PROJECT SUMMARY This application is being submitted in response to the Notice of Special Interest (NOSI) identified as NOT-CA- 21-100. This career development proposal will provide Dr. Syed Nabeel Zafar, a surgical oncologist and expert in gastrointestinal cancer, with the training and mentorship required to develop expertise in using data sciences to improve the outcomes of patients undergoing surgery for pancreas cancer, with the goal of becoming a successful independent surgeon scientist. Pancreas cancer accounts for over 48,000 deaths per year in the US. Pancreas ductal adenocarcinoma (PDAC) constitutes 90% of pancreas cancer, and while surgery with chemotherapy offers the best chance of long-term survival, most patients have poor outcomes: over half survive less than 3 years from surgery, while only 20% of patients live beyond 5 years. A critical barrier to improving survival is the inability to predict which patients will follow favorable vs. unfavorable disease trajectories. This limitation has led to a one-size-fits-all approach for patients with localized disease. To improve survival in patients undergoing surgery for PDAC, it is critical to identify subsets who may benefit from personalized therapies. Prior prognostic research on PDAC has been limited due to the lack of integrated datasets and limitations in analytic techniques. In this proposal, we will use a novel multicenter electronic health record dataset of 64000 patients with pancreas cancer across 82 institutions across the US. In Aim 1, we will derive, validate, and compare machine learning models to arrive at a highly accurate prognostic model for overall survival after surgery for PDAC. In Aim 2, we will integrate genomic information to further test and improve the accuracy of the model. Completion of this proposal will result in a deeper insight into the factors associated with pancreas cancer survival, the identification of patient phenotypes, digital biomarkers, or genomic mutations associated with poor prognosis, and a highly accurate prognostic model. Such individualized prognostication will aid clinical decision making and precision medicine initiatives to improve survival after resection for PDAC. Dr. Zafar will build upon his clinical and research background to acquire new skills in survival analysis, machine learning, and genomic analysis that are crucial for his transition to an independent surgeon scientist. Dr. Zafar is supported by an exceptional team of mentors, led by Dr. Matthew Churpek who is world renowned for his work on using multicenter EHR data and machine learning to build highly accurate predictive models to identify, risk stratify, and guide personalized care of patients with critical illnesses. Dr. Mark Burkard is an expert in genomics and precision therapies; Dr. Sharon Weber, a surgeon leader, senior pancreas surgeon, and section chief who will facilitate achievement of Dr. Zafar’s career goals; and Dr. Ben Zarzaur, an independently funded surgeon ...