Gini Andrea, Buskermolen Maaike, Senore Carlo, Anttila Ahti, Novak Mlakar Dominika, Veerus Piret, Csanádi Marcell, Jansen Erik E L, Zielonke Nadine, Heinävaara Sirpa, Széles György, Segnan Nereo, de Koning Harry J, Lansdorp-Vogelaar Iris
Department of Public Health, Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands.
SC Epidemiology, Screening, Cancer Registry, Città della Salute e della Scienza University Hospital, CPO, Turin, Italy.
MDM Policy Pract. 2021 Jan 29;6(1):2381468320984974. doi: 10.1177/2381468320984974. eCollection 2021 Jan-Jun.
Validated microsimulation models have been shown to be useful tools in providing support for colorectal cancer (CRC) screening decisions. Aiming to assist European countries in reducing CRC mortality, we developed and validated three regional models for evaluating CRC screening in Europe. Microsimulation Screening Analysis-Colon (MISCAN-Colon) model versions for Italy, Slovenia, and Finland were quantified using data from different national institutions. These models were validated against the best available evidence for the effectiveness of screening from their region (when available): the Screening for COlon REctum (SCORE) trial and the Florentine fecal immunochemical test (FIT) screening study for Italy; the Norwegian Colorectal Cancer Prevention (NORCCAP) trial and the guaiac fecal occult blood test (gFOBT) Finnish population-based study for Finland. When published evidence was not available (Slovenia), the model was validated using cancer registry data. Our three models reproduced age-specific CRC incidence rates and stage distributions in the prescreening period. Moreover, the Italian and Finnish models replicated CRC mortality reductions (reasonably) well against the best available evidence. CRC mortality reductions were predicted slightly larger than those observed (except for the Florentine FIT study), but consistently within the corresponding 95% confidence intervals. Our findings corroborate the MISCAN-Colon reliability in supporting decision making on CRC screening. Furthermore, our study provides the model structure for an additional tool (EU-TOPIA CRC evaluation tool: http://miscan.eu-topia.org) that aims to help policymakers and researchers monitoring or improving CRC screening in Europe.
经验证的微观模拟模型已被证明是为结直肠癌（CRC）筛查决策提供支持的有用工具。为了帮助欧洲国家降低结直肠癌死亡率，我们开发并验证了三个用于评估欧洲结直肠癌筛查的区域模型。利用来自不同国家机构的数据，对意大利、斯洛文尼亚和芬兰的微观模拟筛查分析-结肠癌（MISCAN-Colon）模型版本进行了量化。这些模型根据其所在地区筛查有效性的最佳现有证据（如有）进行了验证：意大利的结直肠癌筛查（SCORE）试验和佛罗伦萨粪便免疫化学检测（FIT）筛查研究；芬兰的挪威结直肠癌预防（NORCCAP）试验和基于芬兰人群的愈创木脂粪便潜血试验（gFOBT）研究。当没有已发表的证据时（斯洛文尼亚），则使用癌症登记数据对模型进行验证。我们的三个模型再现了筛查前期特定年龄的结直肠癌发病率和分期分布。此外，意大利和芬兰的模型与最佳现有证据相比，（合理地）很好地复制了结直肠癌死亡率的降低情况。预测的结直肠癌死亡率降低幅度略大于观察到的幅度（佛罗伦萨FIT研究除外），但始终在相应的95%置信区间内。我们的研究结果证实了MISCAN-Colon在支持结直肠癌筛查决策方面的可靠性。此外，我们的研究为另一个工具（欧盟-TOPIA CRC评估工具：http://miscan.eu-topia.org）提供了模型结构，该工具旨在帮助政策制定者和研究人员监测或改善欧洲的结直肠癌筛查。
