结肠镜与两年一次FIT筛查:COLONPREV试验的事后持续策略分析
加权后两组结直肠癌发病率和全因死亡率相近,结肠镜组结直肠癌死亡率数值更低但估计不精确,提示依从性路径在实用筛查试验中的重要性。
要点
- 持续FIT与一次性结肠镜的10年结直肠癌发病率和全因死亡率相似。
- 结肠镜组结直肠癌死亡率数值更低,但置信区间宽、估计不精确。
- 实用筛查试验中依从性路径影响比较结局的解读。
核心结论
- 持续结肠镜与持续FIT策略的10年总体结局相似。
研究设计与样本
- 采用逆概率加权调整观察到的混杂和 informative 方案偏离。
主要发现
- 加权10年结直肠癌发病率:FIT 0.80% vs 结肠镜0.86%,风险差0.06%(95%CI -0.25%至0.37%)。
- 加权10年全因死亡率:FIT 4.02% vs 结肠镜4.09%,风险差0.07%(95%CI -0.76%至0.84%)。
- 加权10年结直肠癌死亡率:FIT 0.072% vs 结肠镜0.042%,风险差-0.03%(95%CI -0.12%至0.06%),估计不精确。
临床怎么用
- 解读实用筛查试验时应考虑依从性路径对比较结局的影响。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: Pragmatic colorectal cancer (CRC) screening trials compare colonoscopy with faecal immunochemical testing (FIT), but differences in adherence complicate interpretation of comparative outcomes. OBJECTIVE: To compare 10-year risks of CRC mortality, CRC incidence and all-cause mortality under one-time colonoscopy versus sustained FIT strategies in the COLONPREV Trial. DESIGN: We conducted a post hoc per-protocol analysis of 17 270 screening initiators aged 50-69 years in eight Spanish regions, using inverse probability weighting to adjust for observed confounding and informative protocol deviations. RESULTS: Weighted 10-year risks were similar for CRC incidence (0.80% FIT vs 0.86% colonoscopy; risk difference (RD) 0.06%, 95% CI -0.25% to 0.37%) and all-cause mortality (4.02% vs 4.09%; RD 0.07%, 95% CI -0.76% to 0.84%). CRC mortality was lower with colonoscopy (0.072% FIT vs 0.042% colonoscopy; RD -0.03%, 95% CI -0.12% to 0.06%), but estimates were imprecise. CONCLUSION: Sustained colonoscopy and FIT showed similar 10-year outcomes overall, highlighting the importance of adherence pathways in pragmatic screening trials. TRIAL REGISTRATION NUMBER: ClinicalTrials.gov: NCT00906997.
引用
- Castells A, Quintero E, Chen H, et al. Colonoscopy versus biennial FIT screening: a post hoc sustained-strategy analysis of the COLONPREV Trial. Gut. 2026. PMID: 42270316. doi:10.1136/gutjnl-2026-338896
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。