OBJECTIVES: In oncology, prevalence (e.g. the number of patients living with a disease), is often assumed synonymous with surviving ever-diagnosed patients. However, with curable cancers, this may overestimate the number of patients with active disease. While counting ever-diagnosed patients may be useful for some purposes, adjusting for cure when estimating prevalence by removing those no longer treated would provide more specific estimates for healthcare planning. The objective here was to demonstrate the impact of cure adjustment in prevalence estimation, using an example of EOC in the United States (US).

METHODS: Data from the Surveillance, Epidemiology, and End Results (SEER) database (1992-2019) were used to develop an epidemiologic model of incidence, survival, and cure to derive EOC prevalence. Mortality, cure rate and timing were estimated using published data for annualized mortality and survival in incident EOC patients, and conditional survival analysis of SEER data with parametric models used to extrapolate survival beyond 2019.

RESULTS: Conditional survival analysis of SEER data resulted in an estimated 81% of patients surviving 9 years’ post-diagnosis further surviving to 14 years, at which point cure was assumed for all patients. Annual incidence of EOC ranged from 17,354 in 2019 to 20,428 in 2015. The unadjusted estimate of prevalent EOC patients in 2022 was 181,379. When patients were removed for cure beginning at 9 years post-diagnosis, the estimated number of active disease cases in 2022 was 106,549; a reduction of 41.3%.

CONCLUSIONS: Prevalence estimates of EOC declined substantially when removing cured patients. Greater specificity in prevalence estimation, in particular the number of patients requiring active care, will be important in understanding clinical and economic burdens of EOC. These findings also have implications for other therapeutic areas with emerging curative treatment options, not only in oncology but also for conditions treatable with cellular and gene therapies for example.