Characteristics, Treatment Patterns And Predictors Of Outcomes In Treatment-Naive Acute Myeloid Leukemia Patients Judged Ineligible For Intensive Chemotherapy In Sweden

Context: There is limited recent population-based data on outcomes for novel agents and different treatment combinations in acute myeloid leukemia (AML) patients judged ineligible for intensive chemotherapy. Objective: To characterize patients, treatment patterns, and predictors of overall survival (OS) in treatment-naive AML patients judged ineligible for intensive chemotherapy. Design: A retrospective population-based study using data from the National Board of Health and Welfare, the Swedish AML Register, and medical records. Patients: Incident AML patients from 2012 to 2019 that were judged ineligible for intensive chemotherapy (n=1,335). Among a subgroup of patients with detailed information on treatment patterns and responses, patients with a hypomethylating agent (HMA; n=120) or low-dose cytarabine (LDAC; n=18) were selected for subgroup analyses. Patients were followed until death or censoring in 2020. Main Outcome Measures: Time-varying hazard ratios (HRs) for OS with 95% confidence interval (CI) for the calendar period for AML diagnosis, first-line treatment type, sex, age, AML etiology, and cytogenetic risk group. Responses were defined as complete remission (CR), including CR with incomplete hematological recovery, or primary induction failure. Results: In total, 122 (19.3%) and 278 (39.6%) patients received HMA during the earlier and later calendar periods (2012–15 and 2016–19), respectively. Among all patients, the median OS from AML diagnosis was 2.52 (95% CI 2.28–2.88) months. First-line treatment exposure to novel agents (HMA or venetoclax) compared to palliative or other low-intensive treatment was associated with lower risk of death at 1 years after AML diagnosis (HR 0.26 [95% CI 0.21–0.31], 0.39 [95% CI 0.29–0.53], and 0.62 [95% CI 0.40–0.95], respectively). Age at AML diagnosis, sex, AML etiology, and calendar period of diagnosis were not significantly associated with OS in the multivariable model, while cytogenetic risk group was borderline statistically significant. In the subgroup study, CR rates were 0% and 20% for patients exposed to LDAC- and HMA-based regimens, respectively. Conclusions: A greater proportion of patients received first-line treatment with HMA over time, which was associated with lower risk of death. However, most patients treated with HMA became refractory to HMA treatment, highlighting the need for optimized treatment strategies.