Identification of new predictive biomarkers and characterization of molecular mechanisms of drug-resistance in chronic myeloid leukemia

Treatment of BCR-ABL1⁺ human leukemia, especially for early phase chronic myeloid leukemia (CML) patients, has been greatly improved by ABL tyrosine kinase inhibitor (TKI) therapies. However, early relapses and acquired drug resistance remain problems. Thus, identification of new biomarkers and therapeutic targets are needed to predict patients’ responses for providing alternative treatment strategy and to overcome drug resistance by developing more effective therapies in CML. To identify new biomarkers that can predict a patient’s response to TKI therapies in CML, the expression of 47 microRNAs (miRNAs) that were differentially expressed between normal bone marrow and CML or in Imatinib (IM)-responders versus nonresponders was evaluated in CD34⁺ CML cells pre- and post-nilotinib (NL) therapy from a cohort of 58 patients enrolled in a clinical trial. Using Cox Proportional Hazard analysis and machine learning algorithms, miR-145 and miR-708 were identified as predictors for NL nonresponse in treatment-naïve cells, while miR-150 and miR-185 were predictors at 1-month and 3-month post-NL treatment. Interestingly, incorporation of in vitro colony formation data into either panel improved the predictive power at each time point. Thus, this new predictive model may be developed into a prognostic tool for use in the clinic. To investigate the molecular functions of the Ahi-1 oncogene and its SH3 domain in regulation of TKI resistance, a high-content antibody microarray was performed in BCR-ABL1⁺ cells expressing different constructs of Ahi-1. This analysis uncovered that the eIF4F complex, the key regulator of the mRNA-ribosome recruitment phase of translation initiation, was differentially expressed in wildtype Ahi-1 and IM-resistant cells. Interestingly, increased expression of several eIF4F complex members was demonstrated in CD34⁺ CML patient cells compared to normal bone marrow, particularly eIF4G1, the scaffolding protein of the complex. Strikingly, inhibition of eIF4G1 by shRNA or a selective inhibitor, SBI-756 impaired survival, increased IM sensitivity and reduced eIF4F complex activity significantly in IM-resistant cells. Additionally, inhibition of eIF4G1 resulted in a significant reduction of BCR-ABL1 protein expression in resistant cells, which may provide a novel strategy of targeting BCR-ABL1. Thus, understanding the mechanism of drug resistance mediated by eIF4G1 could lead to novel strategies to overcome these challenges.Medicine, Faculty ofExperimental Medicine, Division ofGraduat