AFFILIATIONS
1 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
2 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
3 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
4 The First Affiliated Hospital of Xiamen University and Institute of Hematology, School of Medicine, Xiamen University, Xiamen, Fujian, China, Xiamen, Fujian, China
5 Department of Hematology, Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University, Fuzhou, China, Fuzhou, Fujian, China
6 Department of Hematology, The Second Affiliated Hospital, Fujian Medical University, Quanzhou, China., Quanzhou, Fujian, China
7 Department of Hematology, Quanzhou First Hospital, Fujian Medical University, Quanzhou, China, Quanzhou, Fujian, China
8 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
9 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
10 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
11 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
12 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
13 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
14 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
15 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
16 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
17 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
18 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
19 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
20 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
21 Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, China, Fuzhou, Fujian, China
ABSTRACT
Venetoclax plus azacitidine (VA) has transformed the treatment landscape for older or medically unfit patients with acute myeloid leukemia (AML). However, its comparative effectiveness versus intensive induction chemotherapy across the adult age spectrum remains unclear. We conducted a multicenter retrospective cohort study including 998 consecutive patients with newly diagnosed non-M3 AML treated with either VA (n=193) or idarubicin plus cytarabine (IA; n=805). The treatment efficacy, safety, and healthcare resource utilization were anlyzed.
After a median follow-up of 2.62 years, IA was associated with a higher incidence of infections (95.3% vs 87.6%; OR for VA, 0.293, 95% CI 0.125-0.684; P=0.005), whereas organ dysfunction occurred more frequently in the VA group (43.5% vs 27.5%; OR for VA, 1.748, 95% CI 1.099-2.781; P=0.018). Although IA achieved a higher composite complete remission (CR/CRi) rate than VA (78.0% vs 67.4%; OR for VA, 0.582, 95% CI 0.412-0.820; P=0.002), multivariable logistic regression identified adverse European LeukemiaNet (ELN) risk as the only independent predictor of lower CR/CRi rates (OR, 0.320, 95% CI 0.187-0.545; P<0.001).
Landmark survival analyses demonstrated superior overall survival (OS) and relapse-free survival (RFS) with IA beyond the first year, whereas 1-year outcomes were comparable between treatment groups. However, treatment regimen was not independently associated with OS or RFS after multivariable adjustment. Instead, patient age and ELN genetic risk were the dominant determinants of long-term outcomes. Age-stratified analyses identified 35 years as a clinically relevant threshold, above which survival progressively worsened. SHAP-based machine learning further identified age as the strongest predictor of survival and relapse. Interaction analyses demonstrated that the survival advantage associated with IA progressively diminished with increasing age, higher ELN risk, and poorer performance status, suggesting that long-term outcomes are driven primarily by disease biology rather than treatment intensity.
The healthcare resource utilization, including the cost and days of hospitalization. Among patients who did not undergo hematopoietic stem cell transplantation, IA was associated with significantly higher total hospitalization costs during the first year (P<0.001), whereas VA incurred substantially higher drug acquisition costs. The IA group had longer hospitalization days (P< 0.001)
In conclusion, age and ELN genetic risk, rather than induction regimen, were the principal determinants of long-term outcomes in this real-world AML cohort. The identification of a clinically relevant age threshold at 35 years, together with the discordance between treatment costs and long-term benefit, supports biology-driven, age-stratified therapeutic strategies and prospective pharmacoeconomic studies to optimize value-based AML care.