Current Results of Intensive Therapy in Younger Adults with Acute Myeloid Leukemia (AML): The Large Randomized French Backbone Intergroup (BIG)-1 Study on Behalf of the Filo, ALFA, and SFGM-TC Study Groups

Introduction. This multicenter study randomly explored various standard intensive chemotherapy (ICT) strategies in adults with AML, namely higher-dose idarubicin (IDA) vs high-dose daunorubicin (DNR) and intermediate-dose (IDAC) vs high-dose (HDAC) cytarabine. The study was planned large enough to allow patient subgroup analysis.

Treatments. Eligible patients (pts) were randomized (R1) to receive an induction course with either DNR at 90 mg/m²/d for 3 days or IDA at 9 mg/m²/d for 5 days combined with cytarabine at 200 mg/m²/d CIV for 7 days. Since 09/2018, FLT3-mutated AML pts may also receive midostaurin 100 mg/d d8-21. Then, all pts alive without contra-indication for further ICT were randomized (R2) to receive a first consolidation (C1, if in CR/CRp/CRi) or salvage (if induction failure) with either HDAC (3g/m²/12h) or IDAC (1.5g/m²/12h) cytarabine at day 1, 3 and 5. Pts in CR/CRp/CRi after C1/salvage received two additional HDAC/IDAC cycles (C2, C3), according to R2 arm. In intermediate (int) or adverse (adv) risk AML pts as defined in the protocol, allogeneic HSCT from sibling or 10/10 (or even 9/10 if adv-risk) matched donors was offered after C1 (adv-risk) or C2 (int-risk). Conditioning regimens were myeloablative (MAC; Bu-Cy ± ATG) if age<45y or reduced-intensity (RIC; Bu-Flu-ATG) if age≥ 45y or HCTCI> 2 in int-risk pts, and sequential FLAMSA/Bu-Cy-ATG in adv-risk pts. Fav-risk pts were also eligible for HSCT if late CR/CRp/CRi or suboptimal NPM1 MRD response. Alternative stem cells sources and conditioning regimens might also be offered to adv-risk pts.

Patients & Methods. Pts aged 18-60y with de novo, post-MDS or therapy-related AML (CBF, APL, Ph+, and post-MPN excluded) were eligible if ECOG-PS ≤3, no active infection or neoplasia, no organ contra-indication, and a health insurance. Patient characteristics are shown in Table 1. Between 01/2015 and 05/2019, 1405 pts were enrolled and 1357 were evaluable. Among them, 1159 were randomized in R1 with the primary objective to increase 3-year OS from 45% in the DNR to 55% in the IDA arm. A total of 1221 pts were randomized in R2, including 198 pts after the end of the R1 inclusion period (they received DNR or IDA according to centers’ preferences). Primary R2 objective was to show OS non-inferiority after IDAC vs HDAC in a per protocol (PP) set, defined as pts who received the planned HDAC/IDAC dose ± 20% during C1/salvage. The non-inferiority boundary was an upper limit of the hazard ratio [HR] CI below 1.25. Analysis of R1 and R2 effects was adjusted on ELN-2022 risk groups, the other randomization/treatment arms, and time-dependent HSCT in first remission. Median follow-up was 5.4 years.

Results. Overall, 1148 (84.6%) pts achieved CR/CRp/CRi and 564 (41.6%) received HSCT in first remission (Table 1). At 6 years, OS was estimated at 53.4% (95% CI, 50.6-56.2). In the R1 population, IDA did not prolong OS (adjusted HR, 1.04 [0.88-1.23]; p=0.67). This lack of effect was observed in the three ELN-2022 risk groups. During induction, IDA was associated with longer neutropenia and antibiotic administration durations, more RBC/platelet transfusions, more mucositis and fungal infections. Remission and induction death rates were similar in both arms. In the R2-PP set that excluded 39 R2 pts, OS was significantly non inferior in the IDAC as compared to the HDAC arm (adjusted p-value, 0.004) and this was observed in the three ELN-2022 groups (Figure 1). During the first C1/salvage cycle, IDAC was associated with less CNS toxicity and less RBC/platelet transfusions. Only 2 pts died during this cycle. Neither responses, HSCT rates, relapse incidence, nor RFS were significantly impacted by R1 or R2. With respect to the OS endpoint, no interactions were found between induction anthracycline and R2. In pts in induction failure at R2, the salvage rate was 51.6% after HDAC vs 39.5% after IDAC (adjusted p-value, 0.081). As expected, time-dependent HSCT in first remission significantly prolonged RFS (adjusted HR, 0.41 [0.32-0.52]; p<0.0001) and OS from remission (adjusted HR, 0.65 [0.50-0.83]; p=0.001) in int/adv-risk AML pts. In these pts, no interaction was found between the effect of HSCT and the R2 arm (HDAC or IDAC).

Conclusions. In younger adults with non-CBF AML, benefit/risk analysis favors high-dose DNR vs IDA for remission induction and confirms IDAC as the standard for consolidation. The high overall transplant rate may have contributed to the good results of this study.