Ex Vivo Drug Sensitivity Correlates with Clinical Response and Supports Personalized Therapy in Pediatric AML

SIMPLE SUMMARY: Children with acute myeloid leukemia (AML) experience unacceptably poor survival outcomes and are at high risk of relapse. Current treatment options are limited, and standard strategies rely on intensive chemotherapy to achieve remission, frequently resulting in treatment-related morbidities and significant late adverse effects. The use of an ex vivo drug sensitivity platform has potential clinical utility to aid individualized patient risk assignment, and it could allow for personalized treatment regimens, including identifying novel therapies for patients who are identified to be at a very high risk of treatment failure. In this study, we show that ex vivo drug sensitivity correlates with clinical response measures in a cohort of children with AML who received conventional chemotherapy. We also demonstrate preferential sensitivity ex vivo between conventional chemotherapy and the combination of bortezomib and panobinostat in a subset of patient samples. Our results support the value of an ex vivo drug sensitivity platform to identify individualized precision therapy for children with AML. ABSTRACT: Acute myeloid leukemia (AML) is a heterogeneous disease that accounts for ~20% of all childhood leukemias, and more than 40% of children with AML relapse within three years of diagnosis. Although recent efforts have focused on developing a precise medicine-based approach towards treating AML in adults, there remains a critical gap in therapies designed specifically for children. Here, we present ex vivo drug sensitivity profiles for children with de novo AML using an automated flow cytometry platform. Fresh diagnostic blood or bone marrow aspirate samples were screened for sensitivity in response to 78 dose conditions by measuring the reduction in leukemic blasts relative to the control. In pediatric patients treated with conventional chemotherapy, comprising cytarabine, daunorubicin and etoposide (ADE), ex vivo drug sensitivity results correlated with minimal residual disease (r = 0.63) and one year relapse-free survival (r = 0.70; AUROC = 0.94). In the de novo ADE analysis cohort of 13 patients, AML cells showed greater sensitivity to bortezomib/panobinostat compared with ADE, and comparable sensitivity between venetoclax/azacitidine and ADE ex vivo. Two patients showed a differential response between ADE and bortezomib/panobinostat, thus supporting the incorporation of ex vivo drug sensitivity testing in clinical trials to further evaluate the predictive utility of this platform in children with AML.