Modifications in risk classification of acute myeloid leukemias undergoing less intensive chemotherapy comparing the 2022 and 2024 stratifications of the european leukemianet and their impact on treatment decision, response rates and outcomes of a brazilian cancer center. Free

INTRODUCTION: Acute myeloid leukemia (AML) is a heterogeneous hematologic malignancy with a genetic variability that plays a role in determining clinical outcomes. The 2022 European LeukemiaNet (ELN) model uses molecular data to stratify risk in patients receiving high-intensive induction with or without allogeneic hematopoietic stem cell transplantation (allo-SCT), however, its predictive value is limited in those treated with less-intensive induction (Venetoclax plus hypometilating agents - HMA). Due to this gap, a 2024 model based mostly on VIALE-A study stratified patients into three groups: (1) TP53 mutations are classified as adverse risk; (2) those without TP53 mutations but with mutations in FLT3-ITD, NRAS, or KRAS, or other mutations not classified as favorable, are intermediate risk; and (3) the presence of one of the mutations – NPM1, IDH1, IDH2, DDX41 – or the absence of intermediate or adverse risk changes are considered favorable risk. Yet, this prognostic value remains suboptimal considering that mutations classified as adverse in ELN2022 may not carry the same poor prognostic impact in patients treated with less-intensive induction. Notably, HMA plus Venetoclax is now used in a broader and more heterogeneous population than in VIALE-A. Our aim is to evaluate this risk reclassification by mutation profiles and the impact in treatment response and outcomes in AML patients receiving less-intensive induction in a brazilian Cancer Center.

METHODS: A retrospective analysis was made with electronic medical records of patients diagnosed with AML (2020-2024) in our center that received less-intensive induction (Venetoclax and Azacytidine) and had molecular biology evaluated by next-generation sequencing (NGS) or Fluorescence In Situ Hybridization at diagnosis.

RESULTS: We included 29 patients with AML diagnosed between 2018 and 2024, with a median age of 70 years and a male predominance (62%). Sixteen patients (55%) were classified as eligible to proceed with allo-SCT. Seven patients (24%) were therapy-related AML and twelve patients (41%) were considered as a secondary progression from Myelodysplastic Syndrome. The most frequent mutations identified were TP53 (26%) and ASXL1 (24%), mutations in IDH2, SRSF2, and STAG2 were each detected in 21% of cases. According to the ELN2022, 25 patients (86%) were classified as adverse-risk, 3 (10%) as intermediate-risk, and only 1 patient (3%) as favorable-risk. However, based on the ELN2024, 17 patients (59%) were reclassified as intermediate-risk, 8 (26%) as adverse-risk, and 4 (14%) as favorable-risk. One case initially categorized as intermediate-risk by ELN2022 was reclassified as favorable-risk by the 2024 criteria, and 15 patients originally assigned to the adverse-risk group were reclassified as intermediate-risk. Risk classification remained unchanged in only 11 cases (38%). Of the total, 23 patients (79%) achieved a hematologic response, of which 14 presented negative measurable residual disease, the majority (64%) after the first cycle. Only 3 patients (10%) were refractory to treatment, one of them still alive at last follow-up. Fifteen patients (52%) underwent allo-SCT as consolidation therapy. At their last follow-up, 6 patients relapsed, 3 during treatment and 3 post allo-SCT, 9 maintained response and 15 patients (52%) died. Four cases were lost at follow-up. The median time from diagnosis to relapse was approximately 8.3 months, and 2.7 months between relapse and death, with a follow up of 11.1 months from diagnosis to death.

CONCLUSION: The growing use of molecular tests has significantly impacted AML risk stratification, reclassifying nearly two-thirds of our cases. These findings underscore the prognostic value of molecular data and the potential of ELN2024 to guide therapy, though adverse-risk patients remain a challenge. In a developing country hospital without routine NGS access, accurate molecular classification may be limited, and reaching a large number of patients can be challenging, we were still able to demonstrate the importance of adequate risk stratification for these patients.