Oral azacitidine maintenance improved survival for t-AML or AML-MR patients in first remission after CPX-351 and not eligible to HSCT

Abstract Background Oral AZA (Onureg, CC-486) is a maintenance therapy administered at the dose of 300 mg daily, 14 days of a 28-day cycle 1 and approved by the FDA on September 1, 2020 2. Based on the results of the QUAZAR AML-001 trial 3,4, oral AZA is indicated to treat adult patients with AML in complete remission (CR) with or without incomplete blood count recovery (CRi) following intensive induction chemotherapy who are not candidates for hematopoietic stem cell transplant (HSCT). Twenty percent of the oral AZA arm had AML with myelodysplasia-related changes (AML-MRC) and 11% had secondary AML. In NPM1 mutated patients, oral AZA maintenance showed a better median overall survival (OS) 5. Moreover, this maintenance therapy seemed to provide a significant prolongation of OS (mOS 19.9 months) among older patients with AML-MRC according to a post-hoc analysis 6. However, all patients were treated with standard “3+7” induction chemotherapy before entry into the QUAZAR study, leaving uncertainty about the efficacy of oral AZA maintenance in patients with AML-MRC or t-AML who are now routinely treated with CPX-351 as the standard of care. To note, median OS for patients not transplanted after CPX-351 was only 14 months 7. Current guidelines show divergent recommendations for AZA maintenance, with limited evidence specific to patients treated with CPX-351 8,9. Here we report the first patients treated with oral AZA maintenance for t-AML or AML-MR after CPX-351. Methods We conducted a real-world, retrospective, international, multi-center study (ORAZ-351, NCT06349239) in 11 centers in France, US, Germany and Australia. We included AML patients receiving oral AZA maintenance after CPX-351 between 2020 and 2024. Patients were treated according to the US FDA/EMA-approved dosing and schedule of oral AZA. Patient characteristics, treatment patterns, relapse, survival, and occurrence of adverse events (AEs) during oral AZA treatment were analyzed descriptively. Kaplan-Meier methods were used to estimate OS. Results A total of 34 patients with a diagnosis of AML-MR (82.4%) or t-AML (14.7%) were included. Most patients were male (58.8%) with an ECOG of 0 (26.5%) or 1 (61.8%) and a median age of 69 years (range, 39-76) at oral AZA initiation. Genetic risk status using ELN 2022 classification was favorable (23.5%), intermediate (23.5%) and adverse risk (52.9%). Using Lindsley's classification, 35.3% were de novo/pan-AML, 58.8% were secondary and 5.9% were TP53-mutated. The most common gene mutations at AML diagnosis were DNMT3A (23.5%), NPM1 (23.5%), SF3B1 (20.6%), SRSF2 (20.6%), NRAS (17.6%), ASXL1 (17.6%), TET2 (14.7%), U2AF1 (11.8%) and BCOR (11.8%). After first induction, the CR1 rate was 67.6%. 61.8% and 35.3% received one or two cycles of CPX-351 consolidation, respectively. The median number of cycles of oral AZA was 5 (range, 1.0-27.0) with a median duration of 3.6 months (range, 0.1-45.5). The median time from CPX-351 to oral AZA introduction was 4.68 monthsrange, 0.39-17.94). Only 32.4% remained on the drug for ≥ 12 months and 38.2% (n=13) of patients required a dose reduction due to adverse events (intolerance (30.8%) or cytopenia (69.2%)). The two main causes of discontinuation were relapse (44.1%, n=15) and cytopenia (32.3%, n=11). The median OS from diagnosis was 38.5 months (range, 4.2-54.3), and from oral AZA introduction was 28.2 months (range, 1.5-46.8). For patients with NPM1 mutation (n=8), the median OS from oral AZA introduction was not reached (vs 20 months for wild-type patients, P= 0.08). Baseline MRD status at oral AZA introduction was available in 17 patients (50%): 9 patients were MRD positive and 8 were MRD negative. The median OS from oral AZA introduction was 12.2 months and not reached (P=0.027) in MRD positive and MRD negative patients respectively. Conclusion Oral AZA maintenance may improve OS in t-AML and AML-MR after CPX-351 compared to the literature in our cohort. Safety outcomes were consistent with the known safety of oral AZA. MRD at oral AZA introduction could be a strong prognostic indicator of OS. The benefit of this strategy needs to be validated prospectively. More Italian patients will be included at the time of ASH meeting. Bibliography 1 Garcia-Manero, et al. JCO 2011 2 FDA approval 3 Wei et al. ASH 2019 4 Wei et al. NEJM 2020 5 Döhner et al. ASH 2022 6 Voso et al. JCO 2024 7 Lin et al. Blood Adv 2021 8 The NCCN Guidelines (version 3.2024) 9 Heuser et al. Ann Oncol. 2020