Drug sensitivity patterns across FAB subtypes and molecular mutations in AML.

Abstract

e18507 Background: The study analyzed drug response patterns in 186 AML patients utilizing selective Drug Sensitivity Scores (sDSS) derived from ex vivo multi-omics profiling data. Drug sensitivity profiles of 515 oncology agents across different AML subtypes and molecular mutations were studied to identify distinct therapeutic responses. Methods: Patients with identified FAB subtypes (M0, M1, M2, M4, M4 eos, M4/M5, and M5) and mutation statuses NPM1 and FLT3 variants (FLT3-ITD and FLT3-TKD) were investigated. Drug sensitivity was analyzed using mean sDSS scores. Primary endpoints were drug sensitivity patterns across FAB subtypes and mutations, and efficacy of different drug combinations. Results: FAB subtype analysis showed distinct drug response patterns based on mean selective Drug Sensitivity Scores (σ sDSS ). For M0 (n=7 patients), Navitoclax showed highest single-agent sensitivity (σ sDSS = 24), with Navitoclax + Venetoclax combination showing synergistic potential (σ sDSS = 20.67). M1 samples (n=22 patients) demonstrated higher sensitivity to Navitoclax (σ sDSS = 15.89), while combinations with mTOR inhibitors like Navitoclax + PF-04691502 (σ sDSS = 13.97) and Navitoclax + Vistusertib (σ sDSS = 13.72) showed promise. M2 FAB subtype (n=39 patients) responded moderately to Navitoclax (sDSS = 12.39), with Navitoclax + Cerdulatinib combination showing potential (σ sDSS = 11.35). M4 subtypes (n=2 patients) were most sensitive to BAY 87-2243 (σ sDSS = 15.98), with dual combinations like BAY 87-2243 + Cerdulatinib (σ sDSS = 14.21) and BAY 87-2243 + Pevonedistat (σ sDSS = 14.13) maintaining strong responses. However, M4’s small sample size demands cautious interpretation and requires further investigation with larger cohorts. In M4 eos (n=9 patients), Pimasertib demonstrated notable effectiveness (σ sDSS = 14.43), with dual-agent combination such as Pimasertib + SCH772984 (σ sDSS = 14.24) supporting RAS/ERK pathway inhibition. Despite rare M4/M5 subtypes (n=2 patients) showing limited Refametinib sensitivity (σsDSS = 8.75), their minimal sample size precludes definitive conclusions. M5 samples (n=22 patients) demonstrated moderate sensitivity to PF-04691502 (σ sDSS = 13.62). Likewise, mutation analysis revealed that NPM1-mutated samples showed increased sensitivity to Venetoclax (σ sDSS = 13.28) and PF-04691502 (σ sDSS = 12.63). FLT3-ITD mutations were associated with higher responsiveness to PI3K/mTOR inhibitors, with PF-04691502 achieving an σ sDSS of 11.84. FLT3-TKD mutated samples demonstrated better sensitivity to AZD8055 (σ sDSS = 15.94) and PF-04691502 (σ sDSS = 14.01). Conclusions: The study revealed distinct drug sensitivity patterns across AML FAB subtypes and mutations. The findings underscore the importance of integrating FAB classifications and mutation profiles in treatment decisions. Further research is required to validate these findings in larger cohorts and clinical settings.