Development and Potential Implementation of Endometrial Cancer Screening in a Poor-Prognosis Endemic Cancer Community [ID 2683642]

INTRODUCTION: Endometrial cancer incidence and mortality both are increasing in the United States. Certain patient populations are disproportionately affected by this disease. Currently, there is no recommendation from any professional society to screen high-risk patients for premalignancy or cancer. Our patient population is composed of predominantly Black women, a group that experiences worse endometrial cancer outcomes than other demographics. A risk assessment nomogram may be able to be developed using data from this unique patient population that can identify women at high risk of developing endometrial disease. METHODS: Endometrial pathology specimens collected at a large safety-net hospital between the years of 2013 and 2022 were analyzed. A retrospective chart review was performed on this data, looking for the patient characteristics of race, age, body mass index (BMI), parity, hemoglobin A 1C level, and tamoxifen use history at the time of specimen collection. Odds ratios were calculated for those variables individually as well as collectively to determine whether they could be used to predict risk of endometrial disease in a woman with some of all those characteristics. RESULTS: The odds ratios for age, BMI, and parity were statistically significant between benign and abnormal pathologic specimens. The area under the receiver operating characteristic curve was 0.76 when combining age, BMI, and parity. CONCLUSION: Our data suggest that at-risk patients can be reliably identified using simple characteristics. Strengths of our study include a unique patient population and large database. Future directions include refinement of the model and prospective application in the clinical setting.