Bladder cancer is the sixth most common cancer in the United States, with an estimated 83,000 new cases expected in 2024.¹ While most cases are diagnosed at an early stage, the disease remains among the most expensive to manage due to its high recurrence rate.^2,3 Despite reductions in major risk factors such as smoking, progress in lowering bladder cancer incidence and mortality has been limited.^4,5 Currently, routine screening for bladder cancer is not recommended due to insufficient evidence.⁶ Available screening tools–such as urine cytology and biomarker tests–lack the sensitivity and specificity needed to reliably detect early-stage disease.⁷ The gold standard diagnostic tool, cystoscopy, is invasive and costly, making it impractical for widespread population use.⁸ Post-diagnosis surveillance also presents challenges, with questions remaining about how best to tailor the frequency and combination of follow-up tests based on individual risk.^9,10

To address these gaps, the CISNET Bladder Cancer Incubator was established in 2021. The goal is to identify effective and cost-efficient strategies for bladder cancer screening, surveillance, and treatment through advanced population modeling, ultimately supporting future guideline development. The Incubator began with two modeling groups:

COBRAS (Cancer of the Bladder R-based Analytic Simulator), with investigators from the University of Ottawa, Stanford University, and the University of Pittsburgh.
Kystis, with investigators from Brown University and Tufts University.

In 2022, the SCOUT (Simulation of Cancers of the Urinary Tract) group from Columbia University joined as an affiliate member. All three modeling groups maintain methodological independence and are supported by the Coordinating Center at Brown University.

Model Structure

The three CISNET Bladder Cancer models simulate individual life histories to estimate population-level trends in bladder cancer. COBRAS and Kystis operate in continuous time, while SCOUT uses monthly cycles. All models simulate U.S. birth cohorts beginning in 1900, aging them forward with realistic patterns of bladder cancer incidence, progression, and death, but without accounting for migration. The simulations incorporate age-specific bladder cancer hazards, and individuals may develop one or more tumors that can progress from non-invasive to invasive and metastatic stages. Diagnosis is determined by a combination of tumor characteristics, symptoms (e.g., bleeding or voiding), and demographic factors like race and sex to capture observed differences in stage and age at diagnosis.

All models incorporate smoking exposure. COBRAS and Kystis use historically accurate smoking patterns generated by CISNET’s Smoking History Generator,¹¹ while SCOUT applies fixed, age-specific smoking rates. All models simulate key events such as tumor onset, progression, metastasis, diagnosis, and cause-specific death, informed by U.S. mortality data.¹² In addition to bladder cancer natural history, SCOUT simulates chronic kidney disease progression by modeling declines in kidney function and related risk factors including diabetes and hypertension.

Model inputs and calibration targets included the Surveillance, Epidemiology, and End Results (SEER) data on bladder cancer incidence, stage, age distribution, mortality, and relative survival;¹³ individual and published data from nine EORTC trials on early-stage bladder cancer to inform recurrence and progression;¹⁴ cohort-specific U.S. mortality projections from the CDC, stratified by sex and race;¹² population estimates and forecasts from the U.S. Census;^15,16 and additional data sourced from a related project that develops and maintains evidence maps of the bladder cancer literature. Calibration was performed using distinct optimization strategies, and all models underwent extensive verification to ensure biological plausibility, logical consistency, and dimensional correctness. Model outcomes are probabilistic, reflecting randomness in event timing and uncertainty in parameter values, with final results drawn from the top 1,000 best-fitting parameter sets.

Public Health Impact

Understanding the natural history of bladder cancer is crucial for evaluating prevention and control strategies, yet many events, such as the duration between when a cancer is first screen-detectable and when it presents clinically, are not directly observable in routine data. The CISNET Bladder Cancer models address this challenge by using available epidemiological and clinical data to infer unobserved events. These models provide a valuable tool for decision-makers to assess the long-term impact of health policies and newly emergent technologies on bladder cancer incidence, prevalence, and mortality in the U.S.–for example, evaluating how environmental carcinogen control, targeted early detection, intensified surveillance for high-risk groups, or novel treatments for advanced cancers could influence trends in bladder cancer outcomes.

References

Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin. Jan-Feb 2024;74(1):12-49. doi:10.3322/caac.21820
Burger M, Catto JW, Dalbagni G, et al. Epidemiology and risk factors of urothelial bladder cancer. Eur Urol. Feb 2013;63(2):234-41. doi:10.1016/j.eururo.2012.07.033
Clark O, Sarmento T, Eccleston A, et al. Economic Impact of Bladder Cancer in the USA. Pharmacoecon Open. Nov 2024;8(6):837-845. doi:10.1007/s41669-024-00512-8
Leas EC, Trinidad DR, Pierce JP, McMenamin SB, Messer K. Trends in cigarette consumption across the United States, with projections to 2035. PLoS One. 2023;18(3):e0282893. doi:10.1371/journal.pone.0282893
Su X, Tao Y, Chen F, Han X, Xue L. Trends in the global, regional, and national burden of bladder cancer from 1990 to 2021: an observational study from the global burden of disease study 2021. Sci Rep. Mar 5 2025;15(1):7655. doi:10.1038/s41598-025-92033-5
Moyer VA, Force USPST. Screening for bladder cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. Aug 16 2011;155(4):246-51. doi:10.7326/0003-4819-155-4-201108160-00008
Chou R, Dana T. Screening adults for bladder cancer: a review of the evidence for the U.S. preventive services task force. Ann Intern Med. Oct 5 2010;153(7):461-8. doi:10.7326/0003-4819-153-7-201010050-00009
Devlies W, de Jong JJ, Hofmann F, et al. The Diagnostic Accuracy of Cystoscopy for Detecting Bladder Cancer in Adults Presenting with Haematuria: A Systematic Review from the European Association of Urology Guidelines Office. Eur Urol Focus. Jan 2024;10(1):115-122. doi:10.1016/j.euf.2023.08.002
Parrao D, Lizana N, Saavedra C, et al. Active Surveillance in Non-Muscle Invasive Bladder Cancer, the Potential Role of Biomarkers: A Systematic Review. Curr Oncol. Apr 12 2024;31(4):2201-2220. doi:10.3390/curroncol31040163
Su ZT, Florissi IS, Mahon KM, et al. Varying the intensity of cystoscopic surveillance for high-risk non-muscle-invasive bladder cancer. BJU Int. Jan 2025;135(1):148-155. doi:10.1111/bju.16521
Jeon J, Holford TR, Levy DT, et al. Smoking and Lung Cancer Mortality in the United States From 2015 to 2065: A Comparative Modeling Approach. Ann Intern Med. Nov 20 2018;169(10):684-693. doi:10.7326/M18-1250
Prevention CfDCa. National Vital Statistics System – Mortality Data. Accessed March 31, 2025. https://www.cdc.gov/nchs/nvss/deaths.htm.
Institute NC. Surveillance, Epidemiology, and End Results (SEER) Program. SEER*Stat Database. Accessed March 31, 2025. https://seer.cancer.gov/data/
Sylvester RJ, van der Meijden AP, Oosterlinck W, et al. Predicting recurrence and progression in individual patients with stage Ta T1 bladder cancer using EORTC risk tables: a combined analysis of 2596 patients from seven EORTC trials. Eur Urol. Mar 2006;49(3):466-5; discussion 475-7. doi:10.1016/j.eururo.2005.12.031
Bureau USC. Population and Housing Unit Estimates. Accessed March 31, 2025. https://www.census.gov/popest
Bureau USC. Population Projections. Accessed March 31, 2025. https://www.census.gov/popest