Phenotypic Measures of Stemness for Discovery of Novel Therapeutic Targets and Personalized Medicine

Abstract

Healthy stem cells are rare, long-lived cells with unlimited self-renewal potential, enabling organ regeneration. Similarly, cancer stem cells (CSCs) contribute to cancer initiation, post-chemotherapy regeneration and development of drug resistance through their stemness properties of self-renewal highlighting them as a key therapeutic target in relapse prevention. Relapse and drug resistance represent the two biggest clinical hurdles to improving patient outcomes. Since CSCs are rare and governed by different cellular pathways than the bulk of the tumour they require unique ways of targeting. However, CSCs are difficult to isolate and study. CSCs, like all stem cells, are only defined by functional (aka phenotypic) assays that measure self-renewal and differentiation. My research aims to utilize phenotypic measures of stemness to address these clinical hurdles in therapy. First, I conduct a meta-analysis to highlight the importance of phenotypic assays in drug discovery compared to targeted base approaches. This meta-analysis reveals that drugs discovered using phenotypic assays were more likely to be approved for clinical use and less likely to fail due to lack of efficacy. Based on these results I proceed in two directions: using phenotypic measures of stemness for new drug discovery and using it for personalized medicine to improve already clinically approved therapies. First, using a phenotypic chemical genomics approach I uncover a novel vulnerability of CSCs through the targeting of SARNP and the transcription and export (TREX) pathway. This work uncovers a novel role for SARNP in R-loop homeostasis and importantly identified TREX as a targetable pathway for cancer stem cell drug discovery. Secondly, I develop a novel personalized medicine platform that integrates functional measurements of stemness, quantitative drug profiling and genetics which clinicians can use in real-time to improve therapy choices. Importantly, this platform can predict patient response to chemotherapy with 90% accuracy and identify clonal populations that are drug-resistant and relapse-driving leading to refractory disease. Overall, my work highlights the importance of phenotypic approaches to stem cell biology for personalized medicine and cancer drug discovery.