Pooled Analysis of Safety and Efficacy of Romiplostim in Splenectomized and Nonsplenectomized Patients (pts) with Immune Thrombocytopenia (ITP)

Introduction: Romiplostim is a thrombopoietin receptor agonist approved for use in adult pts with ITP. Pooled analyses of combined pt data from romiplostim ITP clinical studies have previously been reported. Here we report updated safety and efficacy data according to baseline splenectomy status.

Methods: Data from adult pts in 13 completed ITP studies with romiplostim were analysed up to June 2014. Pts received romiplostim, placebo or medical standard of care (SOC) and data from the placebo/SOC arms were pooled. All 13 studies were included in analyses of baseline pt characteristics and safety endpoints; 4 early dose-finding studies were excluded from analyses of efficacy endpoints as they do not reflect the current dosing of romiplostim. Adverse events were adjusted for time spent on study and reported as rates per 100 pt-years. For pts who started their parent study in the placebo/SOC group and then went on to receive romiplostim in an extension study, all data from the first dose of romiplostim were included in the romiplostim group. A platelet response was defined as a platelet count ≥50x10⁹/L without rescue medication in the previous 8 weeks; a platelet response for 9 out of any 12 consecutive weeks on-study was considered a sustained platelet response. All analyses were descriptive and no statistical testing was performed.

Results: Data from 1,111 pts were analysed, 395 splenectomized and 716 nonsplenectomized. The splenectomized and nonsplenectomized groups were similar in age (median 52 vs 53 years) and sex (female 64% vs 60%), but in the splenectomized group median baseline platelet counts were slightly lower (14 vs 19x10⁹/L) and a higher proportion of pts were known to have received >3 prior ITP treatments (38% vs 12%) than the nonsplenectomized group. Rates of AEs, serious AEs, fatal AEs, treatment-related AEs, thrombotic events, and hemorrhages were lower in nonsplenectomized than splenectomized pts and were in general lower in romiplostim than placebo/SOC-treated pts in both groups. Bone marrow reticulin occurred in 17 romiplostim-treated pts and one placebo-treated pt, at a slightly increased rate in splenectomized vs nonsplenectomized pts. Bone marrow collagen was reported in one romiplostim-treated nonsplenectomised pt. Data from 1,024 pts were analysed for efficacy (376 splenectomized, 648 nonsplenectomized). The median (Q1, Q3) most frequent weekly dose was 4 µg/kg (2, 9) in splenectomized and 3 µg/kg (2, 7) in nonsplenectomized pts. A platelet response was achieved in 82% of splenectomized and 91% of nonsplenectomized pts and a sustained platelet response in 66% and 79%, respectively. The median time to first response was 2.1 weeks for splenectomized and 2.0 weeks for nonsplenectomized pts. Platelet responses were maintained in those who responded: after the first response the median (Q1, Q3) proportion of time with a response was 97% (79%, 100%) for splenectomized and 100% (91%, 100%) for nonsplenectomized pts.

Conclusions: A relatively large number of nonsplenectomized pts have received romiplostim in clinical studies. Safety of romiplostim was comparable in splenectomized and nonsplenectomized patients with no new safety signals observed, and platelet response rates were high and of sustained duration in both groups.

* Bone marrow collagen reported in one romiplostim-treated nonsplenectomized pt. Excludes study NCT00907478 as bone marrow evaluations were collected differently than in other studies; in the romiplostim arms the N/pt-yrs were 331/560.6 for splenectomised and 546/866.7 for nonsplenectomised pts.