Paper Details

ABSTRACT

Introduction:

For many older or unfit patients (pts) with AML, HMA/VEN improves survival but causes toxicities that compromise quality of life (QoL). Following remission, the standard of care is continuous therapy until disease progression or intolerable side effects. However, retrospective data suggest pts can achieve sustained treatment-free remission (TFR) depending on factors such as depth of response and duration of therapy (Chua et al. Blood Adv. 2022; Gariaz et al. Am J Hematol. 2024). We thus performed a prospective study to determine if planned interruption of frontline HMA/VEN for pre-selected patients safely mitigates toxicities without compromising remission duration and survival.

Methods:

This phase II clinical trial (NCT06511882) enrolls transplant-ineligible adults with newly diagnosed, TP53-wildtype AML on frontline HMA/VEN. After 12 months of therapy, pts who achieved CR/CRi with undetectable MRD (≤0.1% by flow cytometry) qualify for treatment discontinuation. Surveillance during discontinuation includes monthly clinical assessments and bone marrow biopsy every 3 months, with HMA/VEN reinitiated upon MRD relapse (including NPM1 MRD when applicable) or overt morphologic relapse. The primary objective is to determine CR/CRi rate at 18 months post-initial remission. Secondary objectives include evaluation of survivals, TFR duration, re-treatment efficacy, and QoL. A Simon’s two-stage design determines sample size, with interim safety analyses after every 7 pts to ensure sufficient primary endpoint attainment before trial continuation.

Results:

By the July 2026 data cutoff, 11 pts evaluable for TFR (9M/2F) with a median age at diagnosis of 81 years (range: 69-84) were enrolled between 10/2024 and 6/2026. The cohort included 6 (54.5%) pts with de novo AML and 5 (45.5%) with secondary/therapy-related AML. While most pts (81.8%) had intermediate risk cytogenetics (as defined by ELN 2022), integrated risk stratification shifted significantly between classifications, with 7 (63.6%) adverse risk per ELN 2022 and 8 (72.7%) favorable risk per ELN 2024. Molecular profiling identified an average of 3 mutations per pt, with over half the cohort, 6 (54.5%), harboring myelodysplasia-related mutations (sMut). The most common mutations were BCOR, DNMT3A, IDH2, NPM1, RUNX1, and SRSF2, each detected in 27.3% of the pts. There were no pts with FLT3-ITD mutated disease.

With a median follow-up of 21.9 months from diagnosis, median OS was not reached. Post-discontinuation, 5 pts (45.5%) experienced MRD relapse at a median of 4.6 months (range: 2.5-11.3). One pt progressed to overt morphologic relapse 1 month after NPM1 MRD relapse, prior to therapy reinitiation. This pt had ELN 2022/2024 favorable risk disease with DNMT3A, TET2, NPM1, CEBPA (non-bZIP), and KMT2D mutations. Another pt with NPM1 MRD relapse, also favorable risk (IDH2/NPM1), restarted therapy and maintained morphologic remission with persistent NPM1 MRD after 2 months. However, the pt died following a stroke at age 85 prior to subsequent response assessment, with no evidence of overt relapse at the time of death. The remaining 3 pts with MRD relapse resumed therapy and await post-restart bone marrow evaluation. No hematologic or non-hematologic AEs were attributed to treatment discontinuation. 

Six pts (54.5%) maintained CR/CRi with undetectable MRD at a median post-discontinuation follow-up of 7.5 months (range: 0.7-16.6). Notably, 2 pts maintaining this response off treatment for ≥12 months both presented with ELN 2022 adverse risk disease and multiple sMut at diagnosis (1: ASXL1, IDH1, RUNX1, SETBP1, SRSF2 and 2: BCOR, DNMT3A, NF1, PTPN11, RUNX1, U2AF1). Before discontinuation, the first pt had persistent ASXL1 and the second pt achieved NGS clearance. All 6 pts were alive at last contact, and QoL metrics are being evaluated.

Conclusions:

Discontinuing frontline HMA/VEN in transplant-ineligible pts with AML who achieve undetectable MRD is clinically feasible, yielding sustained, MRD negative TFR in over half the cohort. To date, only 1 overt morphologic relapse has occurred. Notably, TFR ≥ 12 months was observed even in pts with adverse risk disease. However, the risk of early MRD relapse necessitates rigorous surveillance to ensure timely re-treatment. Continued trial enrollment and longer follow-up are required to evaluate re-treatment efficacy, further assess QoL outcomes, and refine criteria for safe discontinuation.