Project summary Targeted therapies in melanoma have shown enormous promise in the sense that they can show dramatic reductions in tumor burden, with melanoma being a particularly stark example. However, this promise has failed to be fully realized because of the emergence of resistant tumor cells, which repopulate the tumor and are subsequently difficult or impossible to treat effectively. Typically, scientists have thought of therapy resistance as having genetic origins, with rare mutant tumor cells surviving therapy because of a mutation that causes resistance. Recent work from our labs using advanced single cell analysis, however, suggest that, at the point of attack, there may be other, complementary, non-genetic mechanisms that could also govern exactly why some rare cells are able to evade the effects of the therapy. Subsequently, the targeted therapy itself can reprogram these rare cells into a stably resistant population. This more nuanced “plasticity and reprogramming” view of resistance at the single cell level has opened the possibility of a far richer set of targets that can be exploited for forestalling therapy resistance; however, the current set of tools and models, both experimental and computational, for identifying these targets are underdeveloped and the origin of these biological processes remain mysterious. Here, we propose to develop and apply new concepts and methods in experimental and computational single cell biology to tackle the problem of non-genetic therapy resistance, translating our basic science results towards the clinic through the use of sophisticated in vivo models of melanoma. In Aim 1, we will identify and validate the pathways that govern cellular plasticity in melanoma. We will develop new tools to identify gene networks associated with plasticity, and then deploy new tools, both computational and experimental, to identify vulnerabilities in those networks, ultimately testing whether those vulnerabilities recapitulate in more realistic in vivo settings. In Aim 2, we will develop a tool for revealing the pathways associated with reprogramming. Then, combining this information, we will develop a computational model to predict optimal timed dosing strategies that incorporate these non-genetic rare-cell vulnerabilities into a comprehensive framework. We will then test this framework on patient-derived xenograft models of melanoma to demonstrate the potential clinical impact of our findings on melanoma treatment.

Project narrative The development of new targeted therapies for melanoma have raise the promise of effective treatment, but the emergence of resistance afterwards has remained a major challenge. We have discovered a complementary single-cell view to therapy resistance, and our work aims to exploit this new model of resistance to block the onset of resistance.