Despite tremendous advances in medical therapy for cancer, current treatments often fail to eradicate disease completely, allowing the rare cells that survive to proliferate and drive disease progression. Identifying how these rare cells survive therapy and metastasize is fundamental to our understanding of cancer biology and the development of better therapeutics. Current approaches attempt to identify molecular drivers of these processes by characterizing cancer cells ?after? they become drug resistant or metastatic. However, we and others have shown that as cancer cells grow and adapt to a drug or new environment, they change radically and may bear little resemblance to the ?initial? rare cells that survived treatment and spread. Therefore, even the most sensitive current approach that profiles individual cancer cells may miss fundamental differences that matter at the earliest stages of disease. Instead, we need a way to effectively peer back in time and ask: what was different about those rare cells initially? Here we propose combining high-complexity lineage barcoding, high-throughput sequencing, and RNA FISH to directly isolate and profile the initial cancer cells that give rise to drug resistance and metastasis. In our preliminary work, we created a lentivirus library for delivering unique transcribed barcodes to millions of cells which can be detected by both sequencing and RNA FISH. In parallel, we developed a technique for exponentially amplifying RNA FISH fluorescence, enabling cell sorting based on specific transcripts. Combining these tools as described below will allow us to label and isolate a specific cell lineage based on its future behaviour, such as resistance to drug or invasiveness.
In aim one, we plan to demonstrate this potential by isolating the drug-naive melanoma cells that later give rise to vemurafenib resistant colonies. By performing RNA-seq and ATAC-seq on the isolated cells, we will identify markers and epigenetic regulators of these resistance-progenitor cells and validate them functionally using pharmacologic and CRISPR-based perturbations.
In aim 2, we propose taking our method ?in-vivo? to characterize what is different about the rare melanoma cells that metastasize in human xenograft and syngeneic mouse models. First, by tracking barcodes across tissues forward in time we will identity the order, route and tissue preference of individual metastatic clones. Then, using RNA FISH to label and isolate the ancestors of these clones, we will look back to determine how past differences in gene expression might underlie these behaviours. Successful completion of these aims will reveal how therapy resistance and metastasis arises in melanoma and, more generally, validate a novel method based on FISHable barcodes for studying the origins of rare-cell phenomena.
In cancer, the emergence of cells capable of resisting current therapies and metastasizing to foreign tissues are a major cause of patient suffering and mortality. These initially rare cells continuously evolve during disease progression making it unlikely that what we can measure in their final drug-resistant or metastatic state is representative of what was different about them initially, when they may be most susceptible to adjuvant treatments. Here we propose developing a novel strategy combining single-cell barcoding, RNA FISH and high-throughput sequencing to directly isolate and characterize the initial rare cells that give rise to targeted-therapy resistant and metastatic melanoma.
