Chronic myelomonocytic leukemia: Clinical outcomes from clinical trials submitted to the FDA over the past 25 years

Abstract

Abstract Introduction: Chronic myelomonocytic Leukemia (CMML) is a rare hematological malignancy with features of both myelodysplastic syndromes (MDS) and myeloproliferative neoplasms. Due to its rarity, clinical research in CMML has been limited, with current therapeutic approvals restricted to only three hypomethylating agents (HMAs): azacitidine, decitabine, and decitabine/cedazuridine. Clinical trials have typically included only small numbers of patients with CMML within broader MDS studies, leading to uncertainty about applicability to CMML populations. To address these gaps, this study aims to evaluate overall survival (OS) outcomes and response rates across different treatment modalities and across CMML subtypes. Methods We searched for phase 2 and 3 trials including patients with CMML submitted to FDA as part of New Drug Applications and Investigational New Drug Applications for the treatment of MDS between 2000 and 2025. We performed retrospective patient-level analyses across trials. CMML diagnosis was based on WHO and/or FAB diagnostic criteria. Per WHO classification, myelodysplastic CMML (MD-CMML) was determined based on WBC <13 × 10⁹/L and myeloproliferative CMML (MP-CMML) with WBC ≥13 × 10⁹/L. Overall response rate (ORR) was defined as achieving complete response (CR) or partial response (PR) generally using IWG 2000/2006 criteria (as defined in the protocol). OS was defined as time from randomization to death from any cause. The OS comparisons by treatment, CMML subtype, and ORR comparison groups were analyzed using Kaplan-Meier Methods with log-rank test, and Cox proportional hazards models. Results: We identified 10 trials evaluating 5 investigational agents for MDS treatment that included a total of 196 patients with CMML. These 196 patients were categorized into three treatment groups: HMA monotherapy (n=104), HMA combination therapy (n=63) and control (conventional care regimen or supportive care, n=29). The study population had a median age of 70 years, was 69% male, and included 85% White, 12% Asian, 1.6% African American, and 1.1% Hispanic patients. Geographical distribution was 48% from North America, 35% from Europe, 12% from Asia and 4% from Australia. Among all 196 patients, the median OS was 20.4 months (95% Confidence Interval (CI): 17.5-25.2) with 107 deaths (54.5%). MD-CMML patients (n=154) had a longer median OS of 21.0 months (95% CI: 18.0, 28.3) compared to 15.7 months (95% CI: 10.0, NE) for MP-CMML (n=41), with hazard ratio (HR) 0.82, 95% CI: 0.52-1.31. Median OS for all patients with CMML by treatment group was 22.3 months (95% CI: 16.2-32.3) for HMA monotherapy, 20.0 months (95% CI: 17.2-Not Estimable (NE)) for HMA combination therapy, and 14.8 months (95% CI: 11.1-NE) for control. Overall, 22% (44/196) of patients achieved CR or PR as their best response, with treatment-specific rates of: HMA 26% (27/104), HMA combination therapy 27% (17/63) and control 0% (0/29). Patients who achieved CR or PR (n=44) had longer median OS of 85.8 months (95% CI: 28.3, NE) compared to those who did not achieve CR/PR (n=152), who had a median OS of 15.7 months (95% CI: 12.8, 20.4), with an HR of 0.28 (95% CI: 0.16-0.50). This trend was also observed in MD-CMML, with patients who achieved CR/PR (n=39) having a median OS that was NE (95% CI: 23.7, NE) compared to 17.5 months (95% CI: 13.4, 21.8) for those who did not achieve CR/PR (n=115); the HR was 0.35 (95% CI: 0.19-0.64). However, this trend was not observed in patients with MP-CMML, likely due to the small sample size of only 5 patients who achieved CR/PR. These comparisons do not adjust for differences in time to response or differences between the treatment groups. Conclusion Clinical trials completed over the past 25 years have led to improved understanding and management of CMML. Our data confirm that patients treated with HMA alone or in combination had similar CR/PR rates (26% vs 27%) and median OS (22.3 vs 20.0), which are in line with outcomes reported in patients with MDS treated with HMAs. Patients who achieved CR/PR had longer median OS compared to those who did not achieve CR/PR. Patients with MD-CMML had longer OS than those with MP-CMML. Overall, treatment remains suboptimal with low CR rates and a modest OS outcome, especially in MP-CMML where outcomes were numerically worse. These findings highlight the need for CMML-specific trials and novel therapeutic strategies to improve outcomes for CMML patients, particularly those with MP-CMML.