Tracing and controlling origins, dynamics, and consequences of rare-cell plasticity in cancer

Individual cells within a genetically homogenous population constantly undergo fluctuations in their molecular state which may enable them to adopt new fates in response to signals. Often referred to as “cellular plasticity”, these molecular differences in cell states can translate into dramatically distinct phenotypic fate outcomes in response to external cues. A salient example of variable responses to stressors is cancer cells exposed to targeted and cytotoxic therapies. Not all tumor cells die in response to these therapies, leaving behind a rare, drug resistant subset that promotes relapse in patients. While much effort has gone into probing the genetic determinants of resistance (aka classical Darwinian selection), I and others have recently shown that non-genetic variability or plasticity plays a larger role than hitherto appreciated in driving resistance. Furthermore, my recent work (in revision at Nature) demonstrated the presence of several axes of transient but heritable non-genetic differences in an otherwise homogeneous cancer population, each of which can result in a distinct drug-resistant fate trajectory. Despite enormous therapeutic benefit in probing these non-genetic variabilities, little is known about their mechanistic origins, timescales, and consequences on fate decisions. For the Cancer Research Foundation Young Investigator Award, I am proposing transdisciplinary approaches to address these gaps. First, I will develop 3DMultiFateMap, a novel multi-omic experimental and computational framework to trace and profile rare-cell plasticity and fate decisions across three cancers and three therapies. 3DMultiFateMap will identify universal and developmental-lineage specific regulatory logic driving diverse resistant fates, and present novel therapeutic opportunities. Second, I will develop a new class of imageable synthetic sensors which can simultaneously measure the realtime activity of several signaling pathways while also keeping track of the clonal information. By coupling metadata from these experiments with our novel deep-learning framework, we will predict rare-cell signaling dynamics and fate choices in the face of drug stress. Lastly, beyond passively monitoring rarecell events, I will combinatorically introduce one or more signaling pathways with varying amplitude and duration with optogenetics and systematically screen for inducers and eliminators of rare-cell plasticity in cancer drug resistance. Together, our multiscale approach—integrating ideas from engineering, synthetic biology, mathematics, and computer science—will provide critical conceptual and technical frameworks to control single-cell variabilities in cancer.