Romiplostim for chemotherapy-induced thrombocytopenia (CIT) in colorectal, gastroesophageal, and pancreatic cancers: A global, phase 3, randomized, placebo-controlled trial (RCT).

Abstract

12007 Background: CIT is a common consequence of antineoplastic regimens for gastrointestinal (GI) cancers, occurring in >60% of colorectal cancer patients receiving multiagent chemotherapy. CIT can lead to chemotherapy dose reduction, delay, omission, and discontinuation, potentially worsening outcomes. There are no widely available licensed therapies for this unmet need. Aim: To evaluate the safety and efficacy of the thrombopoietin receptor agonist romiplostim (ROMI) in patients with GI cancers to limit chemotherapy dose modifications from CIT. Methods: This was a phase 3, placebo (PBO)-controlled RCT of patients receiving oxaliplatin-based multiagent regimens for GI cancers with persistent CIT, ie platelets (Plt) ≤85×10 9 /L on day 1 of a scheduled chemotherapy cycle (NCT03362177). Patients from 55 sites in 14 countries were randomized 2:1 to ROMI or PBO for 3 chemotherapy cycles, stratified by baseline Plt (< or ≥50×10 9 /L) and cancer type. Study drug started at 2 μg/kg subcutaneous weekly, adjusted weekly by 1 μg/kg up to 10 μg/kg to target Plt ≥100×10 9 /L in 12 weeks (≤4 weeks at 10 μg/kg). Chemotherapy started when Plt ≥100×10 9 /L (Plt response) or after week 4 per investigator. The primary endpoint was no CIT-induced dose modification of any myelosuppressive agent in either the second or third chemotherapy cycle per independent adjudication committee. Results: Patients (N=165; 109 ROMI, 56 PBO) had colorectal (75%), gastroesophageal (13%), or pancreatic (12%) cancer; 60% were male, 90% White, 4% Black, and 24% Hispanic, with mean (SD) age of 61.4 (11.1) years. Baseline median (range) Plt was 69 (8–85)×10 9 /L; 11% had Plt <50×10 9 /L. Stage IV disease rates were ROMI 65%, PBO 55%. Most (75%) patients completed study drug; 3% discontinued due to adverse events (AEs). The primary endpoint was achieved in 92/109 (84%) patients receiving ROMI vs 20/56 (36%) receiving PBO (odds ratio 10.2; 95% CI 4.6-22.5; P<0.001). Median (range) Plt nadirs were ROMI 87 (14–167)×10 9 /L, PBO 58 (22–95)×10 9 /L; P=0.005. For those with Plt responses (ROMI 97%, PBO 77%), median (95% CI) time to first Plt response was ROMI 1.1 (not estimable) weeks, PBO 2.1 (1.1-3.0) weeks; P<0.001. Treatment-related (TR) AE rates were ROMI: 12%, PBO: 7%, most frequently nausea (2%, 2%) and headache (2%, 0%). TR serious AEs and TRAEs leading to death or discontinuation of study drug or chemotherapy were not observed in either arm. Conclusions: In this first global phase 3 RCT of ROMI vs PBO for CIT, ROMI was well tolerated and efficacious in the treatment and prevention of CIT in GI cancers. These results are potentially practice-changing for a common serious condition encountered routinely in clinical practice worldwide that prevents delivery of on-time, full-dose anticancer therapy. Final results from long-term follow-up will be presented. Clinical trial information: NCT03362177 .