Azacytidine and Venetoclax As a Salvage Therapy in Measurable Residual Disease (MRD) Relapse Post Intensive Chemotherapy or Allogenic Stem Cell Transplantation in Acute Myeloid Leukemia Patients : A Multicentric Study from French Auraml Group

Measurable residual disease (MRD) is recognized as a critical standardized endpoint in numerous hematological neoplasms, and more specifically in acute myeloid leukemia (AML). In intensively (IC) treated patients, poor NPM1 MRD response is associated with a high risk of relapse and may identify patients that benefit most from allogeneic stem cell transplantation (HSCT). In MRD negative patients after intensive treatments, MRD relapse is highly predictive surrogate marker of overt cytologic relapse, more specifically in NPM1_mut patients. Yet, there is no standard of care regarding isolated MRD relapse management. Due to a manageable toxicity profile compared to intensive salvage, azacytidine and venetoclax (AZA/VEN) combination regimen recently emerged as a promising strategy in relapse/refractory settings. We aimed in this retrospective study to evaluate the efficacy of AZA/VEN combination regimen in AML patients previously treated with IC as a first line of treatment and with molecular/flow cytometry (FCM) MRD relapse.

Patients from the VENAURA registry (N=585) with confirmed MRD relapse treated for a myeloid neoplasm that do not received AZA/VEN prior to MRD relapse were included in this study. VENAURA registry collates retrospectively AZA/VEN datas from 12 different French centers in Auvergne Rhône Alpes (AURA) region, between January 2019 and February 2024. Molecular relapse was defined according to the ELN 2022 guidelines. MRD negativity was defined as ≤ 10^-3 by FCM (on bone marrow sample), ≤ 10^-3 for WT1 by RT-qPCR and/or ≤ 10^-4 for NPM1 by RT-qPCR.

We identified 27 patients either with NPM1 (n=23), RUNX1-RUNX1T1 (n=1) and FCM (n=3) with MRD relapse. Median age was 59.7 (range:22.2-72.6). All patients were treated initially with IC. Among them, 11 received a maintenance therapy (6 midostaurine, 1 Gilteritinib, 4 oral azacytidine) and 2 were allotransplanted in first CR. MRD relapse occurred with a median of 10.1 months (range: 1.6-43.8) after first CR. Venetoclax was administred for a median of 21 days [IQR:7-28) at a median dose of 400mg/d [IQR:100-400]. Overall, 70.4% (19/27) reached MRD negativity after a median number 2 cycles (range: 1-6). In NPM1_mut patients, baseline FLT3-ITD mutation and early MRD relapse (<12 months following CR) were not associated with subsequent MRD response failure. Febrile neutropenia (FN) was observed in 22% of cases in cycle 1 and no infectious related death was observed. All FN observed during cycle 1 were associated with a tissular documentation (1 pneumoniae, 2 bartholinithis, 1 cellulitis, 1 upper respiratory tract infection) or bacterial documentation (Proteus mirabilis related urinary tractus infection). FN was observed in 4.7% of patients in the subsequent cycles. Regarding transfusion requirements, 19.6 and 15.4% required red blood cells and platelets transfusion during cycle 1. For patients that received more than 1 cycle (N=21), RBC and platelets transfusion requirements was observed in 15.2 and 9.5% of patients. Four patients received G-CSF after cycle 1 to manage neutropenia after reaching MRD negativity, none of them relapse during follow-up. Overall, 55% (15/27) move to HSCT of which 12/15 were MRD_neg at the time of transplant. After a median follow-up of 18.2 months (range: 1.2-37.3) post HSCT, no patient relapsed and only 1 died from acute gastro-intestinal GVHD. For patients that do not received HSCT (N=12), cumulative incidence of relapse was 44.4 and 66.3% at 12 and 24 months respectively. In non-HSCT patients, median RFS in MRD_neg and MRD_pos was 20.1 and 4.6 months respectively.

Our results confirmed that AZA/VEN combination regimen is a promising therapeutic strategy in case of MRD relapse in AML patients treated with prior IC . We observed low incidence of febrile neutropenia without any toxic related deaths. For patients bridged to HSCT, outcomes post-transplant seems to be favorable with low rate of NRM. For patients not transplanted after salvage AZA/VEN , MRD_neg responding patients might also experience long lasting responses. To further support these results, such strategy should be evaluated prospectively.