Use of a dynamic disease progression model to estimate prevalence and prognosis for patients with urothelial carcinoma (UC) in the United States (US).

Abstract

e19144 Background: Quantifying the distribution and prognosis of patients with UC as a function of disease stage may allow the impact of existing and novel therapies to be assessed in the real-world setting. We present a dynamic progression model that estimates the incidence, prevalence, and mortality of UC clinical state (CS) in the US. Methods: This UC dynamic progression model used US estimates of UC incidence and distribution of stage at diagnosis from the National Cancer Institute’s (NCI) Surveillance, Epidemiology, and End Results (SEER) database to establish new patients. Progression and mortality for each CS were based on published clinical trials and OPTUM claims data. The simulation started in 1990, introducing incident patient cohorts allocated across initial CSs (Table). Historical therapy distributions were applied for each year as additional incident UC cohorts were introduced into the model. This proprietary model built to annual point prevalence dynamically through historical incidence, progression and mortality. Results: Based on the progression model, point estimates of prevalence, incidence and annual mortality hazard are provided by stage of disease (Table). For all stages of UC, the model estimated a prevalence of 719,387 patients in 2019. For stage II/III and metastatic UC (mUC) disease, the model estimated that 5,205 and 12,499 patients will die in 2019, respectively. This combines to 17,704 which closely aligns with the SEER estimate of 17,670. Conclusions: This dynamic UC progression model provides estimates for incidence, prevalence, and mortality of UC by clinical state at diagnosis. Incorporating associated claims and clinical data with this model could estimate the benefits of newer therapies as they become available. [Table: see text]