A systematic review of higher-risk myelodysplastic syndromes clinical trials to determine the benchmark of azacitidine and explore alternative endpoints for overall survival

• Published in: Leukemia research Vol. 104; p. 106555
• Main Authors: Garcia, Jacqueline S., Swords, Ronan T., Roboz, Gail J., Jacoby, Meagan A., Garcia-Manero, Guillermo, Hong, Wan-Jen, Yang, Xiaoqing, Zhou, Ying, Platzbecker, Uwe, Steensma, David P., Wolff, Johannes E., Fenaux, Pierre
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.05.2021
• Subjects: Azacitidine; Higher-Risk; Hypomethylating agent; Myelodysplastic syndromes; Overall survival; Overall survival; Azacitidine; Myelodysplastic syndromes; Higher-Risk; Hypomethylating agent
• ISSN: 0145-2126; 1873-5835; 1873-5835
• DOI: 10.1016/j.leukres.2021.106555

•Survival rates reported among HR-MDS patients are inconsistent across clinical studies.•Real-world OS with azacitidine is consistently shorter compared to that observed in the large pivotal trial AZA-001.•We investigated surrogate endpoints for patients with HR-MDS treated with azacitidine.•This systematic review determined clinical benchmarks for future studies of HMA-based combinations.•Benchmarks of response from 237 clinical studies included complete remission (CR) rate, marrow CR, and overall survival. The hypomethylating agent azacitidine can prolong overall survival (OS) in patients with higher risk-myelodysplastic syndromes (HR-MDS) compared to conventional regimens. However, outcomes differ largely between studies, making it challenging to determine the contribution of novel therapies added to azacitidine. Further, a discrepancy is seen between complete (CR) or partial (PR) response rates and OS improvement with azacitidine, making it challenging to rely on earlier endpoints than OS. We conducted a systematic literature search and study-level systematic review of 237 clinical studies to better understand outcomes for HR-MDS patients treated with azacitidine. Pooled marrow CR was 9% (N = 2654; 95% CI: 6–13 %), CR rate was 17 % (N = 6943; 95% CI: 15–20 %), and median OS (mOS) was 18.6 months (N = 2820; 95% CI: 15.3–21.9). A weak correlation to mOS was detected with CR rate (207 patient cohorts, Pearson’s r = 0.315; P < 0.0005), and a much stronger correlation with median progression-free survival (mPFS) (r=0.88, P = 3 × 10−14). Six-months progression-free survival rates correlated with 1-year OS rates but were only infrequently reported (N = 41 patient cohorts) therefore not allowing a robust recommendation for a surrogate to the established OS endpoint. Larger patient numbers and patient-level data appear necessary, especially for designing future clinical trials using azacitidine combinations.