CPX-351 is effective in secondary Acute Myeloid Leukemia, regardless if defined according to 2016 or 2022 edition of who classification.

Abstract Background and aims: The definition of secondary Acute Myeloid Leukemia (s-AML) has significantly changed from the previous WHO classification in 2016 to the recent 2022 edition. In the previous WHO 2016 classification, s-AML was defined cytogenetically, if myelodysplasia (MDS) related abnormalities were demonstrated, clinically, in the presence of a clinical history of a previous MDS or in patient who already received a previous chemotherapy for un unrelated neoplasms and morphologically if dysplasia was noted. The better understanding of the genetic bases of AML allowed a precise molecular definition of s-AML. As a consequence, in the 2022 WHO classification, s-AML is hierarchically defined by 1. the presence of MDS related gene mutations (SRSF2, SF3B1, U2AF1, ZRSR2, ASXL1, EZH2, BCOR, and STAG2, collectively defined as secondary type mutations, STMs), 2. MDS related cytogenetic aberrations if no MDS-gene mutations are found and 3. clinically, if an antecedent history of MDS is present, recognizing three distinct s-AML categories. In a randomized phase III trial, CPX-351, compared to conventional 3+7, proved to be more effective as frontline treatment for s-AML patients and the drug is currently approved for s-AML patients, as defined by the WHO 2016 classification. More recently, in a sub analysis of the NCRI AML 19 British trial (Othman et al., Blood Advances 2023), CPX-351 was more effective than FLAG-Ida in patients affected by high risk AML who had STMs. However, few data are available on the efficacy of CPX-351 on molecularly defined s-AML as it is in the more recent WHO 2022 classification. The aim of this study was to evaluate the outcome of a cohort of elderly s-AML, defined according to previous WHO 2016, who received CPX-351 treatment, stratifying following both 2016 and 2022 WHO classification, in order to confirm the efficacy of CPX-351 in the molecularly-defined s-AML subgroup. Methods: Next-generation sequencing (NGS) was performed at diagnosis utilizing the Myeloid Solution panel by SOPHiA Genetics, encompassing 34 critical gene mutations. Samples were processed on an Illumina MiSeq platform, and analysis was performed using SOPHiA DDM® Software. Minimal Residual disease (MRD) analysis was conducted in all CR patients using multicolour flow cytometry (MFC), with a 0.1% threshold. Results: We retrospectively included a cohort of 85 patients (median age 69, range 37-77), all affected by s-AML as defined by previous WHO 2016 classification, who received treatment with CPX-351 in our Centre. According to WHO 2016 definition, diagnosis was therapy-related AML (t-AML) in 21 patients (25%), s-AML defined by cytogenetics in 55 (64%) and s-AML defined by morphology alone in 9 (11%). When WHO 2022 definition was applied, 64 patients (75%) had s-AML with MDS-related genetic aberration, 13 patients (15%) had s-AML with MDS-related cytogenetic aberrations whereas 3 patients had a previous history of MDS only (3%), whereas 5 (7%) patients would have no more be considered as affected by s-AML (all of them had a NPM1 mutation and were considered as affected by t-AML in the previous classification). A total of 66 patients (78%) achieved a complete remission (CR) after induction treatment, whereas 30 days-mortality was 3/85 (3.8%). MFC MRD assessment was negative in 48/66 CR patients (72%). Both response rate and MFC MRD negativity were not affected by different s-AML subcategory, either according to WHO 2016 or 2022. In the whole cohort, median follow-up was 42.1 months (CI 95% 31-62), median OS was 19 months (CI 95% 15.89-20.69) and 2-year OS was 40.2%. When WHO 2022 was applied, median OS did not differ between patients harbouring or not STMs (median OS 21 and 18 months in patients with or without STMs, respectively, p=n.s.). Overall, a total of 23 responding patients (35%) underwent allogeneic stem cell transplantation (HSCT) in first CR. Receiving HSCT was the only independent factor related to a longer overall survival, both in univariate and multivariate analysis (median OS not reached in HSCT patients, p0.05). Survival among HSCT was not influenced by the presence of STMs (p=n.s.). Conclusion Our data, together with the British trial, confirm that CPX-351 is effective in s-AML, also if defined according to WHO 2022. Therefore, CPX-351 is a reasonable option for patients with genetically defined s-AML, especially if HSCT in first CR is planned.