven Methods. Stat Med. 1998;17: Wilson EB. Probable Inference, the Law of Succession, and Statistical Inference. J Am Stat Assoc. 1927;22: McCarney R, Warner J, Iliffe S, can Haselen R, Giffin M, Fisher P. The Hawthorne Effect: a randomized, controlled trial. BMC Med Res Methodol. 2007;7:30. Approved

186 Date: 04 April 2016 Page 186 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 23 of Annexes Approved

187 Date: 04 April 2016 Page 187 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 24 of 33 None. Annex 1. List of Stand-alone Documents Approved

188 Date: 04 April 2016 Page 188 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 25 of 33 Annex 2. ENCePP Checklist for Study Protocols Section 1: Research question Yes No N/A Page 1.1 Does the formulation of the research question clearly explain: Why the study is conducted? (eg, to address an important public health concern, a risk identified in the risk management plan, an emerging safety issue) The objectives of the study? 1.2 Does the formulation of the research question specify: The target population? (ie, population or subgroup to whom the study results are intended to be generalised) Which formal hypothesis(-es) is (are) to be tested? if applicable, that there is no a priori hypothesis? Comments: Section 2: Source and study populations Yes No N/A Page 2.1 Is the source population described? 2.2 Is the planned study population defined in terms of: Study time period? Age and sex? Country of origin? Disease/indication? Co-morbidity? Seasonality? Approved 2.3 Does the protocol define how the study population will be sampled from the source population? (eg, event or inclusion/exclusion criteria) Comments:

189 Date: 04 April 2016 Page 189 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 26 of 33 Section 3: Study design Yes No N/A Page 3.1 Does the protocol specify the primary and secondary (if applicable) endpoint(s) to be investigated? 3.2 Is the study design described? (eg, cohort, casecontrol, randomised controlled trial, new or alternative design) 3.3 Does the protocol describe the measure(s) of effect? (eg, relative risk, odds ratio, deaths per 1000 person-years, absolute risk, excess risk, incidence rate ratio, hazard ratio, number needed to harm (NNH) per year) 3.4 Is sample size considered? 3.5 Is statistical power calculated? Comments: Section 4: Data sources Yes No N/A Page 4.1 Does the protocol describe the data source(s) used in the study for the ascertainment of: Exposure? (eg, pharmacy dispensing, general practice prescribing, claims data, selfreport, face-to-face interview, etc) Endpoints? (eg, clinical records, laboratory markers or values, claims data, self-report, patient interview including scales and questionnaires, vital statistics, etc) Covariates? 4.2 Does the protocol describe the information available from the data source(s) on: Exposure? (eg, date of dispensing, drug quantity, dose, number of days of supply prescription, daily dosage, prescriber) Endpoints? (eg, date of occurrence, multiple event, severity measures related to event) Covariates? (eg, age, sex, clinical and drug use history, co-morbidity, co-medications, life style, etc.) 4.3 Is the coding system described for: Diseases? (eg, International Classification of Diseases (ICD)-10) Approved

190 Date: 04 April 2016 Page 190 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 27 of 33 Section 4: Data sources Yes No N/A Page Endpoints? (eg, Medical Dictionary for Regulatory Activities(MedDRA) for adverse events) Exposure? (eg, WHO Drug Dictionary, Anatomical Therapeutic Chemical (ATC)Classification System) 4.4 Is the linkage method between data sources described? (eg, based on a unique identifier or other) Comments: Section 5: Exposure definition and measurement Yes No N/A Page 5.1 Does the protocol describe how exposure is defined and measured? (eg, operational details for defining and categorising exposure) 5.2 Does the protocol discuss the validity of exposure measurement? (eg, precision, accuracy, prospective ascertainment, exposure information recorded before the outcome occurred, use of validation sub-study) 5.3 Is exposure classified according to time windows? (eg, current user, former user, non-use) 5.4 Is exposure classified based on biological mechanism of action? 5.5 Does the protocol specify whether a dosedependent or duration-dependent response is measured? Approved Comments: Section 6: Endpoint definition and measurement Yes No N/A Page 6.1 Does the protocol describe how the endpoints are defined and measured? 6.2 Does the protocol discuss the validity of endpoint measurement? (eg, precision, accuracy, sensitivity, specificity, positive predictive value, prospective or retrospective ascertainment, use of validation sub-study) Comments:

191 Date: 04 April 2016 Page 191 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 28 of 33 Section 7: Biases and Effect modifiers Yes No N/A Page 7.1 Does the protocol address: Selection biases? Information biases? (eg, anticipated direction and magnitude of such biases, validation sub-study, use of validation and external data, analytical methods) 7.2 Does the protocol address known confounders? (eg, collection of data on known confounders, methods of controlling for known confounders) 7.3 Does the protocol address known effect modifiers? (eg, collection of data on known effect modifiers, anticipated direction of effect) 7.4 Does the protocol address other limitations? Comments: Generilizability is discussed. Section 8: Analysis plan Yes No N/A Page 8.1 Does the plan include measurement of absolute effects? 8.2 Is the choice of statistical techniques described? 8.3 Are descriptive analyses included? 8.4 Are stratified analyses included? 8.5 Does the plan describe the methods for identifying: Confounders? Effect modifiers? 8.6 Does the plan describe how the analysis will address: Confounding? Effect modification? Approved Comments: Section 9: Quality assurance, feasibility, reporting Yes No N/A Page 9.1 Does the protocol provide information on data storage? (eg, software and IT environment, database maintenance and anti-fraud protection, archiving) 9.2 Are methods of quality assurance described? 9.3 Does the protocol describe quality issues related to the data source(s)?

192 Date: 04 April 2016 Page 192 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 29 of 33 Section 9: Quality assurance, feasibility, reporting Yes No N/A Page 9.4 Does the protocol discuss study feasibility? (eg, sample size, anticipated exposure, duration of follow-up in a cohort study, patient recruitment) 9.5 Does the protocol specify timelines for Start of data collection? Any progress report? End of data collection? Reporting? (ie, interim reports, final study report) 9.6 Does the protocol include a section to document future amendments and deviations? 9.7 Are communication methods to disseminate results described? 9.8 Is there a system in place for independent review of study results? Comments: Section 10: Ethical issues Yes No N/A Page 10.1 Have requirements of Ethics Committee/Institutional Review Board approval been described? 10.2 Has any outcome of an ethical review procedure been addressed? 10.3 Have data protection requirements been described? Approved Comments: Name of main author of study protocol: Date: / / Signature:

193 Date: 04 April 2016 Page 193 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 30 of 33 Annex 3. Sample Safety Reporting Form Approved

194 Date: 04 April 2016 Page 194 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 31 of 33 Approved

195 Date: 04 April 2016 Page 195 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 32 of 33 Annex 4. Pregnancy and Lactation Notification Worksheets Approved

196 Date: 04 April 2016 Page 196 of 270 Product: Nplate Protocol Number: Date: 13 December 2013 Page 33 of 33 Approved

197 Date: 04 April 2016 Page 197 of 270 Page 1 1. PRAC Outcome 1.1 PRAC Rapporteur assessment In summary, this cross sectional study proposed by the MAH could be considered appropriate to address the study objectives. The MAH is encouraged to initiate the study as soon as possible. Some suggestions are proposed to be addressed in the next RMP. Particularly, to maintain eligibility criteria included in previous version of the protocol (Section 9.2 Setting p.10): The enrolling healthcare professional will ask the patient at start of visit of their willingness to participate in this observational study ( ), instead of: The enrolling healthcare professional will ask the patient after they have completed the HAT pack training for their willingness to participate ( ). Changes to protocol implemented: Old Text New Text Rationale for Change Section 4 Abstract Section 4 Abstract To align the text with what is written in Section 9.2 At each participating site, clinicians will select patients that meet the following criteria for inclusion as a study subject: (...) Section 9.2 Setting At each participating site, patients who meet the following criteria will be included as a study subject The enrolling healthcare professional will ask the patient after they have completed the HAT pack training for their willingness to participate in this observational study of normal clinical practice. At each participating site, patients that meet the following criteria will be enrolled as a study subject: (...) Section Setting At each participating site, patients who meet the following criteria will be enrolled as a study subject The enrolling healthcare professional will ask the patient at start of visit (before HAT pack training) of their willingness to participate in this observational study, and obtain informed consent. Enrolment onto the study will be determined after all eligibility criteria have been met. To clarify the definition and timing of enrolment. To clarify the type of visit at which selection is to take place and address the PRAC s concerns about selection bias. Approved

198 Date: 04 April 2016 Page 198 of 270 Annex 3. Signature of Coordinating Investigator

199 Date: 04 April 2016 Page 199 of 270 Investigator Signature STUDY NUMBER: STUDY REPORT TITLE: A Cross-sectional Study of Patients With Immune Thrombocytopenic Purpura and Caregivers to Estimate the Proportion Who Administer Romiplostim Correctly After Receipt of Home Administration Training Materials I have read the above named Observational Research Study Report and signify my agreement with the overall conclusions. Name of Coordinating Investigator: Institution: Signature of Investigator: Date:

200 Date: 04 April 2016 Page 200 of 270 Annex 4. Statistical Analysis Plan

201 Date: 04 April 2016 Page 201 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 1 STATISTICAL ANALYSIS PLAN FOR OBSERVATIONAL STUDIES A cross-sectional study of patients with immune thrombocytopenic purpura and caregivers to estimate the proportion who administer romiplostim correctly after receipt of home administration training materials Protocol Number: Version: 1.0 Date: 4 March 2014 Authors: James Bennett

202 Date: 04 April 2016 Page 202 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 2 Table of Contents Page TABLE OF ABBREVIATIONS INTRODUCTION OBJECTIVES Primary objective Secondary objectives STUDY OVERVIEW Study Design Data Source Sample Size STUDY ENDPOINTS Primary Endpoint Secondary Endpoints HYPOTHESES OR ESTIMATION DEFINITIONS General Study-level Definitions Time Windows Derivations ANALYSIS SUBSETS Primary Analysis Set Additional Analysis Sets DATA SCREENING AND ACCEPTANCE General Principles Data Handling and Electronic Transfer of Data Handling of Missing and Incomplete Data Detection of Bias Outliers Distributional Characteristics Validation of Statistical Analyses STATISTICAL METHODS OF ANALYSIS General Principles Subject/Patient Accountability Demographic and Baseline Characteristics Analyses of Primary Endpoint Analyses of Secondary Endpoints... 13

203 Date: 04 April 2016 Page 203 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page Safety Analyses Adverse Drug Reactions Exposure to Product CHANGES FROM PROTOCOL OR RPP SPECIFIED ANALYSES LITERATURE CITATIONS / REFERENCES LIST OF PLANNED TABLES, FIGURES AND LISTINGS Planned Tables Planned Listings List of Tables Table 1. Precision in Estimate by Subjects Enrolled.... 7

204 Date: 04 April 2016 Page 204 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 4 TABLE OF ABBREVIATIONS Abbreviation or Term ADR CI EDC EU FAS GSIB HAT HCP ITP MedDRA SAP SmPC TFLSD Defination/Explaination Adverse Drug Reaction Confidence Interval Electronic Data Capture European Union Full Analysis Set Global Safety & Independent Biostatistics Home Administration Training Healthcare Professional Immune Thrombocytopenic Purpura Medical Dictionary for Regulatory Activities Statistical Analysis Plan Summary of Product Characteristics Tables, Figures and Listings Shell Document

205 Date: 04 April 2016 Page 205 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 5 1. INTRODUCTION The purpose of this statistical analysis plan (SAP) is to provide details of the statistical analyses that have been outlined within the Protocol for the NPLATE (romiplostim) study dated 13 December 2013 (v 1.5). This is a cross-sectional study of patients with immune thrombocytopenic purpura (ITP) and caregivers to estimate the proportion who administer romiplostim correctly after receipt of home administration training (HAT) materials. The scope of this document includes the final analyses that are planned and will be executed by the Global Safety and Independent Biostatistics (GSIB) department. 2. OBJECTIVES This study is descriptive in nature and there is no a priori hypothesis. Data collected in this study will provide information to identify possible refinement of the HAT pack and serve as an early indicator of the potential for medication errors associated with self-administration. 2.1 Primary objective To estimate the proportion of subjects and caregivers who administer romiplostim correctly. 2.2 Secondary objectives To estimate the proportion of subjects and caregivers who reconstitute romiplostim correctly. To estimate subjects and caregivers' accuracy in administering the prescribed dose of romiplostim. To estimate the proportion of subjects and caregivers who inject romiplostim successfully. 3. STUDY OVERVIEW 3.1 Study Design This cross-sectional study involves direct observation by healthcare professionals (HCPs) of a series of subjects and caregivers in the act of administering romiplostim at their first standard-of-care visit occurring 4 weeks after training with the HAT pack. Further observations, if they occur, will also be recorded in the study if made within 16 weeks of enrollment. (Additional observations are voluntary and are not required for study participation; they occur only if the HCP requests them). Additionally, data will be collected from the subjects dose diaries at the first standard of care visit to record any problems with administration while not at the clinic. The strength of this study design is that it will describe the risk of medication error with romiplostim self-administration by an

206 Date: 04 April 2016 Page 206 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 6 approach that is reliable (direct observation by trained HCP), and of little burden to patient and HCP. 3.2 Data Source The source population for this study starts with the estimated 3,000 patients (based on sales data) currently compliant to romiplostim therapy within the European Union (EU). Market research predicts approximately 10% of these patients may be eligible and may select self-administration of romiplostim. However, self-administration may not be available in all countries, depending on the reimbursement status in each country, further reducing the number of total patients exposed to self-administration. Hence for this study, the source population of adult ITP patients who initiate romiplostim self-administration is likely to range in the hundreds. To identify study subjects within this source population, clinical sites with clinicians known to treat ITP patients with romiplostim will be approached for possible participation in the study. Sites within each country will become eligible for participation after launch of the HAT pack for self-administration at the country level. The availability of the HAT pack will be staggered at the country level from March 2013 onwards. Because of a small number of patients in the source population and the staggered launch of the HAT pack, it is anticipated that recruitment of subjects to fulfill the target sample size may take up to 18 months. Sites and countries will be selected based on availability and willingness to participate; no minimum number of patients treated with romiplostim, use of self-administration, or other criteria will be applied. At each participating site, patients who meet the following criteria will be enrolled as a study subject: adult ( 18 years of age) ITP patient, treated per EU Summary of Product Characteristics (SmPC), or caregiver new (or at least a 3 month gap) to administering romiplostim. has received HAT pack training. available at standard-of-care medical visit 4 weeks (range 2 to 8 weeks) after HAT pack training. patient provides informed consent. 3.3 Sample Size The sample size should provide an adequate precision of the confidence interval (CI) around the central point estimate and be feasible for achieving the study enrolment with real-world patients in an orphan indication in a reasonable time period. For an expected

207 Date: 04 April 2016 Page 207 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 7 result of at least 80% subjects and caregivers administering romiplostim correctly, the precision of the 95% CI varies little between a sample size of 40 and a sample size of 60 (see Table 1. Hence for this study, the target sample size is 40 subjects. Subjects enrolled Table 1. Precision in Estimate by Subjects Enrolled Upper confidence bounds on proportion of subjects not self-administering correctly if 0 errors are observed* Precision in estimate if 90% of subjects self-administer correctly** Precision in estimate if 80% of subjects self-administer correctly** 20 15% 70% - 97% 58% - 92% % 77% - 96% 65% - 90% 60 5% 80% - 95% 68% - 88% * The upper confidence bounds in a sample of size N is 3/N 1. ** Precision is based on the 95% CI for a proportion using Wilson s score method (no continuity correction) 2,3. 4. STUDY ENDPOINTS 4.1 Primary Endpoint A yes / no indicator for whether the subject or caregiver administers romiplostim correctly at the 4 week visit. The indicator will be "yes" if: the difference between the prescribed and administered dose of romiplostim is within ± 10% (as defined in Section 6.1.1); and the HCP observes that the subject or caregiver reconstitutes romiplostim correctly; and successfully injects romiplostim (as described in the secondary endpoint below); and if the HCP does not have to intervene to correct an error during administration by patient or caregiver (e.g. volume in syringe). The indicator will be "no" if any of criteria above are not met. 4.2 Secondary Endpoints A yes / no indicator for whether the subject or caregiver reconstitutes romiplostim correctly at the 4 week visit / at follow-up visits. The indicator will be yes if: the HCP observes that aseptic techniques were used in preparing the vial; and all the water from the sterile water syringe was injected gently into the vial; and all romiplostim was dissolved. The indicator will be "no" if any of these criteria are not met.

208 Date: 04 April 2016 Page 208 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 8 For accuracy in administering the prescribed dose of romiplostim, the difference between the prescribed and administered dose of romiplostim, will be expressed as a % (as defined in Section 6.1.1). A yes / no indicator for whether the subject or caregiver injects romiplostim successfully at the 4 week visit / at follow-up visits. The indicator will be yes if: the HCP observes that all air bubbles were removed from syringe; and the patient made clinically appropriate use of alcohol wipe at injection site; and clinically appropriate handling of syringe to avoid contamination; and clinically appropriate technique of subcutaneous injection. The indicator will be "no" if any of these criteria are not met. A yes / no indicator for whether the subject or caregiver administers romiplostim correctly at follow-up visits. The indicator will be yes or no as described by the criteria for the primary endpoint (Section 4.1). 5. HYPOTHESES OR ESTIMATION This study is descriptive in nature and does not involve formal hypothesis testing. However, for the binary ( yes / no ) endpoints, a 2-sided 95% CI will be estimated for the associated proportion of subjects recorded as yes or no by the HCP. 6. DEFINITIONS 6.1 General Study-level Definitions Screened subject: an ITP patient who has been selected for HAT pack training. Eligible subject: a screened subject who satisfies the Protocol eligibility criteria. Consenting subject: an eligible subject who agrees to participate in the study. Participating subject: a consenting subject who attends the first standard-of-care visit 2-8 weeks after training with the HAT pack. Study Day 1: subject completes HAT pack training. Administered dose of romiplostim calculation: At each visit the dose administered by the subject or caregiver will be calculated using the following formula based on the variables collected: Romiplostim dose administered (micrograms) = Observed injection volume (milliliter) * 500 (micrograms/milliliter). If a subject uses multiple vials/syringes then the observed injection volume will be summed across these prior to applying the formula above.

209 Date: 04 April 2016 Page 209 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 9 Prescribed dose of romiplostim calculation: At each visit the dose prescribed by the physician will be calculated using the following formula based on the variables collected: Romiplostim dose prescribed (micrograms) = Prescribed injection volume (milliliter) * 500 (micrograms/milliliter). If a subject uses multiple vials/syringes then the prescribed injection volume will be summed across these prior to applying the formula above. Difference between prescribed and administered dose of romiplostim calculation: This secondary endpoint (and component of the primary endpoint) will be calculated at each visit using the following formula: Difference (%) = 100 * [Administered romiplostim dose (micrograms) - Prescribed romiplostim dose (micrograms)] / Prescribed romiplostim dose (micrograms) Time Windows Based on a Baseline of study Day 1, completion of HAT pack training material, the acceptable time window for the Week 4 visit ranges from Week 2 to Week 8 for each subject. In the event of multiple visits in this visit window the first of these will be considered the Week 4 standard of care visit. Follow-up visits will be not be time windowed to a week number and the date of the visit will be displayed in a subject listing Derivations 7. ANALYSIS SUBSETS 7.1 Primary Analysis Set Full Analysis Set (FAS): includes all participating subjects as defined in Section Additional Analysis Sets Screened Subject Set: includes all screened subjects as defined in Section Eligible Subject Set: includes all eligible subjects as defined in Section Consenting Subject Set: includes all consenting subjects as defined in Section

210 Date: 04 April 2016 Page 210 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page DATA SCREENING AND ACCEPTANCE 8.1 General Principles The objective of the data screening is to assess the quantity, quality and statistical characteristics of the data relative to the requirements of the planned analyses. 8.2 Data Handling and Electronic Transfer of Data All data for this study will be housed within the Electronic Data Capture (EDC) database, RAVE. 8.3 Handling of Missing and Incomplete Data Missing data is expected to be minimal for the background and study drug administration variables collected in the standardized data collection form. There is no specific requirement for the subjects to complete the home administration diary, nor is it compulsory to bring the diary to the 4 Week visit, and so this data may be limited and the quality of completion may vary between subjects. For the composite categorical binary ( yes / no ) endpoints if no response has been entered for any of the questions/items comprising the endpoint, then in a conservative approach it will be assumed that the criteria have not all been met and a no imputed. This enables the sum of the numerators across the categories to be equal to the denominator which will be the appropriate analysis set i.e. FAS. For the continuous endpoints if no response has been entered then no imputation will be performed and summary statistics will be calculated based on a reduced denominator. Where the endpoint is comprised of two continuous variables, for example the difference between the prescribed and administered dose of romiplostim, if either of these are missing then the endpoint will be missing. Where individual variables from the standardized data collection form (including transcriptions from the home administration diary) are presented in tables or listings the actual response will be displayed, including missing if applicable. 8.4 Detection of Bias ITP patients eligible to participate in the study may decline to participate for various reasons and thus responses from the participating subjects may not be generalizable to all eligible patients within the EU. To assess for this bias, a 'cooperation proportion' for clinical sites and subjects will be tallied by the number of clinical sites and subjects participating in the study divided by number offered inclusion.

211 Date: 04 April 2016 Page 211 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 11 Sites within each country will become eligible for participation after launch of the HAT pack for self-administration at the country level. The availability of the HAT pack will be staggered at the country level from March 2013 onwards and sites and countries will be selected based on availability and willingness to participate. Thus patients from countries where the HAT pack is launched at an earlier date are more likely to be involved in the study and recruitment will only continue until the required sample size has been met in a consecutive approach. As a result, the study population is likely to be a convenience sample but the small size limits the assessment of sample characteristics. Patients may have been self-administering romiplostim prior to the approval and launch of the HAT pack training so the date of initial prescription of self-administration may be examined to assess if any bias has been introduced as a result of potentially including patients who have been self-administering for longer versus those who were naïve to self-administering. Another possible limitation is the direct observation at first standard-of-care visit may not reflect the subject or caregivers s ability to administer romiplostim between clinical visits when the act of administration is not observed. To assess this bias, data from the self-administration diary is being collected. Incomplete recording of data for all subjects may result in non-differential measurement error; therefore data summaries will be reviewed in terms of the frequency and patterns of missing data to ascertain whether such bias may have been introduced. Follow-up visits will occur only if the HCP requests them, therefore, data may only be available for a selection of subjects and for a varying number and timing of visits. Subsequently, the data captured at these visits will be provided in a subject listing. 8.5 Outliers The endpoints for this study are based on variables collected in the standardized data collection form with almost all having a pre-defined set of responses, and so there is minimal scope for outliers in responses. 8.6 Distributional Characteristics The lower and upper limits of a 95% CI for a proportion must lie within the closed interval [0,1].

212 Date: 04 April 2016 Page 212 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page Validation of Statistical Analyses Programs will be developed and maintained, and outputs will be verified in accordance with current risk-based quality control procedures. Tables and listings will be produced with validated standard macro programs where standard macros can produce the specified outputs. The production environment for statistical analyses consists of Amgen-supported versions of statistical analysis software, for example the SAS System version STATISTICAL METHODS OF ANALYSIS 9.1 General Principles The data analysis for this study will be descriptive in nature. Continuous endpoints Endpoints of a continuous nature will be described by the mean, standard deviation, range (minimum and maximum), median, and lower and upper quartiles. Categorical endpoints Endpoints of a categorical binary nature ( yes / no ) will be summarized as the frequency and proportion (percentage) based on the appropriate analysis set denominator. For the primary and secondary endpoints, as a measure of precision, a 95% CI (Wilson s score method [no continuity correction]) will be calculated around the proportions responding yes or no. 9.2 Subject/Patient Accountability To obtain the proportion of subject participation (also known as the 'cooperation proportion'), each enrolling clinical site will provide the number of subjects participating in the study and the number of eligible subjects offered inclusion but who refused. 9.3 Demographic and Baseline Characteristics Background demographics and baseline characteristics, including sex, age and administerer of drug (subject or caregiver) will be summarized. The background demographic and baseline characteristics for each subject will be displayed in listings.

213 Date: 04 April 2016 Page 213 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page Analyses of Primary Endpoint The categorical binary primary endpoint as described in Section 4.1 will be analysed as outlined in Section 9.1. An associated listing will display the primary endpoint achievement status and the components contributing towards this for subjects. 9.5 Analyses of Secondary Endpoints The secondary endpoints as described in Section 4.2 will be analysed as outlined in Section 9.1. An associated listing will display each secondary endpoint s achievement status for subjects. Assessment of the endpoints at the follow-up visits will only be presented in subject listings. 9.6 Safety Analyses Adverse Drug Reactions The Medical Dictionary for Regulatory Activities (MedDRA) version 16.0 or later will be used to code all adverse drug reactions (ADRs) to a system organ class and a preferred term. ADRs (serious, non-serious and overall) will be tabulated by system organ class and preferred term in descending order of frequency. The tabulations will present pre-and post-completion of the HAT pack training (Day 1) ADRs separately. ADRs with a start date on or after the date of consent but before the date of completion of the HAT pack training (Day 1) will be classified as pre-completion of the HAT pack training, and ADRs with a start date on or after the date of completion of the HAT pack training (Day 1) will be classified as post-completion of the HAT pack training Exposure to Product The dosing administration variables (Section 9.3 of the Protocol) used to derive the composite primary and secondary endpoints will be summarized and listed. All available data transcribed from the home administration diary at the 4 Week visit will be listed for each subject. 10. CHANGES FROM PROTOCOL OR RPP SPECIFIED ANALYSES There are no changes to the pre-specified analyses.

214 Date: 04 April 2016 Page 214 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page LITERATURE CITATIONS / REFERENCES Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version [updated September 2009]. The Cochrane Collaboration, Available from Newcombe R. Two-Sided Confidence Intervals for the Single Proportion: Comparison of Seven Methods. Statistics in Medicine. 1998;17: Wilson EB. Probable Inference, the Law of Succession, and Statistical Inference. Journal of the American Statistical Association. 1927;22:

215 Date: 04 April 2016 Page 215 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page LIST OF PLANNED TABLES, FIGURES AND LISTINGS 12.1 Planned Tables Category Title Description 1. Subject/Patient Accountability 2. Demography and Baseline Characteristics 3. Treatment Compliance -Primary Endpoint 4. Treatment Compliance-Secondary Endpoints 5. Treatment Compliance -Additional Endpoints Summary of Subject Participation by Site and Overall Summary of Demographic and Baseline Characteristics-Full Analysis Set Primary Endpoint: Subject or Caregiver Administers Romiplostim Correctly at the 4 Week Visit Full Analysis Set Secondary Endpoint: Subject or Caregiver Reconstitutes Romiplostim Correctly at the 4 Week Visit Full Analysis Set Summary of Difference Between the Prescribed and Administered Dose (Micrograms) of Romiplostim at the 4 Week Visit-Full Analysis Set Secondary Endpoint: Subject or Caregiver Injects Romiplostim Successfully at the 4 Week Visit-Full Analysis Set Additional Endpoint: Difference Between the Prescribed and Administered Dose (Micrograms) of Romiplostim is Within ± 10% at the 4 Week Visit-Full Analysis Set Additional Endpoint: The HCP Does Not Have to Intervene to Correct an Error During Administration by Patient or Caregiver Full Analysis Set The frequency of subjects screened, eligible, consenting and participating. The frequency and proportion of eligible subjects who participated in the study. The frequency and proportion of screened subjects who participated in the study. The frequency and proportion of subjects for the categorical items; sex (male or female) and administerer of drug (patient or caregiver). Summary statistics for the continuous item; age. The frequency and proportion of subjects in the categories yes and no. Two-sided 95% CI for the yes and no categories. The frequency and proportion of subjects in the categories yes and no. Two-sided 95% CI for the yes and no categories. Summary statistics. The frequency and proportion of subjects in the categories yes and no. Two-sided 95% CI for the yes and no categories. The frequency and proportion of subjects in the categories yes and no. Two-sided 95% CI for the yes and no categories. The frequency and proportion of subjects in the categories yes and no. Two-sided 95% CI for the yes and no categories.

216 Date: 04 April 2016 Page 216 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 16 Category Title Description 6. Treatment Compliance - Exposure to Product Summary of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints at the 4 Week Visit-Full Analysis Set The frequency and proportion of subjects in the categories for the following variables: Healthcare professional observing - clinical site, provider type (clinician, nurse, or other) Healthcare professional observing was same person as trainer (yes or no) Number of vials of romiplostim needed for single administration (1, 2 or 3) Romiplostim dose prescribed by physician-milliliter vial size of romiplostim (250 or 500 microgram kit) Aseptic techniques were used in preparing the vial/s (yes or no) Gently injected all water from sterile water syringe into the vial/s (yes or no) Ensured all romiplostim dissolved (yes or no) Ensured all air bubbles removed from injection syringe (yes or no) Clinically appropriate use of alcohol wipe at injection site (yes or no) Clinically appropriate handling of syringe to avoid contamination (yes or no) Clinically appropriate technique of subcutaneous injection (yes or no) Healthcare professional intervened to correct an error during administration by subject or caregiver (yes or no) If the response was missing then present this additional category.

217 Date: 04 April 2016 Page 217 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 17 Category Title Description 7. Adverse Drug Reactions Summary of All Adverse Drug Reactions Pre-Completion of HAT Pack Training by Preferred Term Consenting Subject Set Summary of Serious Adverse Drug Reactions Pre-Completion of HAT Pack Training by Preferred Term-Consenting Subject Set Summary of Non-Serious Adverse Drug Reactions Pre- Completion of HAT Pack Training by Preferred Term-Consenting Subject Set Summary of All Adverse Drug Reactions Post-Completion of HAT Pack Training by Preferred Term Consenting Subject Set Summary of Serious Adverse Drug Reactions Post- Completion of HAT Pack Training by Preferred Term-Consenting Subject Set Summary of Non-Serious Adverse Drug Reactions Post-Completion of HAT Pack Training by Preferred Term-Consenting Subject Set The frequency and proportion of subjects for each coded ADR. Present by descending frequency. The frequency and proportion of subjects for each coded ADR. Present by descending frequency. The frequency and proportion of subjects for each coded ADR. Present by descending frequency. The frequency and proportion of subjects for each coded ADR. Present by descending frequency. The frequency and proportion of subjects for each coded ADR. Present by descending frequency. The frequency and proportion of subjects for each coded ADR. Present by descending frequency.

218 Date: 04 April 2016 Page 218 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page Planned Listings Category Title Description 1. Subject/Patient Accountability 3. Treatment Compliance-Primary Endpoint 4. Treatment Compliance-Secondary Endpoints 5. Treatment Compliance-Additional Endpoints Listing of Relevant Dates by Subject Consenting Subject Set Listing of Primary, Secondary and Additional Endpoints Status by Subject and Visits (4 Week and Follow-Up)-Full Analysis Set Present the following information: Subject Site Country Date of initial prescription for self-administration Date of consent Date of completion of HAT pack training Date of observation (include the 4 Week and follow-up visits) Reason for follow-up visit (if applicable) Present the following information and endpoints recorded at the 4 Week and follow-up visits: Subject Site Country Sex (male or female) Age (years) Administerer of drug (patient or caregiver) Visit date Subject or caregiver administers romiplostim correctly? (yes or no) Subject or caregiver reconstitutes romiplostim correctly? (yes or no) Subject injects romiplostim successfully? (yes or no) Difference between the prescribed and administered dose of romiplostim is within ± 10%? (yes or no) The HCP does not have to intervene to correct an error during administration by patient or caregiver? (yes or no)

219 Date: 04 April 2016 Page 219 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 19 Category Title Description 6. Treatment Compliance - Exposure to Product Listing of Dose Administration Variables Supporting the Primary, Secondary and Additional Endpoints by Subject and Visits (4 Week and Follow-Up) Full Analysis Set Healthcare professional observing - clinical site, provider type Healthcare professional observing was same person as trainer Number of vials of romiplostim needed for single administration Romiplostim dose prescribed by physician-milliliter vial size of romiplostim Aseptic techniques were used in preparing the vial/s Gently injected all water from sterile water syringe into the vial/s Ensured all romiplostim dissolved Ensured all air bubbles removed from injection syringe Clinically appropriate use of alcohol wipe at injection site Clinically appropriate handling of syringe to avoid contamination Clinically appropriate technique of subcutaneous injection Healthcare professional intervened to correct an error during administration by subject or caregiver Total administered dose of romiplostim (micrograms) Total prescribed dose of romiplostim (micrograms)

220 Date: 04 April 2016 Page 220 of 270 Product: Romiplostim Statistical Analysis Plan: Dated: 4 March 2014 Page 20 Category Title Description 7. Home Administration Diary Listing of Information Transcribed From the Home Administration Diary at the 4 Week Visit by Subject Full Analysis Set Present the following information recorded at the 4 Week visit: Subject Site Country Sex (male or female) Age (years) Administerer of drug (patient or caregiver) Dose diary brought in? (yes or no) Date of administration at home Correct dose administered at home? (yes or no) Any other problems with administration at home? (yes or no) Description of problem?

221 Date: 04 April 2016 Page 221 of 270 Annex 5. Home Administration Training (HAT) Pack Materials Physicians are provided with various HAT pack training materials. These include: Selecting and Training Patients for Home Administration of Nplate (romiplostim) Checklist for Nplate Self-administration Patients are provided with various HAT pack training materials to ensure the correct administration of romiplostim at home. These materials include: a quick guide to home administration of romiplostim which provides and overview of what is involved with home administration and ways to ensure the best result at home; a preparation mat which provides a place to prepare the injection; a step-by-step guide to preparation and administration of the romiplostim injection which is an illustrated guide to self-administration to help ensure the steps are followed in the correct order; a step-by-step self-administration video; and a self-administration diary which is a booklet to keep track of training days, injection days, clinic visits, and space to record weekly dose and any problems experienced with home preparation and administration.

222 Date: 04 April 2016 Page 222 of 270 CHECKLIST FOR NPLATE SELF-ADMINISTRATION Confirm that your patient is... Interested in and suitable for home administration On a stable dose of Nplate For the purposes of self-administration, a stable dose of Nplate means the patient s Nplate dose has not required adjustment for at least 4 weeks. Patients who need dose adjustments cannot return to self-administration until their dose has stabilised that is, until their Nplate dose has not required adjustment for at least 4 weeks Willing to undergo a period of training Inform patients that home administration will involve a period of training, and that they will be required to demonstrate their ability to self-administer Ensure patients return home with... The right dose written and drawn in their Self-administration diary Their reconstitution pack(s) of Nplate (Make sure they have not taken the placebo demonstration kit with them by mistake) All patient materials in their Home Administration Training pack Contact information for a healthcare professional (to be written on the back of their Self-administration diary, in the section titled Just in case you need support... ) An appointment date to return after the first 4 weeks of selfadministration. At this visit they will have their platelet count checked and will again be supervised while reconstituting and administering Nplate. Only patients who demonstrate the ability to reconstitute and self-administer Nplate should be allowed to continue doing so

223 Date: 04 April 2016 Page 223 of 270 PRESCRIBING INFORMATION FOR NPLATE CAN BE FOUND HERE: Adverse events should be reported. Reporting forms and information can be found at Adverse events should also be reported to Amgen Limited on +44 (0) Amgen. All rights reserved. Job code: UKIE-NP Date of preparation: February 2015

224 Date: 04 April 2016 Page 224 of 270 SELECTING AND TRAINING PATIENTS FOR HOME ADMINISTRATION OF NPLATE (romiplostim)

225 Date: 04 April 2016 Page 225 of 270 CONTENTS THE PATIENT HOME ADMINISTRATION TRAINING PACK INCLUDES A SELF-ADMINISTRATION DIARY, WITH SPACE FOR YOU TO RECORD THEIR CORRECT DOSE. IN THE PATIENT S SELF-ADMINISTRATION DIARY, MAKE SURE YOU WRITE AND DRAW THE PATIENT S DOSE, CROSS OUT ANY OLD DOSES TO AVOID CONFUSION, AND COMPLETE THE SECTION TITLED, JUST IN CASE YOU NEED SUPPORT... ON THE BACK OF THE DIARY. Home administration of Nplate may be an option for some of your patients. This guide book will help you select and train appropriate patients to begin self-administration at home: The checklist on the tear-off pad is designed to help ensure your patients take the right materials home CRITERIA FOR APPROPRIATE PATIENTS 4 TRAINING YOUR PATIENTS 5 TRAINING STEPS 6 Planning for healthcare professionals Planning for patients/trainees IMPORTANT FACTORS TO KEEP IN MIND 8 Monitoring: dose adjustments and home administration PATIENT SUPPORT List of patient support materials provided by Amgen 10 The other patient-support items in this Home Administration Training pack will help support the training, confidence, and safety of patients administering Nplate at home See pages 10 and 11 for a complete list of the healthcare professional and patient support materials provided by Amgen as the Home Administration Training pack 3

226 Date: 04 April 2016 Page 226 of 270 CRITERIA FOR SELECTING PATIENTS TRAINING YOUR PATIENTS IN ORDER TO SELF-ADMINISTER NPLATE PATIENTS OR A CARER MUST BE: It is recommended that patients who are eligible for home administration undergo a period of training conducted by you, and certain follow-up steps as outlined below. STEP 1 Demonstration of reconstitution and administration technique (using the Nplate preparation mat and the Step-by-step guide) Interested in home administration On a stable + dose of + Nplate Willing to undergo a period of training STEP 2 The patient is supervised while reconstituting and administering Nplate. Only patients who demonstrate the ability to independently reconstitute and self-administer Nplate are allowed to do so 4 A patient or carer must be competent and capable of self-administration, and not expected to require the presence or assistance of a healthcare professional for self-administration. A patient can only be an appropriate candidate for Nplate selfadministration if they demonstrate: The ability to follow instructions The ability to properly handle all the required elements of selfadministration They have no physical condition that prevents them accurately reconstituting and injecting Nplate For the purposes of self-administration, a stable dose of Nplate means the patient s Nplate dose has not required adjustment for at least 4 weeks (with platelet counts 50 x 10 9 /L) Patients who need dose adjustments cannot return to self-administration until their dose has stabilised that is, until their Nplate dose has not required adjustment for at least 4 weeks The patient or a carer have to undergo training in Nplate preparation and self-administration techniques. They have to commit to undergo this training and accept that they cannot self-administer until they have demonstrated the ability to do so STEP 3 STEP 4 Patient reconstitutes and self-administers Nplate at home. Consider having a healthcare professional (e.g. nurse) call the patient after their first home injection to check for questions or problems Monitoring of platelet count and full blood count on a 4-weekly basis (as per the Summary of Product Characteristics). After the first 4 weeks of self-administration, the patient should again be supervised while reconstituting and administering Nplate. Only patients who demonstrate the ability to reconstitute and self-administer Nplate should be allowed to continue doing so Patient training should also include reminders that: Nplate is a very potent drug, and dose precision is essential. Variability in the total volume withdrawn from the vial following reconstitution has been observed, ranging from 0.5 ml to 0.6 ml extractable volume. In withdrawing the vial contents, a very small amount of liquid will remain in the vial (this volume may vary). If the product has been reconstituted with the self administration kit components, the final concentration will be correct and provide the correct dose for administration Nplate must be stored in the original box (protected from light) and kept cool (in the refrigerator) Reconstitution and administration must be performed safely, on a clean surface and with clean hands, and the injection site should always be cleaned before the injection PATIENTS CAN RECORD THEIR SUCCESS (AND ANY PROBLEMS) WITH SELF-ADMINISTRATION IN THEIR SELF-ADMINISTRATION DIARY. ENCOURAGE PATIENTS TO BRING THIS DIARY WITH THEM TO THEIR 4-WEEKLY MONITORING VISITS. IN THE PATIENT S SELF-ADMINISTRATION DIARY, MAKE SURE YOU WRITE AND DRAW THE PATIENT S DOSE, CROSS OUT ANY OLD DOSES TO AVOID CONFUSION, AND COMPLETE THE SECTION TITLED, JUST IN CASE YOU NEED SUPPORT... ON THE BACK OF THE DIARY. 5

227 Date: 04 April 2016 Page 227 of 270 TRAINING STEPS It is recommended that the training process involves at least two one-on-one sessions between the trainer (the healthcare professional) and trainee (the patient and/or carer), as outlined here: Some patients may be able to learn the steps in a 30-minute session. Others may require more time and/or may require a repetition of the first or second step. STEP 1 Demonstration of reconstitution and administration technique (including actual administration of the correct Nplate dose) The trainer will use the Nplate preparation mat and the Step-by-step guide to demonstrate all the reconstitution and administration steps the patient will undertake at home. This stage of training will end with a healthcare professional administering an actual injection of the correct Nplate dose. This stage of training must include the following steps: Reminder to ensure dose certainty by referring to the dose card Demonstration of how to clean the Nplate preparation mat and a reminder to wash hands Demonstration of how to lay the items out on the mat Demonstration of the reconstitution and administration steps Reminder to double-check the dose before administration Reminder to clean the injection site Instructions on how to administer a subcutaneous injection Reminder to dispose of all items properly STEP 2 The patient is supervised while reconstituting and administering Nplate. Only patients who demonstrate the ability to reconstitute and self-administer Nplate are allowed to do so The trainer will observe the patient reconstituting and administering the correct dose of Nplate, using the Nplate preparation mat and the Step-by-step guide. Ideally, the patient will be able to reconstitute and administer the correct dose of Nplate with minimal instruction from the trainer. At this stage, the patient must demonstrate their ability to inject the Nplate dose safely and correctly. This stage of training is used to assess each patient s ability to successfully and correctly reconstitute and administer their Nplate dose. If the patient requires significant assistance during the process, and/or the trainer feels the patient will not be able to perform the processes successfully at home (i.e. without assistance), then the patient should not be allowed to progress to home administration until the trainer feels the trainee is prepared. This step needs to be repeated weekly until trainer and trainee are both confident that self-administration can be performed correctly. STEP 3 Patient reconstitutes and self-administers Nplate at home. Consider having a healthcare professional (e.g. nurse) call the patient after their first home-injection to check for questions or problems The trainee will prepare and self-administer the correct dose of Nplate at their own home. The exact date and time of the injection should be agreed upon by the trainer and trainee, and written in the patient s Self-administration diary. If the trainee could not successfully reconstitute and administer the correct dose of Nplate, the reason(s) for failure should be assessed and, if considered necessary, further training should be performed. The trainer should report the failure and the reason(s) for failure to Amgen. Details of how to report this to Amgen and the yellow card reporting system can be found on the reverse of this booklet. STEP 4 Remember to schedule 4-weekly monitoring visits with your self-administering patients. After the first 4 weeks of self-administration, the patient should again be supervised while reconstituting and administering Nplate. Only patients who demonstrate the ability to reconstitute and self-administer Nplate should be allowed to continue doing so. 6 7

228 Date: 04 April 2016 Page 228 of 270 IMPORTANT FACTORS TO KEEP IN MIND The practicalities of home reconstitution and administration of Nplate were assessed in a Human Factors Engineering study. The study identified the following issues, all of which can be managed by training: Underdosing or overdosing Incorrect storage of Nplate Contaminated work surface Contamination of the Luer-lok connector Poor attachment of the syringe Failure to prepare injection site During training, these issues should be verbally addressed and correct behaviour demonstrated (where appropriate). PATIENT MONITORING: DOSE ADJUSTMENTS AND HOME ADMINISTRATION Patients platelet counts must be monitored on a 4-weekly basis. Such monitoring will enable identification of patients who need dose adjustments. Patients who need dose adjustments cannot return to self-administration until their dose has stabilised that is, until their Nplate dose has not required adjustment for at least 4 weeks. 4-weekly monitoring visits may also be a good time to review the patients Self-administration diary, in which they will have recorded their experiences of home administration. After the first 4 weeks of self-administration, the patient should again be supervised while reconstituting and administering Nplate. Only patients who demonstrate the ability to reconstitute and self-administer Nplate should be allowed to continue doing so. THE IDENTIFIED RISKS OUTLINED HERE CAN ALL BE MANAGED BY THOROUGH TRAINING CONDUCTED AND SUPERVISED BY A HEALTHCARE PROFESSIONAL, AND THROUGH FOLLOW-UP MONITORING (POTENTIALLY BY PHONE). 8 9

229 Interested in and suitable for home administration On a stable dose of Nplate For the purposes of self-administration, a stable dose of Nplate means the patient s Nplate dose has not required adjustment for at least 4 weeks. Patients who need dose adjustments cannot return to self-administration until their dose has stabilised that is, until their Nplate dose has not required adjustment for at least 4 weeks Willing to undergo a period of training Inform patients that home administration will involve a period of training, and that they will be required to demonstrate their ability to self-administer The right dose written and drawn in their Self-administration diary Their reconstitution pack(s) of Nplate (Make sure they have not taken the placebo demonstration kit with them by mistake) All the patient materials in their Home Administration Training pack Contact information for a healthcare professional (to be written on the back of their Self-administration diary, in the section titled Just in case you need support... ) An appointment date to return after the first 4 weeks of self-administration. At this visit they will have their platelet count checked and will again be supervised while reconstituting and administering Nplate. Only patients who demonstrate the ability to reconstitute and self-administer Nplate should be allowed to continue doing so Reporting of side effects If you get any side effects, talk to your doctor, pharmacist or nurse. This includes any possible side effects not listed in the package leaflet. You can also report side effects directly via the Yellow Card Scheme at yellowcard. By reporting side effects you can help provide more information on the safety of this medicine Amgen. All rights reserved. Job code: NPO-GBR-AMG NP Date of prep: April Reporting of side effects If you get any side effects, talk to your doctor, pharmacist or nurse. This includes any possible side effects not listed in the package leaflet. You can also report side effects directly via the Yellow Card Scheme at By reporting side effects you can help provide more information on the safety of this medicine Dosecard V4.indd 1 5/20/14 12:32 PM BEFORE YOU BEGIN REPORTING OF SIDE EFFECTS Read all instructions for use thoroughly. If you get any side effects, talk to your doctor, pharmacist or nurse. This includes You should establish a regular treatment schedule, even when travelling. any possible side effects not listed in the package leaflet. You can also report side Make sure you have all the supplies that you will need (refer to the images below). effects directly via the Yellow Card Scheme at Check the expiry date of the self-injection kit. By reporting side effects you can help provide more information on the safety of Do not use if expired or left at room temperature for more than 24 hours. this medicine. Wash your hands with soap. Use an alcohol swab to clean the surface where you will be preparing product. NOTE: Please use this mat along with the Step-by-step guide and/or Package leaflet and Self-administration DVD to prepare your Nplate injection. You may be required to use more than one injection of Nplate. If your doctor has instructed you to take more than one injection of Nplate, follow the steps as defined in the Step-by-step guide to complete your prescribed dose of product using a new self-administration kit(s). A Product vial (1) B1 Plunger rod (1) B2 Prefilled sterile water syringe (1) C Disposable 1 ml syringe with Luer-lok (1) Store product vials in their carton to protect from light until use. This medicine may be removed from the refrigerator for a maximum period of 24 hours at room temperature (up to 25 C). It must be returned to the refrigerator for storage longer than 24 hours. D Safety needle: 27 G, 13 mm (1) E Vial adapter (1) F Alcohol swab packages (x4) G Sharps container (1; not supplied) Only for use by people who have been trained in self-administration by their healthcare professional. NOTE: These images are for reference only. Job code: NPO-GBR-AMG NP Date of prep: April Prep mat update V3.indd 1 5/20/14 12:41 PM Reporting of side effects If you get any side effects, talk to your doctor, pharmacist or nurse. This includes any possible side effects not listed in the package leaflet. You can also report side effects directly via the Yellow Card Scheme at By reporting side effects you can help provide more information on the safety of this medicine Amgen. All rights reserved. Product or Therapeutic Area: Romiplostim Date: 04 April 2016 Page 229 of 270 PATIENT SUPPORT In addition to training and monitoring, patient support is integral to reducing the risks outlined on the previous page. Patient support may include: HOME ADMINISTRATION TRAINING PACK: PATIENT SUPPORT MATERIALS PROVIDED BY AMGEN The option to return to the clinic and perform reconstitution and self-administration again under the observation of a healthcare professional The option to have a dose administered at the clinic rather than at home Distribution of additional copies of the materials from the Home Administration Training Pack HOME ADMINISTRATION TRAINING PACK: HEALTHCARE PROFESSIONAL SUPPORT MATERIALS PROVIDED BY AMGEN SELECTING AND TRAINING PATIENTS FOR HOME ADMINISTRATION OF NPLATE (romiplostim) CHECKLIST FOR NPLATE SELF-ADMINISTRATION Confirm that your patient is... Ensure patients return home with... QUICK GUIDE TO HOME ADMINISTRATION OF NPLATE For people who have chosen to give themselves Nplate injections at home. The materials in this Home Administration Training pack will help ensure you can successfully inject Nplate at home. Quick guide booklet (with a preadministration checklist, an overview of Nplate, an introduction to self-administration, and an overview of the training programme) Step-by-Step guide to preparation ANd AdMiNiStRAtioN of the NplAte injection A visual aid for your home-administration of Nplate SELF-ADMINISTRATION DIARY To support your home administration of Nplate Your healthcare professional should write your most up-to-date dose in the front of this Self-administration diary Your healthcare professional should write the name of a contact person in this Self-administration diary, in the section titled Just in case you need support... (in the back of this diary) Use this Self-administration diary to help you remember what to tell your healthcare professional at your next appointment Self-administration diary (including a page for recording both dose prescribed and dose administered) NPLATE PREPARATION MAT You must be trained by your healthcare professional before you can administer Nplate Make sure you have the following materials before you begin NPO-GBR-AMG NP 2014 Amgen Inc. Date of preparation: April 2014 Step-by-Step guide: preparation and SeLF-adMiniStration of the nplate injection Self-administration DVD this booklet will guide you through all the steps you ll need to perform to successfully prepare and administer your Nplate injection at home. Selecting and training patients for home administration of Nplate (this guide) Checklist for Nplate self-administration (a guide to selecting patients and ensuring they receive the right materials for home administration) during training, your healthcare professional will have shown you all of the steps in this Step-by-step guide. When you administer at home, this booklet will remind you of all the preparation and administration steps, so you can be confident you re performing every step correctly. Reporting of side effects if you get any side effects, talk to your doctor, pharmacist or nurse. this includes any possible side effects not listed in the package leaflet. you can also report side effects directly via the yellow card Scheme at by reporting side effects you can help provide more information on the safety of this medicine. Please contact your HealtHcare ProFessIonal with any questions or concerns you Have YOUR PREPARATION AREA Step-by-step guide Nplate preparation mat 10 11

230 Date: 04 April 2016 Page 230 of 270 PRESCRIBING INFORMATION FOR NPLATE CAN BE FOUND HERE: Adverse events should be reported. Reporting forms and information can be found at Adverse events should also be reported to Amgen Limited on +44 (0) Amgen. All rights reserved. Job code: UKIE-NP Date of prep: February 2015

231 Date: 04 April 2016 Page 231 of 270 Self-administration diary To support your home administration of Nplate Your healthcare professional should write your most up-to-date dose in the front of this Self-administration diary Your healthcare professional should write the name of a contact person in this Self-administration diary, in the section titled Just in case you need support... (in the back of this diary) Use this Self-administration diary to help you remember what to tell your healthcare professional at your next appointment Reporting of side effects If you get any side effects, talk to your doctor, pharmacist or nurse. This includes any possible side effects not listed in the package leaflet. You can also report side effects directly via the Yellow Card Scheme at By reporting side effects you can help provide more information on the safety of this medicine.

232 Date: 04 April 2016 Page 232 of 270 Dose recorder Use this page to keep track of the dose you will administer at home. Your healthcare professional will fill this page in for you. If two vials are needed to administer the correct total dose, your healthcare professional should write the dose for each vial (ml) used. Welcome to home administration of Nplate. This Self-administration diary will help you to keep track of your home administration of Nplate through recording the following: Your up-to-date dose training days for home administration (at the clinic) Dates you should receive your injection (either at home or in the clinic) Dates you had your injection (either at home or in the clinic) The dose that was injected each treatment Any problems you experienced with self-administration It is important to keep a record of these as it will help you and your healthcare professional ensure that you take the right dose of Nplate at the right time. Correct dose (ml) Date Nplate dose prescribed... / /... / /... / / Visual record of correct dose... / /

233 Date: 04 April 2016 Page 233 of 270 Dose recorder Use this page to keep track of the dose you will administer at home. Your healthcare professional will fill this page in for you. If two vials are needed to administer the correct total dose, your healthcare professional should write the dose for each vial (ml) used. Training diary Use this scheduler to keep track of the days you attended the clinic to learn how to prepare and administer your own Nplate injections. Correct dose (ml) Date Nplate dose prescribed Visual record of correct dose Day Date Time Type of training (delete as appropriate)... / / Mon Tue Wed Thu Fri Sat Sun / /... :... Demonstration given to me by a healthcare professional Healthcare professional observed me self-administer Mon Tue Wed Thu Fri Sat Sun / /... :... Demonstration given to me by a healthcare professional Healthcare professional observed me self-administer... / / Mon Tue Wed Thu Fri Sat Sun / /... :... Demonstration given to me by a healthcare professional Healthcare professional observed me self-administer... / / Mon Tue Wed Thu Fri Sat Sun / /... :... Demonstration given to me by a healthcare professional Healthcare professional observed me self-administer Mon Tue Wed Thu Fri Sat Sun / /... :... Demonstration given to me by a healthcare professional Healthcare professional observed me self-administer... / / Mon Tue Wed Thu Fri Sat Sun / /... :... Demonstration given to me by a healthcare professional Healthcare professional observed me self-administer

234 Date: 04 April 2016 Page 234 of 270 Self-administration diary If you administer the wrong dose, contact your healthcare professional immediately. They may want to monitor you for a time. If two vials are needed to administer the correct total dose, write the dose for each vial (ml) used. Self-administration diary If you administer the wrong dose, contact your healthcare professional immediately. They may want to monitor you for a time. If two vials are needed to administer the correct total dose, write the dose for each vial (ml) used. Day & date Nplate dose is due Administered dose (ml) Did you take the right dose on the right date? Note any problems with self-administration. If scheduled dose was missed, include revised date of dose and reason for the change. Day & date Nplate dose is due Administered dose (ml) Did you take the right dose on the right date? Note any problems with self-administration. If scheduled dose was missed, include revised date of dose and reason for the change. Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No 4-weekly clinic Day Date 4-weekly clinic Day Date

235 Date: 04 April 2016 Page 235 of 270 Self-administration diary If you administer the wrong dose, contact your healthcare professional immediately. They may want to monitor you for a time. If two vials are needed to administer the correct total dose, write the dose for each vial (ml) used. Self-administration diary If you administer the wrong dose, contact your healthcare professional immediately. They may want to monitor you for a time. If two vials are needed to administer the correct total dose, write the dose for each vial (ml) used. Day & date Nplate dose is due Administered dose (ml) Did you take the right dose on the right date? Note any problems with self-administration. If scheduled dose was missed, include revised date of dose and reason for the change. Day & date Nplate dose is due Administered dose (ml) Did you take the right dose on the right date? Note any problems with self-administration. If scheduled dose was missed, include revised date of dose and reason for the change. Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No 4-weekly clinic Day Date 4-weekly clinic Day Date

236 Date: 04 April 2016 Page 236 of 270 Self-administration diary If you administer the wrong dose, contact your healthcare professional immediately. They may want to monitor you for a time. If two vials are needed to administer the correct total dose, write the dose for each vial (ml) used. Self-administration diary If you administer the wrong dose, contact your healthcare professional immediately. They may want to monitor you for a time. If two vials are needed to administer the correct total dose, write the dose for each vial (ml) used. Day & date Nplate dose is due Administered dose (ml) Did you take the right dose on the right date? Note any problems with self-administration. If scheduled dose was missed, include revised date of dose and reason for the change. Day & date Nplate dose is due Administered dose (ml) Did you take the right dose on the right date? Note any problems with self-administration. If scheduled dose was missed, include revised date of dose and reason for the change. Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No 4-weekly clinic Day Date 4-weekly clinic Day Date

237 Date: 04 April 2016 Page 237 of 270 Self-administration diary If you administer the wrong dose, contact your healthcare professional immediately. They may want to monitor you for a time. If two vials are needed to administer the correct total dose, write the dose for each vial (ml) used. Self-administration diary If you administer the wrong dose, contact your healthcare professional immediately. They may want to monitor you for a time. If two vials are needed to administer the correct total dose, write the dose for each vial (ml) used. Day & date Nplate dose is due Administered dose (ml) Did you take the right dose on the right date? Note any problems with self-administration. If scheduled dose was missed, include revised date of dose and reason for the change. Day & date Nplate dose is due Administered dose (ml) Did you take the right dose on the right date? Note any problems with self-administration. If scheduled dose was missed, include revised date of dose and reason for the change. Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No 4-weekly clinic Day Date 4-weekly clinic Day Date

238 Date: 04 April 2016 Page 238 of 270 Self-administration diary If you administer the wrong dose, contact your healthcare professional immediately. They may want to monitor you for a time. If two vials are needed to administer the correct total dose, write the dose for each vial (ml) used. Questions for your healthcare professional: Day & date Nplate dose is due Administered dose (ml) Did you take the right dose on the right date? Note any problems with self-administration. If scheduled dose was missed, include revised date of dose and reason for the change. Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No Day... Date / /... Yes No 4-weekly clinic Day Date

239 Date: 04 April 2016 Page 239 of 270 Just in case you need support... Your healthcare professional should write the information for your Nplate self-administration contact person here. Contact name: Name of healthcare institution: Telephone: For any information about this medicine, please contact: Amgen Limited, 240 Cambridge Science Park, Milton Road, Cambridge, CB4 0WD Amgen. All rights reserved. Job code: NPO-GBR-AMG NP Date of prep: April 2014

240 Date: 04 April 2016 Page 240 of 270 NPLATE PREPARATION Mat You must be trained by your healthcare professional before you can administer Nplate Before you begin Read all instructions for use thoroughly. You should establish a regular treatment schedule, even when travelling. Make sure you have all the supplies that you will need (refer to the images below). Check the expiry date of the self-injection kit. Do not use if expired or left at room temperature for more than 24 hours. Wash your hands with soap. use an alcohol swab to clean the surface where you will be preparing product. Make sure you have the following materials before you begin Reporting of side effects If you get any side effects, talk to your doctor, pharmacist or nurse. This includes any possible side effects not listed in the package leaflet. You can also report side effects directly via the Yellow Card Scheme at By reporting side effects you can help provide more information on the safety of this medicine. NOTE: Please use this mat along with the Step-by-step guide and/or Package leaflet and Self-administration DVD to prepare your Nplate injection. You may be required to use more than one injection of Nplate. If your doctor has instructed you to take more than one injection of Nplate, follow the steps as defined in the Step-by-step guide to complete your prescribed dose of product using a new self-administration kit(s). A Product vial (1) B1 Plunger rod (1) B2 Prefilled sterile water syringe (1) 0AMUS012_55_MAT_L18alt:250/500 mcg 6/14/10 7:53 PM Page 1 resistance. eference Mat Product Preparation and Injection Reference Mat You must be trained by your healthcare provider before you or your caregiver can administer product. Before you begin You will need: Patients should establish a regular treatment schedule, Product Vial (500 mcg) (Visual A) even when traveling. One Prefilled Sterile Water Syringe and Plunger (Visuals B-1 and B-2) C One Disposable Disposable 1 ml Syringe 1 (Visual ml Make sure you have all the supplies that you will need. C) Check the expiration dates on all vials. Safety Needle:27 G,½ inch (Visual D) Do not use if expired. syringe with One Vial Adapter (Visual E) Wash your hands with soap. Alcohol Swab Packages (Visual F) Luer-lok (1) Use alcohol swab to clean the surface where you will Sharps Container (not supplied) (Visual G) be preparing product. Enter your required injection volume here (from HCP): ml NOTE: You may be required to use more than one injection of product.if your doctor has instructed you to take more than one injection of product,follow the steps as defined in the Patient Instructions for Use to complete your prescribed dose of product using a new reconstitution kit(s). A.Product Vial (500 mcg) C.One Disposable 1 ml Syringe 8. Once the powder is completely dissolved,remove the syringe (B-1, areaofskinandholditfirmly.withtheotherhand,holdthesyringe(c) Store product B-2) from the vials adapter in their (E) bycarton twisting counterclockwise.discard to protect from light until at a 45-degree use. angle This to the medicine skin. may be removed from the refrigerator for a maximum period the syringe (B-1,B-2) into the sharps container (G). 19. After injecting your product dose,activate the pink needle safety of 24 hours 9. Removeat theroom 1 ml syringe temperature (C)from its package(up and pull to the 25 C). plunger toit must be returned to the refrigerator for storage longer D.Safety than Needle:27 24 hours. G,½ inch shield(d) with one hand until you hear it click into place,and then NOTE:These images are for reference only. E.One Vial Adapter F.Alcohol Swab Packages Preparation and Administration of Product 1. Remove the cover from the product vial (A)and clean the rubber 10. Attach the 1 ml syringe (C)to the vial adapter (E)of reconstituted stopper with an alcohol swab (F). solution by twisting the syringe tip clockwise onto the vial adapter Plastic cover E 2. Peel Vial off theadapter protective cover of (1) the vial adapter (E),but do not remove until you feel a slight resistance,and slowly F expel Alcohol air into the vial(a). swab packages B-1.One Plunger (x4) Rod B-2.One Prefilled Sterile Water G Syringe Sharps container (1; not supplied) it from the package.keeping the product vial (A) on a table,push the vial adapter (E) down onto the center of the vial s rubber stopper until it is firmly in place. 3. To assemble the prefilled sterile water syringe (B-1,B-2),attach the plunger rod (B-1) to the syringe (B-2) by twisting the plunger rod (B-1) clockwise onto the syringe (B-2) until you feel a slight resistance. 4. To remove the white plastic cover of the prefilled syringe (B-2),hold the syringe at its base with one hand and bend the tip of the white plastic cover downward with your other hand.this will break the seal of the white plastic cover. 5. Double-check that the vial adapter (E)is securely in place,and remove the packaging.keeping the product vial(a) onthetable,attachthe water-filledsyringe(b-1,b-2) to the vial adapter (E) by twisting the syringe tip clockwise onto the vial adapter until you feel a slight 6. Slowly and gently expel water into the product vial (A).Water should flow slowly onto the powder.do not force water into the vial. 7. Gently swirl the product vial to dissolve the powder.do not shake the vial.this may take as long as 2 minutes.once fully dissolved, product should be clear and colorless. the 1 ml marking.do not pull plunger past the 1 ml marking. Please call X-XXX-XXX-XXXX (X-XXX-XXX-XXXX) for more information. *Yourhealthcareprovidermayrecommendasmaller29-or31-gaugeneedleforinjection. Storeproductvialsintheircartontoprotectfromlightuntiluse. 11. Keeping the plunger at the base of the syringe (C),turn vial (A) assembly and syringe upside down,so the vial is above the syringe. 12. Withdraw all of the liquid into the syringe (C).Do not pull plunger past the 1 ml marking. 13. Remove all air bubbles by gently tapping the barrel of the syringe(c). Once the air bubbles have risen to the top,gently push them back into the vial (A) by slowly pushing the plunger. 14. Ensure the syringe(c)has the correct amount for your dose (doublecheck the injection volume entered above) by pushing the plunger of the syringe to expel any excess liquid back into the vial (A). 15. Remove syringe (C)from the vial adapter (E)by twisting counterclockwise,but make sure to keep the syringe in your hand. 16. While holding the syringe (C)in your hand with the tip facing up, remove the 27 G needle*(d) from its packaging by peeling apart the tabs. 17. Attach the needle (D)onto the filled syringe (C)by twisting clockwise until you feel a slight resistance.pull back the pink needle safety shield,and then remove the white needle cover by pinching it firmly and pulling in the opposite direction of the syringe assembly. 18. Open a new alcohol swab package (F)and clean the injection site. When giving the injection,use one hand to gently pinch the cleaned place the entire syringe assembly(c,d) into the sharps container (G). D Safety needle: 27 G, 13 mm (1) G.Sharps Container (not supplied) 2010 Company. All rights reserved. 5/10 MC50370-A P44178 NOTE: These images are for reference only. A.Product Vial (500 mcg) E.One Vial Adapter F.Alcohol Swab Packages B-1.One Plunger Rod Your Plastic cover preparation area Plastic cover B-2.One Prefilled Sterile Water Syringe C.One Disposable 1 ml Syringe G.Sharps Container (not supplied) D.Safety Needle:27 G,½ inch Job code: NPO-GBR-AMG NP Date of prep: April 2014

241 Date: 04 April 2016 Page 241 of 270 Quick guide to home administration of Nplate For people who have chosen to give themselves Nplate injections at home. The materials in this Home Administration Training pack will help ensure you can successfully inject Nplate at home. Reporting of side effects If you get any side effects, talk to your doctor, pharmacist or nurse. This includes any possible side effects not listed in the package leaflet. You can also report side effects directly via the Yellow Card Scheme at By reporting side effects you can help provide more information on the safety of this medicine Amgen. All rights reserved. Job code: NPO-GBR-AMG NP Date of prep: April 2014

242 Date: 04 April 2016 Page 242 of 270 Materials in your Home Administration Training pack Table of contents Item This Quick guide An Nplate preparation mat A Step-by-step guide to preparation and administration of the Nplate injection A Self-administration DVD A Self-administration diary Purpose Gives you an overview of what is involved with home administration, including a description of the training process and ways to ensure you get the best result at home Gives you a place on which to prepare your injection. Use in combination with your Step-by-step guide and Self-administration diary An illustrated guide to self-administration, to help ensure you follow the steps correctly and in the right order A step-by-step self-administration video A booklet so you can keep track of your schedule for training days, injection days, and clinic visits; also provides space for you to record your weekly dose and any problems you may experience with home preparation and administration Nplate for ITP (a brief summary) 4 Why platelet counts are lowered in ITP How Nplate helps your body make more platelets Treatment with Nplate 5 Taking the right dose at the right time Possible side effects Return of ITP symptoms? Training for home administration 6 The 4 steps of home administration training 7 Quick guide checklist 8 Before training you will probably have questions for your healthcare professional around some of the issues listed below. All of these can be managed effectively through training. If you do have any remaining concerns, make sure you address these with your healthcare professional so that you feel confident and competent to prepare and administer the injection without their direct supervision. Also, make sure your healthcare professional completes the section titled Just in case you need support in your Self-administration diary. Underdosing or overdosing Please contact your healthcare professional with any questions or concerns you may have about ITP or Nplate. Incorrect storage of Nplate Contaminated work surface Contamination of the luer connector Poor attachment of the syringe Failure to prepare injection site 2 3

243 Date: 04 April 2016 Page 243 of 270 Nplate for ITP (a brief summary) Treatment with Nplate Why platelet counts are lowered in ITP Adult ITP is a type of condition called an autoimmune disease. An autoimmune disease is when the body s defence (immune) system attacks a normal, healthy part of the body. For people with ITP, the healthy part of the body being attacked are the platelets. The immune system attack on platelets that occurs in ITP causes two things: The body destroys platelets much more quickly than usual, causing a low platelet count and lower levels of an important natural chemical that makes your body produce new platelets The body damages the special cells that create platelets, so each one makes fewer platelets than healthy cells would How Nplate helps your body make more platelets Nplate treats ITP by increasing your platelet count, and thereby reducing your risk of bleeds. Nplate raises platelet counts by mimicking the natural processes in the body responsible for making platelets. Nplate is a type of medicine known as a TPO-receptor agonist (TPO-RA). TPO (thrombopoietin) is the natural chemical that your body produces to tell your bone marrow to make more platelets. TPO-RAs are treatments that act like your body s own TPO to increase the number of platelets that are produced. Taking the right dose at the right time The right dose Because of the way Nplate works, very small doses may have a strong effect on your platelet count. Therefore, it s important that you get the right dose every time you re treated. Make sure that you administer the exact dose your healthcare professional has prescribed. The right time It s important that you never skip a dose of Nplate. If you have missed a dose, contact your healthcare professional immediately for advice about when you should have your next dose. Also talk to your healthcare professional first if you want to stop taking Nplate for any reason. Nplate should not be stopped without advice and guidance from a healthcare professional. It is important to have a regular treatment schedule, even when you are travelling away from home. Possible side effects As with any medicine, Nplate may be associated with side effects. Headache is very common, and may affect more than 1 in 10 people. You can find a full list in the package leaflet which accompanies every pack of Nplate. If you do experience side effects let your healthcare professional know. Read your package leaflet for more information about possible side effects, and be sure you discuss any concerns with your healthcare professional. 4 Return of ITP symptoms? If you experience any symptoms that you think may be a sign of a low platelet count, contact your healthcare professional immediately. Your healthcare professional may need to run some tests to see if there is a reason why you may not be responding so well to Nplate. It is possible that your Nplate dose may need to be changed. If this happens, you will not be able to administer Nplate at home until your dose is stabilised. 5

244 Date: 04 April 2016 Page 244 of 270 Training for home administration The 4 steps of home administration training Your healthcare professional will give you the training you need to successfully prepare and inject Nplate by yourself. This training involves 4 steps as outlined on the following page. Remember to write your training schedule in your Self-administration diary Before you begin home administration, you should feel completely comfortable with preparing and administering your own weekly dose of Nplate. Although the training involves 4 steps, the time required to learn how to prepare and administer Nplate will vary across patients. The most important thing is that you feel confident to prepare and administer the injection without the direct supervision of a healthcare professional. The training process involves the 4 steps described in the following table. After you ve begun treating yourself with Nplate at home, you ll need to attend the clinic once every 4 weeks for blood tests (or more often as required). STEP 1 STEP 2 STEP 3 Your health professional shows you how to prepare and administer Nplate Your healthcare professional will use the Nplate preparation mat and Step-by-step guide to demonstrate all the preparation and administration steps you will perform at home. You will receive your Nplate dose at this time. Your healthcare professional will complete the section titled Just in case you need support... in the your Self-administration diary, so you know who to contact in case of any questions. You prepare and administer Nplate under the direct supervision of your health professional Your healthcare professional will watch as you prepare and administer the correct dose of Nplate. Before you begin home administration, they will make sure you can prepare and administer the right dose of Nplate without additional assistance. You may perform this step more than once, until both you and your healthcare professional are confident in your ability to selfadminister Nplate. You prepare and administer Nplate at home You will prepare and self-administer Nplate at home. Be sure to let your healthcare professional know if you had any problems, if you were able to administer the right dose, if you had any problems or have any concerns. To make sure this step goes smoothly, write down the date and time you plan to administer your first at-home injection of Nplate. Record the date and time in your Self-administration diary. 6 Step 4: return for a blood test once every 4 weeks after you ve begun home administration of Nplate, or as often as required. At your first 4-weekly blood test, you will need to show your healthcare professional that you can prepare and self-administer Nplate correctly. 7

245 Date: 04 April 2016 Page 245 of 270 Quick guide checklist Each time you self-administer Nplate please make sure that you follow this checklist. The right steps The right materials Correct storage of Nplate The right dose of Nplate Cleanliness Firm connections Disposal of needles and vials Follow all the preparation and administration instructions in the Step-by-step guide Use the Nplate preparation mat to make sure you have all the materials you need, and check to make sure that the Nplate self-administration kit has not expired Nplate should be kept cool and out of direct light. Store your Nplate in the refrigerator (but not the freezer) and in its original box. This medicine may be removed from the refrigerator for a maximum period of 24 hours at room temperature (up to 25 C). It must be returned to the refrigerator for storage longer than 24 hours Even in small doses Nplate can have an influence on platelet production, which is why you need to take the correct dose so remember to check that you are administering the exact dose that your healthcare professional has prescribed. Your healthcare professional will tell you the correct dose to administer. This will be recorded in your Self-administration diary If you inject too much or too little Nplate, contact your healthcare professional immediately. Use the information your healthcare professional wrote in your Self-administration diary, in the section titled Just in case you need support... Clean hands: wash your hands with soap and water and dry them thoroughly with a clean towel before you begin your preparation and administration of Nplate Clean work surface: clean your Nplate preparation mat with an alcohol swab before placing the materials on it Clean tools: All the preparation and injection materials in your Nplate self-administration kit are clean and sterile. To ensure the materials remain sterile, remove them from the packaging only when instructed (as in the Step-by-step guide), and keep items in the air while preparing them Clean skin: clean the injection site with a new alcohol swab immediately before you inject Nplate Make sure you follow the instructions and your training when attaching parts of the kit. For the connection between the vial and the syringe you may feel a small resistance when they twist together, however, when attaching the syringe tip (Luer-lok) to the needle you will need to apply stronger force After use, please dispose of the needles, syringes and vials safely in a sharps container

246 Date: 04 April 2016 Page 246 of 270 Step-by-step guide to PREPARATION AND ADMINISTRATION of the Nplate injection A visual aid for your home-administration of Nplate This booklet will guide you through all the steps you ll need to perform to successfully prepare and administer your Nplate injection at home. During training, your healthcare professional will have shown you all of the steps in this Step-by-step guide. When you administer at home, this booklet will remind you of all the preparation and administration steps, so you can be confident you re performing every step correctly. Reporting of side effects If you get any side effects, talk to your doctor, pharmacist or nurse. This includes any possible side effects not listed in the package leaflet. You can also report side effects directly via the Yellow Card Scheme at By reporting side effects you can help provide more information on the safety of this medicine. Please contact your HEALTHCARE PROFESSIONAL with any questions or concerns you have

247 Date: 04 April 2016 Page 247 of 270 Before you begin Read all instructions for use thoroughly. The instructions in this Step-by-step guide are intended for patients who have already been trained in self-injection by their healthcare professional. If you have not been trained, please contact your healthcare professional and do not self-administer Nplate. A few notes on storage and use Keep your Nplate in the original package and in the refrigerator until you re ready to use it Nplate should not be frozen Nplate should be protected from light Nplate should be kept cool (refrigerated at 2ºC to 8ºC), but should not be frozen. Nplate may be removed from the refrigerator for a maximum period of 24 hours at room temperature (up to 25 C). It must be returned to the refrigerator for storage longer than 24 hours. Do not use this medicine after the expiry date. The expiry date is stated on the carton and vial label after EXP. The expiry date refers to the last day of that month Once Nplate has been dissolved, inject immediately you may have excess Nplate left over after administering your prescribed dose. Do not re-use Nplate! Any excess dissolved Nplate must be thrown away immediately after completing the injection process. Left over Nplate in the vial must NEVER be re-used for another injection

248 E.One Vial Adapter F.Alcohol Swab Packages Plastic cover 2010 Company. All rights reserved. 5/10 MC50370-A P44178 Product or Therapeutic Area: Romiplostim Date: 04 April 2016 Page 248 of 270 Step 1: Set up materials for an injection nplate preparation Mat 0AMUS012_55_MAT_L18alt:250/500 mcg 6/14/10 7:53 PM Page 1 First, prepare your workspace and the Nplate preparation mat: Place the mat on a clean, well-lit, flat surface (such as a table) Wash your hands with soap and water, and dry them thoroughly with a clean towel Clean the mat with an alcohol swab (item F) Mat You must be trained by your healthcare professional before you can administer Nplate Product Preparation and Injection Reference Mat Before you begin You will need: Before you Begin reporting of side effects read all instructions for use thoroughly. if you get any side effects, talk to your doctor, pharmacist or nurse. this includes You should establish a regular treatment schedule, even when travelling. any possible side effects not listed in the package leaflet. You can also report side Make sure you have all the supplies that you will need (refer to the images below). effects directly via the Yellow Card Scheme at Check the expiry date of the self-injection kit. By reporting side effects you can help provide more information on the safety of Do not use if expired or left at room temperature for more than 24 hours. this medicine. Wash your hands with soap. use an alcohol swab to clean the surface where you will be preparing product. You must be trained by your healthcare provider before you or your caregiver can administer product. Product Vial (500 mcg) (Visual A) One Prefilled Sterile Water Syringe and Plunger (Visuals B-1 and B-2) One Disposable 1 ml Syringe (Visual C) Safety Needle:27 G,½ inch (Visual D) One Vial Adapter (Visual E) Alcohol Swab Packages (Visual F) Sharps Container (not supplied) (Visual G) note: Please use this mat along with the Step-by-step guide and/or Package leaflet and Self-administration DVD to prepare your Nplate injection. You may be required to use more than one injection of Nplate. If your doctor has instructed you to take more than one injection of Nplate, follow the steps as defined in the Step-by-step guide to complete your prescribed dose of product using a new self-administration kit(s). Enter your required injection volume here (from HCP): ml Preparation and Administration of Product 1. Remove the cover from the product vial (A)and clean the rubber stopper with an alcohol swab (F). Make sure you have the following materials before you begin 2. Peel off the protective cover of the vial adapter (E),but do not remove it from the package.keeping the product vial (A) on a table,push the vial adapter (E) down onto the center of the vial s rubber stopper until it is firmly in place. 3. To assemble the prefilled sterile water syringe (B-1,B-2),attach the plunger rod (B-1) to the syringe (B-2) by twisting the plunger rod A product vial (1) B1 plunger rod (1) B2 prefilled sterile water syringe (1) (B-1) clockwise onto the syringe (B-2) until you feel a slight resistance. 4. To remove the white plastic cover of the prefilled syringe (B-2),hold the syringe at its base with one hand and bend the tip of the white plastic cover downward with your other hand.this will break the seal of the white plastic cover. 5. Double-check that the vial adapter (E)is securely in place,and remove the packaging.keeping the product vial(a) onthetable,attachthe water-filledsyringe(b-1,b-2) to the vial adapter (E) by twisting the syringe tip clockwise onto the vial adapter until you feel a slight resistance. 6. Slowly and gently expel water into the product vial (A).Water should flow slowly onto the powder.do not force water into the vial. 7. Gently swirl the product vial to dissolve the powder.do not shake the C Disposable 1 ml vial.this may take as long as 2 minutes.once fully dissolved, product should be clear and colorless. 8. Once the powder is completely dissolved,remove the syringe (B-1, the syringe (B-1,B-2) into the sharps container (G). syringe with Luer-lok (1) B-2) from the vial adapter (E) by twisting counterclockwise.discard Patients should establish a regular treatment schedule, even when traveling. Make sure you have all the supplies that you will need. Check the expiration dates on all vials. Do not use if expired. Wash your hands with soap. Use alcohol swab to clean the surface where you will be preparing product. NOTE: You may be required to use more than one injection of product.if your doctor has instructed you to take more than one injection of product,follow the steps as defined in the Patient Instructions for Use to complete your prescribed dose of product using a new reconstitution kit(s). 10. Attach the 1 ml syringe (C)to the vial adapter (E)of solution by twisting the syringe tip clockwise onto the reconstituted until you feel a slight resistance,and slowly expel air into vial the adapter vial(a). 11. Keeping the plunger at the base of the syringe(c),turn vial (A) assembly and syringe upside down,so the vial is above the syringe. 12. Withdraw all of the liquid into the syringe (C).Do not pull plunger past the 1 ml marking. 13. Remove all air bubbles bygentlytapping the barrel of the syringe(c). Once the air bubbles have risen to the top,gently push them back into the vial(a) by slowly pushing the plunger. 14. Ensure the syringe(c)has the correct amount for your dose (doublecheck the injection volume entered above) by pushing the plunger of the syringe to expel any excess liquid back into the vial (A). 15. Remove syringe (C)from the vial adapter (E)by twisting counterclockwise,but make sure to keep the syringe in your hand. 16. While holding the syringe (C)in your hand with the tip facing up, remove the 27 G needle*(d) from its packaging by peeling apart the tabs. 17. Attach the needle (D)onto the filled syringe (C)by twisting clockwise until you feel a slight resistance.pull back the pink needle safety shield,and then remove the white needle cover by pinching it firmly and pulling in the opposite direction of the syringe assembly. 18. Open a new alcohol swab package (F)and clean the injection site. When giving the injection,use one hand to gently pinch the cleaned areaofskinandholditfirmly.withtheotherhand,holdthesyringe(c) at a 45-degree angle to the skin. A.Product Vial (500 mcg) B-1.One Plunger Rod B-2.One Prefilled Sterile Water Syringe C.One Disposable 1 ml Syringe E.One Vial Adapter F.Alcohol Swab Packages D Safety needle: 27 G, 13 mm (1) G.Sharps Container (not supplied) 9. Remove the 1 ml syringe (C)from its package and pull the plunger to the 1 ml marking.do not pull plunger past the 1 ml marking. Please call X-XXX-XXX-XXXX (X-XXX-XXX-XXXX) for more information. 19. After injecting your product dose,activate the pink needle safety place the entire syringe assembly(c,d) into the sharps container (G). shield(d) with one hand until you hear it click into place,and then D.Safety Needle:27 G,½ inch *Yourhealthcareprovidermayrecommendasmaller29-or31-gaugeneedleforinjection. Storeproductvialsintheircartontoprotectfromlightuntiluse. E Vial adapter (1) F alcohol swab packages (x4) G Sharps container (1; not supplied) NOTE:These images are for reference only. Store product vials in their carton to protect from light until use. This medicine may be removed from the refrigerator for a maximum period of 24 hours at room temperature (up to 25 C). It must be returned to the refrigerator for storage longer than 24 hours. B-2.One Prefilled Sterile Water Syringe Plastic cover NOTE: These images are for reference only. Your preparation area G.Sharps Container (not supplied) only Company. All rights reserved. 5/10 MC50370-A P44178 Job code: NPO-GBR-AMG NP Date of prep: April 2014 Step 1

249 Date: 04 April 2016 Page 249 of 270 Step 1: Set up materials for an injection Once your workspace is clean and ready, take the Nplate self-injection kit out of the refrigerator. Look closely at the powder in the Nplate vial, to see if it s frozen. Do not use if frozen. If you have any questions about storage, contact your healthcare professional for further instructions Check the expiry date on the self-injection kit. If the expiry date has passed, do not use. Stop and contact your healthcare professional Note: If your healthcare professional has instructed you that your Nplate dose requires more than one injection of Nplate, you will need to use more than one self-injection kit. Follow the steps as described in this Step-by-step guide and use as many self-injection kits as necessary to complete your prescribed dose of Nplate.

250 Date: 04 April 2016 Page 250 of 270 Make sure you have the following items. These are the items you will need to prepare and administer Nplate. These items are organised the same way they are on your Nplate preparation mat, which you should use whenever you prepare your Nplate injection. Note: Do not open items until directed to do so in the instructions. Do not use components that have evidence of tampering or damage. Do not re-use items. The Sharps container is not included in your self-administration pack, and should be obtained before initiating self-administration. A Product vial (1) B1 Plunger rod (1) B2 Prefilled water syringe (1) xc Disposable 1 ml syringe with Luer-lok D Safety needle: 27 G, 13 mm (1) E Vial adapter (1) F Alcohol swabs (x4) G Sharps container (1; not supplied) Step 1

251 Sterile Tampon de preparation a l alcool Sterile Tampon de preparation a l alcool Sterile Tampon de preparation a l alcool Sterile Tampon de preparation a l alcoo Sterile Tampon de preparation a l alcool Sterile Tampon de preparation a l alcool Sterile Tampon de preparation a l alcool Sterile Tampon de preparation a l alcoo Product or Therapeutic Area: Romiplostim Date: 04 April 2016 Page 251 of 270 Step 2: Prepare vial for use, attach vial adapter This step uses 1 alcohol swab (item F), 1 Nplate vial (item A), and 1 vial adapter (item E) Remove red (250 micrograms) or blue (500 micrograms) plastic cap from vial Using a new alcohol swab clean vial stopper Do not touch vial stopper after cleaning it Peel off paper backing slowly from vial adapter while keeping vial adapter in the plastic package Do not touch vial stopper or spike of vial adapter FLIP OFF Alcohol Prep Pad Alcohol Prep Pad

252 Date: 04 April 2016 Page 252 of 270 Keeping the vial on a table, and keeping the vial adapter in the plastic packaging, line up spike on the vial adapter to the centre of the stopper on the vial Push the vial adapter down onto the vial until it is firmly in place and you can t push down any more Lift off plastic vial adapter packaging, leaving vial adapter on vial Do not touch the syringe connector Step 2

253 Product or Therapeutic Area: Romiplostim Date: 04 April 2016 Page 253 of 270 Step 3: Prepare sterile water syringe This step uses 1 plunger rod (item B1) and 1 pre-filled water syringe (item B2) Before you begin step 3 please note the following: The clear plastic plunger rod MUST always be attached first before breaking the white tip off of the pre-filled water syringe. Perform step 3a before step 3b. Step 3a: Attach clear plastic plunger rod to pre-filled sterile water syringe by placing the threaded end of the plunger rod into the syringe and carefully twisting the rod clockwise onto the grey syringe plunger, until you feel a slight resistance. Do not over tighten Step 3b: Holding the syringe with one hand, bend the tip of the white plastic cover downward with your other hand. This will break the seal of the white plastic cover Once the seal is broken, pull the white plastic cover off. You will see grey rubber in the cap

254 Date: 04 April 2016 Page 254 of 270 Step 4: Dissolving Nplate by injecting water into vial This step uses the syringe you just assembled in step 3, plus the vial and adapter assembly from step 2 Before you begin step 4 please note the following: Do dissolve Nplate powder slowly and carefully. This is a protein product and proteins can be easily damaged by improper mixing and excessive shaking. Keeping the vial on the table, attach water-filled syringe to vial adapter by holding the side of the vial adapter with one hand and twisting the syringe tip clockwise onto the adapter with the other hand until you feel a slight resistance Step 3 Step 4

255 Date: 04 April 2016 Page 255 of 270 Step 4: Dissolve Nplate by injecting water into vial Very slowly and gently push down on plunger rod to inject all water in the syringe into the vial. Water must flow slowly onto powder Do not force the water into the vial Note: After injecting the water into the vial it is common for the plunger to move back up. You do not have to maintain pressure on the plunger for the rest of step 4.

256 Date: 04 April 2016 Page 256 of 270 Correct Incorrect Before continuing: Do ensure that all water is injected from the syringe into the vial before dissolving. Holding the area where the vial and vial adapter connect between your fingers, gently swirl the vial by rotating your wrist until all of the powder has dissolved and the liquid in the vial is clear and colourless Do not shake the vial Do not roll vial between palms Note: It may take up to 2 minutes for the powder to completely dissolve Step 4

257 Date: 04 April 2016 Page 257 of 270 Step 4: Dissolve Nplate by injecting water into vial Before continuing: do visually inspect the dissolved liquid for particles and/or discolouration. It must be clear and colourless and fully dissolved Note: If there is any colour or particles in the liquid, contact your healthcare professional do make sure liquid is fully dissolved before removing syringe When Nplate is completely dissolved, remove the empty syringe by twisting it anti-clockwise off of the vial adapter Discard the empty syringe into sharps or hazard container. Keep the dissolved Nplate vial. Immediately prepare new syringe for injection Do not delay injecting Nplate