Impact of midostaurin in younger AML patients intensively treated with high-dose anthracycline

Abstract

Abstract Introduction: Midostaurin (MIDO) was approved by the FDA in 04/2017 for the treatment of FLT3 mutated AML patients in combination with intensive chemotherapy (ICT) with daunorubicin (DAUNO) administered at 60 mg/m² for 3 days based on the findings of the RATIFY trial (Stone, NEJM 2017). Moreover, the UK NCRI AML17 trial (Burnett, Blood 2016) demonstrated that higher DAUNO exposure at 90 mg/m² (without MIDO) provided a particular benefit for patients with FLT3 mutated AML. The aim of this post-hoc study was to assess the impact of MIDO in combination with higher-dose anthracyclines (DAUNO or idarubicin) in the BIG-1 trial (Hunault, NEJM Ev 2025). Methods: Between 01/2015 and 07/2021, the BIG-1 trial (NCT02416388) included patients (pts) aged 18-60 years with newly diagnosed AML treated with ICT (CBF-AML and APL excluded). DNA fragment analysis (FA) detected FLT3-ITD (AR≥0.05 for positivity) and FLT3-TKD mutations were detected depending on each center's usual procedures. The protocol planned single and first induction cycle containing anthracycline. Pts may receive either DAUNO (90 mg/m², d1-3) or idarubicin (9 mg/m², d1-5), combined with cytarabine 200 mg/m² (d1-7). After its approval, MIDO has been introduced in 07/2018 during the course of the trial and provided by Novartis. This offered the opportunity to assess the role of MIDO in this context using an internal control group. Of note, MIDO was omitted during the post-induction cycles in the few pts who entered nested randomized studies evaluating dexamethasone (N=46) or vosaroxin (N=13) in combination with HDAC or IDAC, respectively. Results: Overall, 382 (84.7%) pts had FLT3-ITD, 83 (18.4%) had a TKD mutation and 14 had both, leading to the inclusion of 451 pts in this analysis. 282 (62.5%) pts received ICT without MIDO (ICT group: internal control) and 169 (37.4%) received ICT with MIDO (ICT+MIDO group). Median age was 50.1y and 263 pts were female. ELN-2022 genetic risk was favorable, intermediate and adverse in 49 (10.9%), 319 (70.7%) and 77 (17.1%) pts. 302 (67%) pts carried also a NPM1 mutation without significant differences between the two groups. Following induction, the rate of CR/CRi was 77.3% vs 88.7% in ICT vs ICT+MIDO groups (p=0.002), respectively. Early death rate at d30 was 4.3% vs 1.2% (p=0.006). After adjustment on confounding factors including allo-HSCT in CR1 as a time-dependent variable in multivariate analysis, MIDO was significantly and independently associated with a decreased risk of relapse (sHR 0.63 [0.46-0.85]; P=0.003). At 2 and 5 years, cumulative incidence of relapse (CIR) was 43.6% vs 35% and 48.1% vs 40.9%, in the ICT and ICT+MIDO group respectively. The two other independent predictive factors for relapse were ELN-2022 genetic risk and allo-SCT in CR1 as protective factor. Anthracycline, gender, age and WBC did not significantly influence CIR. MIDO was also significantly and independently associated with a decreased risk of death or relapse (aHR 0.74 [0.55-0.98]; P=0.036), as well as ELN-2022 genetic risk and allo-SCT in CR1. At 2 and 5 years, RFS was 49.7% vs 55.4% and 42.5% vs 46.4%, in the ICT and ICT+MIDO group respectively. Again, anthracycline, gender, age and WBC did not significantly influence RFS. MIDO was significantly and independently associated with a decreased risk of death, relapse or failure (aHR: 0.65 [0.50-0.84]; P=0.001), as well as WBC, ELN-2022 genetic risk and allo-SCT in CR1. At 2 and 5 years, EFS was 43.3% vs 54.2% and 37.6% vs 44%, in ICT and ICT+MIDO groups respectively. Anthracycline, gender and age did not significantly influence EFS. Finally, MIDO was significantly and independently associated with a decreased risk of death (aHR: 0.70 [0.50-0.96]; P=0.02), as well as WBC, age and ELN-2022 genetic risk but neither allo-HSCT in CR1, nor sex, nor the type of anthracycline was associated with OS. At 2 and 5 years, OS was 62.8% vs 73.2% and 52.8% vs 62%, in ICT and ICT+MIDO group respectively. Conclusion: Subject to the limitations of this non-randomized study, adding MIDO to high-dose anthracycline-based chemotherapy improves CIR, RFS, EFS and OS independently of other factors, resulting in notable 5-year cure rates.