Final Analysis for the Primary End-Point of Gimema AML1718, a Safety Run-in and Phase 2 Open-Label Study of Venetoclax, Fludarabine, Idarubicin and Cytarabine (V-FLAI) in the Induction Therapy of Non Low-Risk Acute Myeloid Leukemia

Background: Acute myeloid leukemia (AML) poses significant challenges, especially in intermediate and high-risk cases due to modest rates of complete remission (CR) of 50-60% yelding a 2-year survival probability of about 30-40%. Preliminary results from the AML1718 part 1 trial have demonstrated remarkable improvements, with a composite complete remission (CCR) rate reaching an impressive 82%, and a 1-year survival probability of 72%. These findings represent a substantial advancement compared to historical data. Hereby, we present the comprehensive results of the AML1718 study, with a primary focus on the CCR as the primary endpoint.

Methods: The safety and efficacy of venetoclax (VEN) in combination with FLAI (fludarabine, cytarabine, and idarubicin as a first-line treatment for newly diagnosed adult patients with ELN 2017 intermediate- or high-risk AML) was investigated in the GIMEMA AML1718 phase 1/2 multicenter trial (NCT03455504). The study followed a modified two-stage Simon's design, including safety run-in cohorts (SRI-C1 and SRI-C2), an extension phase (Part 1 with P1-C1 and P1-C2 cohorts), and a confirmatory cohort (P2). VEN was administered at a daily dose of 400 mg in combination with FLAI (V-FLAI 400) in SRI-C1 and P1-C1, while in SRI-C2 and P1-C2, the VEN dose was escalated to 600 mg/day (V-FLAI 600). In P2 we used the lower effective dosage of V-FLAI 400, with centralized measurable residual disease (MRD) assessment with flow cytometry following ELN guidelines. After V-FLAI induction, consolidation therapy with cytarabine-based regimens and consideration for allogeneic stem cell transplant (HSCT) were conducted. Adjustments to the VEN dose were made for patients concurrently receiving posaconazole. VEN administration was discontinued until recovery for patients achieving remission by day 21 and during consolidation courses.

Results: Since February 2019, a total of 124 patients were treated in this study, 6 in SRI-C1, 6 in SRI-C2, 22 in P1-C1, 23 in P1-C2, and 67 in P2. Among them, 95 patients received V-FLAI 400 and 29 patients received V-FLAI 600. Posaconazole was administered to 115 patients (93%) as antifungal prophylaxis during the first course. The patient population had a median age of 55 years (ranging from 18 to 66), with 70 patients (56%) being male. According to ELN 2017 risk, 67 patients (54%) were classified as intermediate- and 57 patients (46%) as high-risk. Sixteen (13%) patients had a seconday AML. Genetic mutation analysis showed that 4/56 tested patients (7.1%) had IDH1, 11/55 (20%) IDH2, 16/108 (14.7%) FLT3 ITD, 3/112 ( 2.8%) NPM1 mutation; TP53 data are not yet available.

The primary endpoint, the CCR, was achieved in 93 out of 124 patients (75%) after the 1^st course (table 1). One patient initially in partial remission (PR) achieved CR after the second V-FLAI induction. In P2, 47/67 patients achieved CCR (70.0%, with 9 patients not yet tested). Central MRD analysis showed that that 31 out of 48 tested patients obtained MRD negativity at day 28 of course 1 (64.5%, treshold 0.1% by flow).

With a median follow-up of 10 months, 60 patients (49%) underwent hematopoietic stem cell transplant (HSCT) in CR, and 1 patient underwent HSCT in partial remission. The median overall survival (OS) was 22.4 months (95% confidence interval [C.I.] 13.4 months - not reached, Figure 1), with a 12-month OS probability of 64% (95% C.I. 54% - 76%). The median disease-free survival (DFS) was 20.7 months (95% C.I. 12.1 months - not reached), and the 12-month DFS probability was 64% (95% C.I. 50% - 76%).

The treatment was found to be safe, with only 4 deaths recorded during induction (3.2%), and a 60-day mortality rate of 6% including disease progression-related deaths. Infections were the most commonly reported grade 3+ adverse events. Overall, the safety profile of the treatment regimen was consistent with any intensified AML induction regimen. Among transplanted patients, no instances of graft failure or higher-than-expected incidence of graft-versus-host disease (GVHD) were observed. No significant differences were observed between the 400 mg and 600 mg arms.

Conclusions: V-FLAI demonstrated remarkable efficacy without any safety concerns. Impressive CCR rate and MRD-negativity qualify the combination for randomized comparisons . A virtual-randomized phase 3 trial is currently being prepared.