Azacitidine and Venetoclax As a Salvage Therapy for Relapse after Allogeneic Transplantation for Acute Myeloid Leukemia : A Multicentric Study from French Auraml Group

Even if therapeutic arsenal has improved the prognosis of acute myeloid leukemia (AML), relapse remains a challenge especially in post allogenenic stem cell transplantation (ASCT) settings. Therapeutic options in this situation are usually limited and prognostic remains poor. While azacitidine and venetoclax (AZA/VEN) combination regimen is now a standard of care in frontline settings for unfit AML patients, it is nowadays used as an off-label therapeutic approach in relapse/refractory settings after ASCT. In this multicentric study, we aimed to evaluate the efficacy and safety of AZA/VEN regimen for relapse/refractory (R/R) AML patients after ASCT in real life setting.

Patients from the VENAURA registry (N=585) with following criteria were included in this study: patients transplanted for AML or other high-risk myeloid neoplasm (MDS/AML, CMML-2) who cytologically relapse after ASCT. Data were retrospectively collected in the VENAURA registry from 12 different French centers (Saint-Etienne, Clermont-Ferrand, Lyon (Hopital Lyon Sud, Centre Léon Bérard), Vichy, Annecy, Chambery, Valence, Bourgoin-Jallieu, Grenoble, Roanne) in Auvergne Rhône Alpes (AURA) region, between January 2019 and February 2024. Composite complete remission CRc was defined as in VIALE-A trial. Measurable residual disease (MRD) negativity was defined as ≤ 10^-3 by flow cytometry (in the bone marrow), ≤ 10^-3 for WT1 by RT-qPCR and/or ≤ 10^-4 for NPM1 by RT-qPCR.

Overall, 52 patients (AML=49, myeloid sarcoma=1, MDS/AML=1, CMML-2=1) received AZA/VEN combination regimen for relapse after ASCT. Median time between ASCT and relapse was 12.8 months (1-182). Eleven patients received azacitidine as monotherapy prior venetoclax introduction for mixed chimerism/molecular relapse (n=7) or cytologic relapse (n=4). Patients received a median number of 3 AZA/VEN cycles (1-23). Median time between relapse and AZA/VEN onset was 1.5 months (0.2-24) with a median duration of cycle 1 of 21 days (7-28). During C1, febrile neutropenia and tumor lysis syndrome was observed in 40.7% and 7.8% of cases respectively. During follow-up, 27 patients presented with acute GVHD including 12 grade 1, 11 grade 2 and 4 grade 3. No grade 4 GVHD occurred. Fourteen patients developed chronic GVH, 5 of them in extensive form. For patients evaluated at the end of cycle 1 (N=38), overall response rate (CRc) was 47.3% (18/38) with 12 CR and 6 CRi respectively. In responding patients evaluable for MRD at the end of cycle 1 (N=13), 30.7% were MRD negative. For non-responding patients at the end of C1 (N=21), 13 received subsequent cycles of which 4/13 (30.7%) reached CR/CRi/MLFS. Overall, 18.3%, 26.5% and 12.2% of evaluable patients reached CR_MRD-, CR_MRD+, and CR with unknown MRD status. Median duration of CRc was 12.2 and 3.9 months in CR_MRD- and CR_MRD+ patients respectively (p=0.02). With a median follow-up of 8.7 months since AZA/VEN onset, median overall survival (OS) was 17.1 months. Patients with CR_MRD- have a significant superior outcome compared to those in CR_MRD+ (p=0.023). During follow-up, 40.4% of patients received donor lymphocyte infusions (DLIs) either prior AZA/VEN onset (N=9) or after (N=8) which did not seem to influence ASCT outcome. Five patients received a second ASCT in CRc following AZA/VEN salvage of which 2 ultimately relapsed after the second transplantation.

To our knowledge, this is one of the largest cohort reporting AZA/VEN combination regimen efficacy and safety in post-ASCT AML relapse settings. In this real-life settings study, those results suggest that AZA/VEN combination regimen yielded promising results with a reasonable toxicity profile in post-ASCT AML relapse settings. Reaching CR_MRD- was associated with a better outcome and it might represent a good option to optimize graft versus leukemia through immunomodulatory approaches before considering DLI or a second allogeneic transplantation.