Molecular landscape and outcomes of Black and White adults diagnosed with acute myeloid leukemia.

Abstract

6519 Background: Acute myeloid leukemia (AML) is characterized by poor survival with Black patients (pts) experiencing worse outcomes compared to White pts. While patient-specific factors such as race have emerged as important predictors for survival in AML, the biologic mechanisms driving these factors are not well characterized. Here, we analyzed the molecular profiles of adults diagnosed with AML to identify potential differences in the frequency of genes commonly mutated in AML and their impact on survival. Methods: This was a retrospective cohort study of adults diagnosed with AML at the Cleveland Clinic between 2008 and 2022. Patients were identified as Black or White based on self-reported race/ethnicity. Molecular data were obtained from next-generation sequencing (NGS) performed at the time of diagnosis. Survival analysis was limited to pts receiving intensive induction chemotherapy and estimated using the Kaplan-Meier method and compared with the log-rank test. Results: 1,144 pts were included in this study: 10% (n = 115) Black and 90% (n = 1029) White. Median age at diagnosis (years) was lower in Black pts compared to White pts (62 vs 66) (p = 0.005). Cytogenetics analysis (MRC 2010) showed a higher proportion of Black pts with favorable-risk disease compared to White pts (15% vs. 8%; p = 0.04) while rates of poor-risk cytogenetics were similar (31% vs. 30%). Mutations in IDH1 , CBL , CEBPA , and NOTCH1 were more common in Black pts than in White pts (p < 0.05). Conversely, mutations in NPM1 were lower in Black pts (16%) vs. White pts (24%) (p = 0.22). We did not observe a difference in the type of induction chemotherapy received (p = 0.23). However, more White pts underwent hematopoietic cell transplant (HCT) compared to Black pts (32% vs. 21%; p = 0.02). The median overall survival (OS) for the whole cohort was 18 months (95% CI: 16-21 months) with a 5-year OS of 31% (95% CI: 27-34%); median follow up time of 63 months. Univariable analysis showed that worse overall survival was associated with older age, poor risk cytogenetics, therapy-related and secondary AML, and not undergoing HCT (p < 0.05), but median OS and 5-year OS rates were similar between Black and White pts (24 months, 28% vs. 17 months, 31%). After adjusting for prognostic variables, age, HCT status and cytogenetic risk remained significant (p < 0.05). Conclusions: Unlike previously reported studies, we did not observe a survival disparity in race in our study cohort. However, we observed lower rates of treatment with HCT in Black pts. We also identified a higher frequency of mutations in genes involved in myeloid transcriptional regulation, epigenetic modification, and cellular signaling. Adequate assessment of the prognostic relevance of these genes is limited by a small sample size and variability in NGS platforms over time. Future studies may explore pooled genetic data across several institutions and genetic ancestry testing over self-reported race/ethnicity.