Abstract
BACKGROUND: ASXL1 mutation acute myeloid leukemia represents a clinically aggressive subtype with heterogeneous outcomes. Current evidence remains inconclusive regarding the prognostic relevance of the KRAS mutation partner in AML with ASXL1 mutation. The comprehensive mutational landscape and prognostic implications of co-occurring driver mutations remain poorly characterized. METHODS: A total of 2788 consecutive AML patient records were reviewed. A comprehensive clinicogenomic analysis was conducted on 451 AML patients with ASXL1 or KRAS mutations from the discovery cohort (n = 394) and the independent validation cohort (n = 57) to assess the correlation between molecular profiles and clinical outcomes. RESULTS: The KRAS mutation was observed in 22 (9.9%) AML cases with the ASXL1 mutation. Notably, survival analysis revealed that the ASXL1(mut)/KRAS(mut) subtype demonstrated trends toward inferior overall survival (OS) and relapse-free survival (RFS) relative to ASXL1 single subgroups. Stratified by mutational status, patients with ASXL1(mut)/KRAS(mut) exhibited significantly inferior 2-year OS rates (30.5% vs. 59.1% vs. 73.9%; p < 0.001) and short 2-year RFS (32.7% vs. 59.4% vs. 69.5%; p = 0.002) compared to ASXL1(mut)/KRAS(wt) or ASXL1(wt)/KRAS(mut) mutation counterparts. This association persisted in the BeatAML invalidation (OS: 0% vs. 43.1% vs. 69.3%; p = 0.026). This association persisted in the PSM analysis. This co-mutation confers an exceptionally poor prognosis comparable to that of TP53 mutations or complex karyotypes. HSCT showed no significant survival benefit after landmark analysis (OS; p = 0.292). CONCLUSIONS: These results demonstrate the independent prognostic value of ASXL1(mut)/KRAS(mut) co-mutation and define a novel ultra-adverse-risk subtype of acute myeloid leukemia.