Understanding Bladder Cancer Screening Limits Through Comparative Modeling: The Maximum Clinical Incidence Reduction...

Area of Science:

• Oncology
• Epidemiology
• Health Economics

Background:

• Bladder cancer presents significant clinical and economic challenges.
• Key natural history parameters for screening effectiveness, like preclinical reservoir size and sojourn time, are difficult to observe.
• The Cancer Intervention and Surveillance Modeling Network (CISNET) uses standardized stress tests to compare microsimulation models and understand intervention impacts.

Purpose of the Study:

• To compare three independently developed bladder cancer microsimulation models using the CISNET Maximum Clinical Incidence Reduction (MCLIR) and Realistic Clinical Incidence Reduction (RCLIR) frameworks.
• To assess the impact of one-time screening interventions on bladder cancer incidence reduction.
• To understand how model differences influence predictions of screening effectiveness.

Main Methods:

• Applied the CISNET MCLIR/RCLIR protocol to three models (COBRAS, Kystis, SCOUT) calibrated to U.S. epidemiologic targets.
• Simulated a 1950 U.S. birth cohort under no-screening, perfect screening (MCLIR), and realistic screening (RCLIR-1 and RCLIR-2) scenarios.
• Analyzed age-specific clinical incidence, incidence-reduction curves, and cumulative reductions.

Main Results:

• Median preclinical sojourn times varied across models (2.1 to 3.3 years).
• Peak incidence reductions under MCLIR ranged from 20% to 32% at age 65 for White men, with varying persistence.
• Realistic screening (RCLIR) yielded smaller reductions, and most residual incidence was due to new lesion emergence.

Conclusions:

• Three bladder cancer microsimulation models suggest a short detectable preclinical phase, limiting one-time screening effectiveness.
• Differences in model predictions are linked to variations in implied sojourn time and detectable reservoir size.
• Findings support evaluating risk-targeted and repeated screening, and identify priorities for improving bladder cancer natural history inference.