Cytogenetic and molecular risk predict survival in AML independent of age and treatment intensity

Abstract

Abstract Background: Cytogenetic and molecular features are critical to prognostication and treatment planning in acute myeloid leukemia (AML). However, the interaction of these factors with age and induction regimen remains less clearly defined. This study evaluates the impact of cytogenetic and molecular risk on survival outcomes across age and treatment strata. Methods: We conducted a retrospective review of 152 adult patients (age ≥18 years) diagnosed with AML between January 2010 and December 2024 at our institution. Patients were excluded if they were under 18 years or did not undergo treatment. Cytogenetic risk was classified as favorable, intermediate, or poor. Mutation status (FLT3, IDH, or none) was recorded where available. Patients were stratified by age (≤60 vs >60) and induction regimen (Anthra+Cyt, Ven+Aza, or others). Kaplan-Meier survival curves with log-rank testing were used to compare OS. Multivariable Cox proportional hazards modeling assessed the impact of cytogenetic risk, mutation status, age, and treatment intensity. Results: Cytogenetic Risk and Age-Based Survival: Age ≤60 with poor-risk: Median OS 9.7 months Age >60 with poor-risk: 15.4 months; intermediate-risk: 22.3 months Favorable-risk (all ages): OS not reached Survival differences by cytogenetic risk were significant in younger patients (p=0.008) Mutation-Based Survival: FLT3-mutated: 44.1 months; IDH-mutated: Not reached; No mutation: 51.9 months (p=0.323) Multivariable Cox Model Findings: Poor-risk cytogenetics: HR = 4.48, p = 0.0098 Intermediate-risk cytogenetics: HR = 2.40, p = 0.1181 Age >60: HR = 1.91, p = 0.0734 Induction regimen: Not independently predictive Limitations: The sample size for FLT3 and IDH mutation subgroups was small (n = 7–11), reducing power to detect survival differences. Cytogenetic subgroups were unevenly distributed across age and treatment arms, potentially introducing bias. Conclusions: Cytogenetic risk remains the most powerful predictor of survival in AML, regardless of patient age or induction regimen. While molecular markers such as FLT3 and IDH did not show significant associations in this cohort, limited sample sizes may obscure their true prognostic value. These findings reinforce the central role of cytogenetic evaluation in guiding treatment decisions and highlight the need for larger, mutation-enriched datasets to clarify the impact of molecular risk.