Efficacy and safety of venetoclax and azacitidine for acute myeloid leukemia in China: a real-world single-center study

Abstract

Background: Venetoclax (VEN) and azacitidine (AZA) are used to treat patients with newly diagnosed acute myeloid leukaemia (AML) who are unfit for intensive chemotherapy and those with relapsed or refractory AML. Understanding the real-world usage patterns and outcomes after VEN-AZA therapy failure is crucial, yet poorly studied.

Methods: This single-centre retrospective cohort study included 50 patients with AML (29 newly diagnosed and 21 relapsed or refractory cases) who were treated between January 2020 and November 2023. The primary endpoint was overall survival (OS), and secondary endpoints included composite complete remission (CR), partial remission (PR), overall response rate (ORR), event-free survival (EFS), minimal residual disease (MRD), adverse events (AEs), and post-VEN-AZA failure.

Results: Among the newly diagnosed patients (median age, 74 years; follow-up 10.1 months), the median EFS was 9.87 months (95% CI, 6.54-13.2 months) and OS was 11.93 months (95% CI, 7.6-16.29 months). The ORR was 85.7%, CR/CR with incomplete haematologic recovery (CRi) was achieved in 67.9% of patients, and MRD negativity was observed in 26.3% of the cohort. Post-treatment failure included VEN-AZA combined with gilteritinib, chidamide, or selinexor, which resulted in PR or CRi. The median OS after post-failure was 1.6 months. Among relapsed or refractory cases (median age 65 years; follow-up 8.53 months), median EFS was 5.2 months (95% CI, 1.8-8.6 months), and OS was 9.1 months (95% CI, 3.01-15.19 months). The ORR was 52.4%, CR/CRi was achieved in 42.9% of patients, and MRD negativity was observed in 11.11% of the cohort. Post-failure treatments include induction chemotherapy, VEN-AZA combined with enasidenib or gilteritinib, and participation in clinical trials, which yielded varying responses. The median OS after failure was 0.67 months, and the most common AEs were haematological and infectious complications.

Conclusions: VEN-AZA demonstrated high efficacy and manageable toxicity in patients with AML. Following VEN-AZA failure, the combination of VEN-AZA with targeted therapies has shown better efficacy than other VEN-AZA alone, whereas induction chemotherapy or clinical trials were preferred after second-line failure. Larger multicentre studies are warranted to validate these findings.

Keywords: Acute myeloid leukemia; Hypomethylating agents; Refractory; Relapse; Venetoclax.

Fig. 1

Waterfall plot of gene mutation or indel in AML patients. The waterfall plot illustrating gene mutations or indel in 26 new diagnosed AML patients (A) and in 19 relapse/refractory AML patients (B). The horizontal axis represents individual patients, while the vertical axis displays the mutated genes. Different types of squares signify various mutation types, with gray squares indicating that the genes are wild-type. The bar graph at the top shows the total number of mutations per gene across all patients. The numbers on the left represent the proportion of mutations relative to the total patient sample, and the bar graph illustrates the proportion of different mutation types within each gene. The plot identifies WT1, IDH1/IDH2 and MLL as the top three genes with the highest mutation/indel frequencies in new diagnosed AML patients (A) and DNMT3A, FLT3 and RUNX1 as predominant in relapse/refractory AML patients (B)

Fig. 2

Event-free survival and overall survival of patients with new diagnosed AML patients treated with venetoclax-azacitidine. (A) Event-free survival curves for patients with new diagnosed AML treated with VEN -AZA. (B) Overall survival curves for patients with new diagnosed AML treated with VEN-AZA. (C) Event‐free survival curves for patients with relapse/refractory AML treated with VEN-AZA. (D) Overall survival curves for patients with relapse/refractory AML treated with VEN-AZA. Abbreviations: EFS: event‐free survival; OS, overall survival; AZA, azacitidine; VEN, venetoclax

Fig. 3

Univariate Cox regression analysis of event-free survival and overall survival. The hazard ratio for treatment failure, relapse from CR/CRi or death from any cause was estimated with the univariate Cox regression analysis. Data included are subject to a cutoff date of March 2, 2024. This plot displays the relative risks (hazard ratios) and 95% confidence intervals for various variables in relation to the study outcomes. Each point represents the estimated hazard ratio, with the lines indicating the 95% confidence intervals. The dashed vertical line represents a hazard ratio of 1.0. Abbreviations: WBC: white blood cell; Hb: hemoglobin; PLT: platelet; LDH: lactate dehydrogenase; ELN: European Leukemia Net; MDS: myelodysplastic syndromes; HMA: hypomethylating agent; ECOG: Eastern Cooperative Oncology Group performance-status scores; CR: complete remission; CRi: Complete Remission with Incomplete Hematologic Recovery; MRD: Measurable Residual Disease; (A) Univariate Cox regression analysis of event-free survival for patients with new diagnosed AML treated with VEN -AZA. (B) Univariate Cox regression analysis of overall survival curves for patients with new diagnosed AML treated with VEN-AZA. (C) Univariate Cox regression analysis of event‐free survival curves for patients with relapse/refractory AML treated with VEN-AZA. (D) Univariate Cox regression analysis of overall survival curves for patients with relapse/refractory AML treated with VEN-AZA