PROJECT SUMMARY/ABSTRACT Locally advanced rectal cancer, defined by spread to lymph nodes or extension through the full rectal wall thickness, is diagnosed in over 15,000 US patients yearly. Standard treatment for locally advanced rectal cancer in the US involves chemotherapy and radiation therapy (chemoradiation, CRT) prior to surgical removal of the entire rectum. Patients may experience undesirable post-surgical consequences such as permanent colostomy or altered anorectal function. In light of this, much ongoing research focuses on strategies to avoid surgical intervention in the subset of patients who might be cured with CRT alone. At the time of surgery, 10- 30% of tumors are eradicated by CRT (i.e., they demonstrate a pathologic complete response) while other tumors show little to no response. This variability in response is incompletely understood, but likely relates to the fact that one tumor may contain numerous different tumor cell subpopulations, all with different genotypes and epigenetic (or non-genetic) features. These diverse subpopulations and variable genetic/epigenetic features can interact, resulting in a dynamic and evolving tumor cell network that is poorly understood. In an increasing number of cancers, there is evidence supporting the emergence of treatment-induced evolutionary “traps,” or “collateral sensitivities”, where resistance to one therapy results in sensitivity to another therapy. Whether this occurs in rectal cancer, and how to leverage it into treatment strategies, is unclear. To address these questions, it is crucial to select a representative experimental model. Thus, the current proposal leverages an already-established co-clinical pipeline from which patient derived tumor samples are used to generate paired, patient-specific organoid and xenograft models. Two broad analytic approaches will be undertaken to identify trends in CRT-induced tumor evolution and elucidate potential synergistic approaches to improve CRT response. Specific Aim 1 will dissect CRT-induced changes using genomic, epigenomic, and single cell profiling techniques, including an advanced method combining single cell RNA-seq with lineage tracing. Specific Aim 2 employs an innovative iterative paired CRISPR/high throughput drug screen approach to identify evolving collateral sensitivities in CRT-treated rectal cancer. To validate the findings in both Aims, advanced patient-derived rectal cancer organoid and xenograft models will be utilized. Research, scientific instruction, and career development will be supported by an expert panel of mentors and collaborators and a structured training plan. Successful completion of both Aims will be promoted by the expertise and existing screening and genomic infrastructure in the co-mentors’ laboratories and leverages a unique co-clinical platform already established by close institutional collaborators. This research will establish a patient-derived platform for interrogating CRT-induced tumor evolution...