COLOFIT:利用年龄、性别、粪便免疫化学及血液检测优化有症状患者结直肠癌诊断的模型开发与内外部验证
本研究基于2018至2022年诺丁汉大学医院NHS信托基金收治的≥18岁疑似结直肠癌且有粪便免疫化学检测(FIT)的有症状成人,开发并验证了结合年龄、性别、粪便血红蛋白浓度、平均红细胞体积和血小板计数的Cox比例风险模型。在结直肠癌风险阈值0.6%(相当于粪便血红蛋白10 μg Hb/g)下,验证模型的整体C统计量为93%(95% CI 92%-95%),校准可接受,所需检查数量较粪便血红蛋白≥10 μg/g策略减少约20%,预测每年约10万疑似患者中可节省超过4500例结肠镜检查,且无证据表明会漏诊更多癌症。
要点
- 模型纳入年龄、性别、平均红细胞体积、血小板计数和粪便血红蛋白浓度预测1年结直肠癌诊断风险。
- 风险阈值0.6%(相当于粪便血红蛋白10 μg/g)时验证模型C统计量93%(95% CI 92%-95%),校准可接受。
- 该模型较粪便血红蛋白≥10 μg/g策略减少约20%检查,预测每年约10万疑似患者可节省超过4500例结肠镜检查。
核心结论
- 将年龄、性别、平均红细胞体积、血小板计数和粪便血红蛋白浓度纳入生存分析模型预测结直肠癌风险,其诊断效用优于简单的粪便血红蛋白≥10 μg/g二分类截断值。
研究设计与样本
- 研究纳入2018年至2022年间诺丁汉大学医院NHS信托基金收治的所有≥18岁、有疑似结直肠癌症状且接受粪便免疫化学检测的成人。
- 使用Cox比例风险模型计算预测的1年结直肠癌诊断,并对年龄、粪便血红蛋白浓度、平均红细胞体积、血小板计数和性别进行选定的多重分数多项式变换,进行外部验证。
主要发现
- 在结直肠癌风险阈值0.6%(相当于粪便血红蛋白10 μg Hb/g)下,验证模型按年龄分层使用Harrell's C指数的整体性能≥0.91%(整体C统计量93%,95% CI 92%-95%),校准可接受。
- 使用该模型检测和漏诊的癌症数量相似,但所需检查数量较粪便血红蛋白≥10 μg/g策略减少约20%。
- 对于每年约100,000名有疑似结直肠癌症状的人群,预测使用该模型可能节省超过4500例结肠镜检查,且无证据表明与粪便血红蛋白≥10 μg/g策略相比会漏诊更多癌症。
关键数值与推荐
- 结直肠癌风险阈值0.6%相当于粪便血红蛋白10 μg Hb/g(μg/g)。
- 验证模型整体C统计量为93%(95% CI 92%-95%),按年龄分层Harrell's C指数≥0.91%。
临床怎么用
- 在初级保健中,可利用粪便免疫化学检测时已有的年龄、性别、平均红细胞体积、血小板计数和粪便血红蛋白浓度信息,通过该模型优化有症状患者结直肠癌检查的选择。
局限性
- 原文未报告局限性。
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: Colorectal cancer (CRC) is the third most common cancer in the United Kingdom and the second largest cause of cancer death. AIM: To develop and validate a model using available information at the time of faecal immunochemical testing (FIT) in primary care to improve selection of symptomatic patients for CRC investigations. METHODS: We included all adults (≥ 18 years) referred to Nottingham University Hospitals NHS Trust between 2018 and 2022 with symptoms of suspected CRC who had a FIT. Predicted 1-year CRC diagnosis were calculated, and externally validated, using Cox proportional hazards modelling with selected multiple fractional polynomial transformations for age, faecal haemoglobin concentration (f-Hb) value, mean corpuscular volume (MCV), platelet count and sex. RESULTS: At a CRC risk threshold of 0.6% (equivalent to f-Hb = 10 μg Hb/g (μg/g)) overall performance of the validated model across age strata using Harrell's C index was ≥ 0.91% (overall C-statistic 93%, 95% CI 92%-95%) with acceptable calibration. Using this model yields similar numbers of detected and missed cancers, but requires ~20% fewer investigations than a f-Hb ≥ 10 μg/g strategy. For approximately 100,000 people per year with symptoms of suspected CRC, we predict it might save > 4500 colonoscopies with no evidence that more cancers would be missed if we used our model compared to using FIT f-Hb ≥ 10 μg/g. CONCLUSIONS: Including age, sex, MCV, platelets and f-Hb in a survival analysis model to predict the risk of CRC yields greater diagnostic utility than a simple binary cut off f-Hb ≥ 10 μg/g.
引用
- Crooks CJ, West J, Jones J, et al. COLOFIT: Development and Internal-External Validation of Models Using Age, Sex, Faecal Immunochemical and Blood Tests to Optimise Diagnosis of Colorectal Cancer in Symptomatic Patients. Aliment Pharmacol Ther. 2025. PMID: 39764729. doi:10.1111/apt.18459
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。