Comparative Modeling of Recent and Projected Trends in the Incidence and Mortality of Uterine Cancer.

Abstract

To aid in cancer control and prevention activities by developing, calibrating, and validating three distinct natural history simulation models of uterine cancer, a growing public health concern. To perform comparative analyses, we developed two state-transition microsimulation models and a multistage clonal expansion model of uterine cancer. The models simulate uterine cancer incidence and mortality. All three models were calibrated to common data on the incidence of uterine cancer from the Surveillance, Epidemiology, and End Results (SEER) 18 database. Each model accounts for changing trends in hysterectomy and obesity over time and simulates incidence and mortality for endometrioid and nonendometrioid tumors and uterine sarcoma. After calibration, we projected the incidence and mortality of uterine cancer to 2050. The three uterine cancer models were well calibrated to population data and produced comparable results for projecting the burden of disease through 2050. Among non-Hispanic White women aged 40 years or older, the models project that by 2050 the incidence of uterine cancer will rise to 76.1-81.8 per 100,000 woman-years, up from 2018 SEER incidence of 60.0 per 100,000 woman-years. Among non-Hispanic Black women, new cases will rise to 90.3-107.2 per 100,000 woman-years, up from 2018 SEER incidence of 61.3 per 100,000 woman-years. Within these populations, incidence-based mortality will increase to 11.3-12.3 deaths per 100,000 woman-years for non-Hispanic White women and to 28.2-35.7 deaths per 100,000 woman-years for non-Hispanic Black women. Three distinct mathematical simulation models of uterine cancer have been calibrated to observed population-based incidence and mortality. All three models project substantial and continued increases in the incidence and mortality of uterine cancer.