MDS-519 Outcome Predictors and Timing Considerations for Allogeneic Stem Cell Transplantation in Chronic Myelomonocytic Leukemia

Abstract

Chronic myelomonocytic leukemia (CMML) outcomes are poor in general with a median overall survival (mOS) of 3 years. Allogeneic transplant (HSCT) offers the only curative option. Considerations as to when to pursue HSCT, identification of relapse predictors and assessment of whether transplant-related, clinical and molecular disease features have an impact on outcomes remain crucial. To characterize best timing to HSCT, best therapy to achieve CR prior to transplant, and analyze associations between molecular and transplant-related variables and relapse risk Design: Retrospective collection of clinical and molecular data of transplanted CMML patients treated at Moffitt Cancer Center between 1995-2022. Median duration of follow-up was 93 months. A total of 84 patients went to HSCT: 30 (36%) in BP-CMML (sAML), 36(43%) in CMML-0/ CMML-1(EarlyStage) and 15(18%) in CMML-2. Median age at HSCT was 63years (18-76). mOS for the transplant cohort was 62 months (95%CI 35-89 months). Patients who went to transplant earlier had better outcomes: mOS for ES-CMML: 202 months (95%CI NR), vs 62 months for BP-CMML (95%CI 39-84 moths) (p=0.019). There was no association between relapse rate and number of treatment lines prior to HSCT (p=.112), type of donor (p=.491), stem cell source (p=.768) or type of conditioning regimen (MA vs RIC/NMA) (p=.565). There was a trend toward better outcomes for those receiving intensive chemotherapy prior to transplant (p=.08) and with lower blast burden (mOS:80 vs 64 months, (p=.92)). Regarding specific therapies both standard 7+3/AML-like induction and CLAG regimens were equally effective in inducing CR and superior to single-agent HMA (p<0.05). 28 patients (34%) experienced relapse. Relapse rate for those with lower blast percentage at time of HSCT was 30% (n=18) compared to 40% (p=0.458). Patients who developed GvHD were less likely to relapse (85.7% vs 14.3%, p=0.000). There was an association between DNMT3A mutations and risk of relapse (p=0.04) with a trend also favoring NRAS (p=0.08). Patients with CMML who went to transplant earlier in the disease course had better outcomes. There was a trend toward better outcomes for those receiving intensive chemotherapy. CR was more likely with CLAG and 7+3 induction. Relapse was less likely for those who developed GvHD and more likely to be associated with DNMT3A mutations.