Kuter 2008a.

Methods	Randomized, double‐blind, placebo‐controlled trial. Phase III. 6‐month study. Multicenter (35 sites in the USA and Europe).
Participants	Chronic ITP, a history of ITP for at least 6 months, PLT≤ 30 × 109/l, had a splenectomy for the treatment of ITP greater than or equal to 24 weeks prior to study entry, the dose of maintenance immunosuppressive regimens, primarily glucocorticoids must have been stable for at least 1 month, age ≥ 18 years, n = 63
Interventions	Romiplostim versus placebo The starting dose (romiplostim or placebo) was 1 μg/kg. To achieve the target platelet count of 50 × 10/L to 200 × 10/L, doses could be increased according to the following algorithm: 2 μg/kg every week if the count was 10 × 10/L or less and 2 μg/kg every 2 weeks if 11 × 10/L to 50 × 10/L. Once platelets reached more than 50 × 10/L, the maintenance algorithm was used: dose was increased by 1 μg/kg every week if 10 × 10/L or less; increased by 1 μg/kg after 2 weeks if 11 × 10/L to 50 × 10/L; reduced by 1 μg/kg after 2 consecutive weeks at 201 × 10/L to 400 × 10/L; withheld if more than 400 × 10/L and subsequent doses reduced by 1 μg/kg and given after count was less than 200 × 10/L. The maximum allowed dose was 15 μg/kg.
Outcomes	Primary: durable platelet response Key secondary endpoints: overall platelet response; number of weeks with platelet response; proportion of subjects requiring rescue medication; incidence of durable platelet response with stable dose Descriptive secondary endpoints: adverse events; proportion of subjects able to reduce or discontinue concurrent ITP therapies during the first 12 weeks of treatment; health‐related quality of life (HRQoL)
Notes	Trial closed May 2009 Study supported by Amgen Amgen designed the study, did statistical analyses and interpreted the data, which Amgen holds. Amgen collected the data and representatives (JLN and DMG) participated in the writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication. Study was also reported in George 2009 and Pullarkat 2009 (see Kuter 2008a for references)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	A ‐ Adequate. Quote: "The random allocation sequence was generated by Amgen Inc with the blocked randomisation method".
Allocation concealment (selection bias)	Low risk	A ‐ Adequate. Quote: "Clinphone was used to randomly assign patients into the study with the interactive voice response system (IVRS)".
Blinding (performance bias and detection bias) Physician assessed	Low risk	Quote: "Double blind (subject, investigator)". "Romiplostim and placebo were supplied in identical vials containing a lyophilised powder that was reconstituted with sterile water for subcutaneous injection."
Blinding (performance bias and detection bias) Patient assessed	Low risk	Quote: "Double blind (subject, investigator)". "Romiplostim and placebo were supplied in identical vials containing a lyophilised powder that was reconstituted with sterile water for subcutaneous injection."
Incomplete outcome data (attrition bias) Efficacy outcomes	Low risk	95% romiplostim group and 90% placebo group completed the trial. Judged a low risk of bias as efficacy analysis included all patients treated.
Incomplete outcome data (attrition bias) Safety outcomes	Low risk	95% romiplostim group and 90% placebo group completed the trial. Judged a low risk of bias as safety analysis included all patients treated.
Selective reporting (reporting bias)	Low risk	Quote: "All outcome measures were prospectively defined before the start of the studies. A weekly platelet response was defined as a platelet count of 50 × 10/L or more at a weekly study visit. Unless otherwise noted, platelet responses that occurred within 8 weeks after rescue drugs were used were not included in the efficacy analyses or in the determination of any other measures for platelet outcome. Rescue medication was defined as an increased dose of concurrent ITP therapy, or the use of any new drug to increase platelet counts. A durable platelet response (primary efficacy measure) was defined as weekly platelet responses during 6 or more weeks of the last 8 weeks of treatment. Patients who received rescue medication at any time during the study could not be counted as having a durable response. A transient platelet response was defIned as four or more weekly platelet responses without a durable platelet response from week 2 to 25. Additional secondary endpoints were the frequency of overall platelet response (durable plus transient rates of platelet response), the number of weekly platelet responses, the proportion of patients needing rescue drugs, and the frequency of durable platelet response with a stable dose (dose maintained within 1 μg/kg during the last 8 weeks of treatment). We also assessed changes in concurrent ITP therapies."
Other bias	High risk	Funded by drug company. There is evidence that industry‐sponsored trials may overestimate the treatment effect (Bhandari 2004).