Relapse risk and survival benchmarks for non-transplanted acute myeloid leukemia patients in first complete remission

Abstract

Survival outcomes of patients with acute myeloid leukemia (AML) who achieve complete remission (CR) without allogeneic stem cell transplantation (allo-SCT) remain largely undefined, yet these benchmarks are essential for developing post-remission strategies. We retrospectively analyzed 362 adults with newly diagnosed de novo AML who achieved CR/Incomplete CR (CRi) (between 2016–2023) and did not undergo allo-SCT in first remission. Patients with APL and corebinding factor AML, which are rarely consolidated with allo-SCT, were excluded. The cohort was stratified by low-intensity (LIT, n=257) and intensive therapy (IT, n=105), and survival was evaluated across predefined subgroups. Median relapse-free survival (RFS) and overall survival (OS) were 11 and 19 months, respectively. Venetoclax was associated with significantly lower 24- month relapse rates in both LIT (68% to 45%) and IT (49% to 27%) (p<0.01 for both). ELN 2022 classification in IT and ELN 2024 in LIT showed clear prognostic separation across favorable, intermediate, and adverse groups, with median RFS of 58.0, 25.5, and 9.6 months for IT, and 13.7, 10.2, and 5.3 months for LIT, respectively. On multivariable analysis, ELN 2024 and measurable residual disease (MRD) positivity were independently associated with RFS in venetoclax-treated LIT patients, while ELN 2022 risk classification was most strongly associated with survival outcomes in IT. Notably, within the favorable-risk LIT subgroup, RUNX1 mutations were independently associated with inferior RFS. These findings establish benchmarks for postremission strategies in non-transplanted AML.