Recurrent Drug Sensitivity Patterns in Myelodysplastic Syndrome Patients Are Recapitulated by Ex Vivo Drug Response Profiling

Myelodysplastic syndromes (MDS) are a collection of clonal diseases of dysfunctional hematopoietic stem cells, characterized by ineffective hematopoiesis, cytopenias, and dysplasia. Limited conventional treatment options exist for these patients, with hypomethylating agents remaining the standard of care for higher-risk MDS patients. Drug sensitivity and resistance testing on myelodysplastic syndromes (MDS) samples should provide important functional information to guide actionable target and biomarker discovery. We provide proof-of-concept data by profiling the effects of 50 common oncology drugs on 14 myelodysplastic (MDS) samples from both treatment-naïve and refractory cases. Ex vivo high-throughput functional screening was performed on lysed peripheral blood and/or bone marrow aspirate patient samples. This set of samples was analyzed 72 hours post incubation and blasts were quantified by flow cytometry. Unsupervised hierarchical clustering was performed to determine similar responses between patient samples, and produced two distinct groups with differing sensitivity to tested drugs. We defined these groups as “sensitive” and “resistant” clusters based on the prevailing drug sensitivity patterns these groups displayed, the latter being strongly enriched for HMA refractory samples. Ex vivo testing recapitulated known patient drug sensitivity patterns, with 12/14 samples from known HMA refractory vs. sensitive patients showing similar ex vivo phenotypes. Interestingly, sensitive and resistant clusters strongly correlated with known features associated with therapy resistance and poor prognosis in MDS patients, including number of somatic driver mutations (average 2.38 vs. 4.5, P <0.05), absence vs. presence of RUNX1 or ASXL1 mutations (P 0.015 and 0.0256, respectively), and HMA sensitive vs. refractory status (P <0.05). Interestingly, percent bone marrow blasts, cytogenetic risk groups and other somatic mutations had no effect on sensitivity patterns. Furthermore, while the resistant cluster showed increased resistance to most drugs, some drugs seemed to have increased activity in this cluster, including calcitriol and poly ADP ribose polymerase (PARP) inhibitor rucaparib. This unique platform, applied to predict ex vivo therapeutic response of MDS patient samples to various classes of drugs, recapitulates known clinical and molecular predictors of therapeutic response, and possible new therapeutic targets. These data suggest the possible utility of using this methodology to aid decision making for therapeutic selection in the management of MDS patients.