Developing and Validating Prognostic Models for Black women with Endometrial Cancer following Hysterectomy - PROJECT SUMMARY Endometrial cancer (EC) incidence and mortality are rising rapidly in the U.S. Black women are experiencing the sharpest increase, with incidence and mortality rising 46% and 29%, respectively, from 1999-2015. While the 5-year survival for EC overall is 80%, for Black patients it is only 63%. Black women are more commonly diagnosed with aggressive histologic subtypes (non-endometrioid; ~35% vs ~17% non-Black) not amenable to surgical resection alone. Primary staging and treatment for EC is surgical hysterectomy with bilateral salpingo- oophorectomy; several adjuvant therapies allow for the tailoring of post-surgical treatment, requiring careful balancing of prognosis/recurrence risks with adverse effects of therapy to improve outcomes and avoid mistreatment. However, there are no prognostic tools to inform discussion of adjuvant therapies for Black women diagnosed with EC. This Early K99/R00 application leverages years of NIH-funding and collaboration to develop and externally validate prognostic tools for Black women with EC post-hysterectomy. This research will inform discussion of tailored adjuvant therapies in clinical settings and is foundational for Dr. Peeri to achieve research independence focused on creating clinical decision-making tools inclusive of minority populations to reduce cancer health disparities. Our specific aims are to: 1A) build a robust prognostic tool tailored for Black EC patients using readily available clinical and epidemiologic factors to predict survival 3- and 5- years post-hysterectomy, and to externally validate this tool, 1B) compare model performance for Black patients with an established tool developed primarily in White women, 1C) develop a web-based survival tool for clinician and patient use; 2A) assess the utility of adding existing TCGA molecular subtypes to our prognostic tool for Black women and externally validate this tool, and 2B) use an established clinical tool in our population as a point of reference for evaluating the performance of our model; 3) Build biomarker-based prognostic tools to predict overall survival at 3- and 5-years post-hysterectomy in 3A) Black women, expanding upon the prognostic tool developed in Aim 1, and 3B) White women, expanding on an established tool. The training goals are to 1) receive training in rigorous prognostic model development and validation methods, 2) receive training in cancer genomics including advanced methods for whole-exome sequence data, 3) expand knowledge of biomarkers of clinical significance and machine-learning methods for model development, 4) lay a foundation to support an independent research career in cancer health disparities and cancer survivorship. Ultimately, this proposal will advance the career of a young investigator with a strong research background from postdoctoral fellow to independence.
