Abstract 515P
Background
Colorectal cancer (CRC) screening models are valuable tools to inform policymakers on decisions for national CRC screening strategies. A new Discrete Event simulation model for the natural history of Colorectal cancer from the Adenoma and Serrated neoplasia pathways (DECAS) was recently developed and calibrated using the data of the German screening colonoscopy registry, which consisted of 5.2 million average-risk individuals aged 55 years and older in the period of 2003-2014. It was validated against the data from the German Center for Cancer Registry Data (ZfKD). We aimed to use DECAS to estimate the long-term effectiveness of colonoscopy screening and perform a between-model comparison.
Methods
We used DECAS to simulate a cohort of averaged-risk individuals from age of 20 to 90 for three scenarios: (1) no screening; (2) one colonoscopy screening at age of 55; (3) two colonoscopies at age of 55 and 65. We assumed the adenoma pathway accounted for 85% of the CRC development and the serrated pathway the other 15%, as well as 100% uptake of screening and surveillance colonoscopy for positive precancerous findings. Outcomes were reductions in CRC cumulative incidence and mortality in the two screening scenarios compared to no screening. We compared the results with incidence reductions estimated from another German CRC Markov model by Brenner et al.
Results
The base case DECAS predicted relative risks of 25-year (age of 55 to 80 years) CRC cumulative incidence at approximately 0.3 and 0.2 for once colonoscopy and twice colonoscopies, respectively. The relative risks of mortality were approximately 0.3 and 0.25 for the two strategies, respectively. The trend in incidence reduction was comparable to those resulting from Brenner et al.’s model. Additional sensitivity analyses on the proportions of CRC developing from adenoma and serrated pathways were performed.
Conclusions
We confirmed the long-term colonoscopy screening effectiveness by modeling with DECAS, with the implications overall consistent with another German modeling study. We will further apply DECAS to perform economic evaluations on various CRC screening strategies and policies.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
German Cancer Research Center (DKFZ).
Disclosure
All authors have declared no conflicts of interest.