Colonoscopic surveillance is recommended in patients with colonic inflammatory bowel disease (IBD) given their increased risk of colorectal cancer (CRC). We aimed to develop and validate a dynamic prediction model for the occurrence of advanced colorectal neoplasia (aCRN, including high-grade dysplasia and CRC) in IBD. We pooled data from 6 existing cohort studies from Canada, The Netherlands, the United Kingdom, and the United States. Patients with IBD and an indication for CRC surveillance were included if they underwent at least 1 follow-up procedure. Exclusion criteria included prior aCRN, prior colectomy, or an unclear indication for surveillance. Predictor variables were selected based on the literature. A dynamic prediction model was developed using a landmarking approach based on Cox proportional hazard modeling. Model performance was assessed with Harrell’s concordance-statistic (discrimination) and by calibration curves. Generalizability across surveillance cohorts was evaluated by internal–external cross-validation. The surveillance cohorts comprised 3731 patients, enrolled and followed-up in the time period from 1973 to 2021, with a median follow-up period of 5.7 years (26,336 patient-years of follow-up evaluation); 146 individuals were diagnosed with aCRN. The model contained 8 predictors, with a cross-validation median concordance statistic of 0.74 and 0.75 for a 5- and 10-year prediction window, respectively. Calibration plots showed good calibration. Internal–external cross-validation results showed medium discrimination and reasonable to good calibration. The new prediction model showed good discrimination and calibration, however, generalizability results varied. Future research should focus on formal external validation and relate predicted aCRN risks to surveillance intervals before clinical application.

中文翻译：

---

炎症性肠病晚期结直肠肿瘤的动态预测

鉴于结肠炎性肠病 (IBD) 患者患结直肠癌 (CRC) 的风险增加，建议对其进行结肠镜监测。我们的目的是开发和验证 IBD 中晚期结直肠肿瘤（aCRN，包括高度不典型增生和 CRC）发生的动态预测模型。我们汇集了来自加拿大、荷兰、英国和美国的 6 项现有队列研究的数据。如果患有 IBD 且有 CRC 监测指征的患者接受了至少 1 次随访手术，则将被纳入其中。排除标准包括既往 aCRN、既往结肠切除术或监测指征不明确。根据文献选择预测变量。使用基于 Cox 比例风险模型的地标方法开发了动态预测模型。模型性能通过 Harrell 的一致性统计（歧视）和校准曲线进行评估。通过内部-外部交叉验证评估了监测队列的普遍性。监测队列由 3731 名患者组成，在 1973 年至 2021 年期间进行入组和随访，中位随访期为 5.7 年（随访评估为 26,336 患者年）； 146 人被诊断患有 aCRN。该模型包含 8 个预测变量，5 年和 10 年预测窗口的交叉验证中位数一致性统计量分别为 0.74 和 0.75。校准图显示良好的校准。内部-外部交叉验证结果显示中等区分度和合理到良好的校准。新的预测模型表现出良好的辨别力和校准能力，但普遍性结果有所不同。未来的研究应侧重于正式的外部验证，并将预测的 aCRN 风险与临床应用前的监测间隔联系起来。