Background: The AML-Score has been validated to predict complete remission (CR) and early mortality (EM) in patients receiving traditional induction chemotherapies but not daunorubicin/cytarabine liposome (Vyxeos; Jazz Pharmaceuticals; Palo Alto, CA). We conducted a retrospective analysis to evaluate, among patients with newly diagnosed secondary acute myeloid leukemia (sAML) who received intensive induction with daunorubicin/cytarabine liposome, how accurately the AML-Score predictions associate with observed 1) CR after intensive induction and 2) EM within 60 days after induction initiation. Methods: We abstracted demographic and clinical data from consecutive patients receiving daunorubicin/cytarabine liposome at the Sidney Kimmel Cancer Center at Thomas Jefferson University Hospital between September 2017 and November 2020. We used descriptive statistics with 95% confidence intervals (CIs) and receiver operating curves to characterize the associations between AML-Score predictions and observed rates of CR and EM. Results: In total, 40 patients were included for analysis. 27 (67.5%) were male, 27 (67.5%) were white, 36 (90.0%) were not Hispanic or Latino, and 28 (70.0%) were aged 60-72 years. Twenty-seven patients (67.5%) had a CR, and 4 (10%) experienced early mortality. Observed rates of CR and EM generally increased with increasing predicted risk (Table). The area under the curve was 0.75 (95% CI 0.60-0.90) for CR and 0.82 (95% CI 0.68-0.96) for EM. Conclusions: The AML-Score tool trends in the correct direction in terms of predicting rates of CR and EM, and thus may contribute to oncologist prognostication and treatment planning for patients receiving daunorubicin/cytarabine liposome. However, its clinical utility is limited by the fact that it may underestimate the risk of CR and overestimate the risk of EM. Further study in a larger cohort is needed to update the AML-Score to give accurate point estimates of CR and EM risk in this population. Predicted CR and EM versus Observed CR and EM.

CI = Confidence Interval.