Abstract
Background In higher-risk myelodysplastic syndromes (MDS), the standard doses of hypomethylating agents are azacitidine 75 mg/m2 for 7 days and decitabine 20 mg/m2 for 5 days. In lower-risk MDS, shorter durations of hypomethylating agents may be effective with reduced toxicity. We aimed to show how abbreviated durations of hypomethylating agents impact the natural history of lower-risk MDS.
Methods We conducted a randomized, multicenter study of 3-day decitabine, 3-day azacitidine, and 5-day azacitidine in low- or intermediate-1-risk MDS or chronic myelomonocytic leukemia (NCT02269280). The protocol separated patients into transfusion-dependent and -independent cohorts. Transfusion dependence was defined as requiring a red blood cell or platelet transfusion at any point before therapy. Transfusion-independent patients were also randomized to a best supportive care arm. Patients were assigned through Bayesian adaptive randomization to the arm with better event-free survival (EFS).
The primary end point was EFS, defined as the time from randomization until the date of an event (no response after six cycles, loss of response, increasing bone marrow blast count, or discontinuation due to adverse events). Key secondary end points were overall survival (OS), transfusion independence, and response attainment by the International Working Group 2006 criteria. The overall response rate (ORR) was defined as complete response (CR), hematological improvement, and marrow CR divided by the number of evaluable patients.
Results We randomized 247 patients: 151 transfusion-dependent and 96 transfusion-independent. The median age was 70.8 years (range, 24 to 88 years), 25% were female, and 13% were from Black, Asian, Hispanic, or Middle Eastern backgrounds. The median International Prognostic Scoring System (IPSS) score was 0.5 (range, 0 to 1.5) and the median IPSS-R score was 3 (range, 0 to 5.5). Among 230 (93%) patients with next-generation sequencing, mutation frequencies were: 31% ASXL1, 19% SF3B1, 14% RUNX1, 13% DNMT3A, and 8% TP53. Among transfusion-dependent patients, the ORR was 53% (95% CI, 40% to 65%) for 3-day decitabine, 53% (95% CI, 36% to 69%) for 3-day azacitidine, and 48% (95% CI, 35% to 61%) for 5-day azacitidine (P = 0.89). The proportion of red blood cell or platelet transfusion-dependent patients who became transfusion-independent was 44% (95% CI, 31% to 57%) among patients treated with 3-day decitabine, 42% (95% CI, 26% to 59%) with 3-day azacitidine, and 40% (95% CI, 28% to 54%) with 5-day azacitidine (P = 0.97).
Among transfusion-independent patients, the ORR was 50% (95% CI, 27% to 73%) for 3-day decitabine, 54% (95% CI, 35% to 71%) for 3-day azacitidine, 70% (95% CI, 49% to 84%) for 5-day azacitidine, and 17% (95% CI, 9% to 56%) for best supportive care (P = 0.14). Among transfusion-independent patients, the response duration with 3-day decitabine or 5-day azacitidine was longer than 3-day azacitidine (36.9 months versus 28.9 months versus 15.3 months, respectively; P = 0.05).
Among treated patients, the 30-day mortality rate was 1% and the 60-day mortality rate was 2%. Regardless of transfusion dependence, grade ≥3 thrombocytopenia occurred more frequently with decitabine (45%) than with 3-day azacitidine (27%; P = 0.04) or 5-day azacitidine (31%; P = 0.06).
In a multivariate Cox model of transfusion-dependent patients controlling for age, performance status, disease risk, and molecular characteristics, the EFS of 5-day azacitidine was better than 3-day azacitidine (hazard ratio [HR], 0.41; 95% CI, 0.19 to 0.90; P = 0.03); there was no difference between 5-day azacitidine and 3-day decitabine (HR, 0.74; 95% CI, 0.41 to 1.33; P = 0.31). However, the OS of 5-day azacitidine was better than 3-day decitabine (HR, 0.46; 95% CI, 0.26 to 0.82; P = 0.01). In transfusion-independent patients, the EFS of 5-day azacitidine was better than 3-day decitabine (HR, 0.28; 95% CI, 0.10 to 0.77; P = 0.01), 3-day azacitidine (HR, 0.25; 95% CI, 0.09 to 0.71; P = 0.01), and best supportive care (HR, 0.09; 95% CI, 0.01 to 0.57; P = 0.01). The OS of 5-day azacitidine was better than 3-day decitabine (HR, 0.13; 95% CI, 0.04 to 0.45; P < 0.01) and best supportive care (HR, 0.21; 95% CI, 0.06 to 0.77; P = 0.02).
Conclusions Shorter courses of hypomethylating agents are safe and effective in lower-risk MDS. Five-day azacitidine provides the best balance of safety and efficacy in lower-risk MDS, producing the most consistent EFS and OS benefit.