Scientific programme

OS01  Oral Session
ORAL SESSION 1

04-May-2017 15:00 16:30
 
 
Abstract:
ROMIPLOSTIM IN LOW/INT-1-RISK MDS RESULTS IN REDUCED BLEEDING WITHOUT IMPACTING LEUKEMIC PROGRESSION: FINAL RESULTS FROM A RANDOMIZED, DOUBLE-BLIND, PLACEBO-CONTROLLED STUDY

Background: In a placebo (PBO)-controlled study, 250 patients with MDS were randomized 2:1 to receive weekly romiplostim or PBO.  In the original 2011 analysis, romiplostim reduced clinically significant bleeding events (P=0.13) and platelet transfusions (P<0.001) and increased IWG HI-P rates (P<0.001).  Peripheral blast count increases >10% were more frequent with romiplostim (15% vs. 3.6%) and mostly resolved after discontinuation.  In 2011, the DMC recommended study drug be discontinued as the potential benefit did not outweigh the potential risk for disease progression to AML or diagnosis of and treatment for AML (from increased blasts).  Previously reported (Giagounidis et al, Cancer 2014) 58-week incidence of AML was 6.0% (N=10) for romiplostim and 4.9% for PBO (N=4); HR 1.20 (95%CI:0.38−3.84).  

Methods: Eligible patients had IPSS low/int-1 risk MDS and platelets 1) ≤20×109/L or 2) ≤50×109/L with a history of bleeding.  Disease progression to AML was defined as 1) ≥20% blasts in bone marrow or peripheral blood 4 weeks after romiplostim discontinuation; 2) per pathology; or 3) antileukemic treatment.  Results are presented by treatment group.

Results:  At baseline, 27.6% of patients were IPSS low risk and 72.4% int-1, WHO classifications were RCMD:67.6%, RAEB-1:13.2%, MDS-U:11.2%, RA:4.4%, RCMD-RS:2.4%, RARS:0.8%, and RAEB-2:0.4%.  Of 250 patients in the study, 210 entered long-term follow-up (LTFU); median (Q1, Q3) follow-up was 27.5 (10.8, 58.7) months.  During the active study period and LTFU, mortality rates were romiplostim:55.7% and PBO:54.2%, AML rates were romiplostim:11.9% and PBO:11.0% (HR 1.06, 95%CI:0.48−2.33), and rates of AML/death were romiplostim:56.9% and PBO:55.4% (HR 1.04, 95%CI:0.73−1.48).  Nearly half (N=14, 48%) of the 29 AML cases occurred in patients who were RAEB-1 at screening (none RAEB-2), and 6 cases were diagnosed because of anti-AML treatment use alone.   

Conclusions: Following the 2011 decision to stop study drug secondary to increased AML rates at that time and transient blasts increases, final 5-year LTFU HRs (romiplostim vs placebo) for death or progression to AML, respectively, are 1.03 (95%CI:0.72−1.47) and 1.06 (95%CI:0.48−2.33).   In conclusion, romiplostim reduced bleeding events and platelet transfusions, with no increase in AML incidence or impact on survival.


 
Co-authors
H. Kantarjian 1, P. Fenaux 2, M. Sekeres 3, J. Szer 4, U. Platzbecker 5, A. Kuendgen 6, G. Gaidano 7, W.W. Jedrzejczak 8, N. Carpenter 9, B. Mehta 10, J. Franklin 10, A. Giagounidis 11
1MD Anderson Cancer Center, Leukemia, Houston, USA
2Hopital Avicenne Université Paris XIII, Service d'Hématologie Clinique, Bobigny, France
3Cleveland Clinic, Leukemia Program- Taussig Cancer Institute, Cleveland, USA
4Royal Melbourne Hospital, Clinical Haematology, Melbourne, USA
5Medizinische Klinik und Poliklinik I, University Hospital Carl Gustav Carus Dresden, Dresden, Germany
6Medizinische Klinik und Poliklinik A, Heinrich-Heine-University of Düsseldorf, Dusseldorf, Germany
7Amedeo Avogadro University of Eastern Piedmont, Division of Hematology- Department of Translational Medicine, Novara, Italy
8Medical University of Warsaw, Independent Public Central Hospital Clinic, Warsaw, Poland
9Amgen Ltd., Global Biostatistics, Uxbridge, United Kingdom
10Amgen Inc., Global Development, Thousand Oaks, USA
11Marien Hospital, Clinic for Oncology- Hematology- and Palliative Medicine, Düsseldorf, Germany