Phase II trial of venetoclax in combination with azacitidine as maintenance therapy for high-risk acute leukemia following allogeneic stem cell transplantation

Abstract Background: Allogeneic stem cell transplantation (alloSCT) remains the only potentially curative therapy for patients with acute myeloid leukemia (AML), although outcomes are suboptimal in high-risk populations. Methods: This is a single-center, open-label, Phase II clinical trial evaluating the safety and efficacy of venetoclax (VEN) plus azacitidine (AZA) as maintenance therapy following alloSCT. The trial includes four high-risk cohorts. This abstract reports results for Cohort 1 (high-risk AML). Inclusion criteria: 1) High-risk AML defined as undergoing SCT beyond first complete remission (CR1), having adverse-risk disease per ELN2017, or measurable residual disease (MRD) at transplant despite CR. 2) Absolute neutrophil count ≥ 1.0 × 10⁹/L and platelet count ≥ 30 × 10⁹/L without transfusion support. 3) No active graft-versus-host disease or infection 4) Enrollment between day +42 and +100 post-SCT and 5) MRD negativity within 30 days of enrollment. All donor and stem cell sources were included. Treatment: Up to 12 cycles of VEN 200 mg on Days 1–7 and AZA 32 mg/m² SC on Days 1–5 every 28 days. VEN was reduced for patients on CYP3A inhibitors (e.g., 40 mg with posaconazole). Further dose adjustments were made for hematological toxicity. Treatment delays of up to 70 days were permitted. Primary endpoints were relapse-free survival (RFS) and safety/tolerability. Results: 60 patients were enrolled, and 58 were evaluable. Median age was 60 years (range, 19–73). Per design, this cohort included high-risk features: 6 patients (10.3%) underwent a second SCT, 29 (50%) were in CR1, 15 had primary induction failure (25.9%), and 4 (6.9%) were in CR2 at the time of alloSCT. Among CR1 patients, 18 (62.1%) had adverse-risk disease per ELN 2017 19 (65.5%) had MRD detectable at SCT. Most patients (75.9%) received myeloablative conditioning. Median time to initiation of maintenance was 73 days post-SCT (range, 43-100). The median number of cycles was 6 (range, 1-12). At analysis, 7 patients remained on treatment; 51 had discontinued due to: disease progression (n=18), treatment completion (n=12), toxicity/performance status decline (n=13), patient request (n=6), or physician decision (n=2). RFS was 63.3% at 1 year (95% CI, 51.4–78.1%) and 54.1% at 2 years (95% CI, 41.6–70.2%) as shown in the figure. Overall survival (OS) time was 82.5% at 1 year and 56% at 2 years. Grade ≥3 adverse events occurred in 51 patients. The most common grade 3–4 toxicities included: anemia (n=17, 29.3%), neutropenia (n=23, 39.7%), and thrombocytopenia (n=19, 32.8%). Infections included neutropenic fever (n=1), pneumonia (n=4), and other infections (n=6). One patient experienced graft failure following the fourth cycle of maintenance therapy and subsequently relapsed 386 days after initiation of maintenance. No deaths unrelated to relapse occurred before day 90. Two patients experienced non-relapse mortality (NRM) at days 201 and 807 after start of maintenance. Conclusion: VEN/AZA maintenance post-SCT in high-risk AML appears to be feasible and demonstrates encouraging RFS and OS rates at 1- and 2-years. Despite frequent hematological toxicity, the regimen was generally well tolerated with low rates of NRM. Our findings support further investigation of VEN+AZA maintenance in this setting to improve alloSCT outcomes, including randomized trials and expansion to intermediate risk AML cohorts.