文献速递流行病学与生物统计医学人工智能与信息学

乳腺癌预后统计模型与机器学习模型的开发及内部-外部验证:队列研究

Development and internal-external validation of statistical and machine learning models for breast cancer prognostication: cohort study.
PubMed 文献 · BMJ · 2023-05-10

基于英格兰QResearch初级保健数据库141765名20岁及以上侵袭性乳腺癌女性,开发并比较两种回归模型与两种机器学习模型估计10年乳腺癌死亡风险,采用内部-外部交叉验证和随机效应Meta分析评估性能。Cox模型Harrell's C指数最高(0.858),回归模型校准与临床效用优于机器学习模型。

要点

  • Cox模型区分度最高,汇总Harrell's C指数0.858(95%CI 0.853-0.864)。
  • 竞争风险回归模型区分度良好,汇总Harrell's C指数0.849(95%CI 0.839-0.859)。
  • 决策曲线分析提示Cox与竞争风险回归模型临床效用可能高于两种机器学习方法。

核心结论

  • 在任意分期乳腺癌女性中,基于本数据集可用预测因子,回归方法比机器学习方法性能更好且更一致,可能值得进一步评估用于潜在临床用途如分层随访。

研究设计与样本

  • 基于人群的队列研究,使用英格兰QResearch初级保健数据库,个体水平链接国家癌症登记、医院统计和国家死亡登记。
  • 纳入141765名20岁及以上、2000年1月1日至2020年12月31日诊断为侵袭性乳腺癌的女性。
  • 构建四种模型:两种回归(Cox比例风险回归和竞争风险回归)和两种机器学习(XGBoost和人工神经网络),采用内部-外部交叉验证评估,随机效应Meta分析汇总区分度和校准指标。

主要发现

  • 中位随访4.16年(四分位距1.76-8.26),发生21688例乳腺癌相关死亡和11454例其他原因死亡。
  • 限制乳腺癌诊断后最长10年随访,总随访688564.81人年中发生20367例乳腺癌相关死亡,粗乳腺癌死亡率为每10000人年295.79例(95%CI 291.75-299.88)。
  • Cox模型随机效应Meta分析汇总Harrell's C指数最高,为0.858(95%CI 0.853-0.864,95%预测区间0.843-0.873),校准图显示可接受校准。
  • 竞争风险回归模型区分度良好:汇总Harrell's C指数0.849(95%CI 0.839-0.859,95%预测区间0.821-0.876),汇总指标上缺乏系统性校准不良证据。
  • 机器学习模型总体区分度可接受:XGBoost Harrell's C指数0.821(95%CI 0.813-0.828,95%预测区间0.805-0.837);神经网络0.847(95%CI 0.835-0.858,95%预测区间0.816-0.878),但校准不良模式更复杂,区域和分期特异性性能变异更大。

关键数值与推荐

  • Cox和竞争风险模型均纳入诊断时年龄、体重指数、吸烟状态、诊断途径、激素受体状态、癌症分期和乳腺癌分级作为预测因子。

临床怎么用

  • 作者认为回归方法可能值得进一步评估用于潜在临床用途,如分层随访。

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

OBJECTIVE: To develop a clinically useful model that estimates the 10 year risk of breast cancer related mortality in women (self-reported female sex) with breast cancer of any stage, comparing results from regression and machine learning approaches. DESIGN: Population based cohort study. SETTING: QResearch primary care database in England, with individual level linkage to the national cancer registry, Hospital Episodes Statistics, and national mortality registers. PARTICIPANTS: 141 765 women aged 20 years and older with a diagnosis of invasive breast cancer between 1 January 2000 and 31 December 2020. MAIN OUTCOME MEASURES: Four model building strategies comprising two regression (Cox proportional hazards and competing risks regression) and two machine learning (XGBoost and an artificial neural network) approaches. Internal-external cross validation was used for model evaluation. Random effects meta-analysis that pooled estimates of discrimination and calibration metrics, calibration plots, and decision curve analysis were used to assess model performance, transportability, and clinical utility. RESULTS: During a median 4.16 years (interquartile range 1.76-8.26) of follow-up, 21 688 breast cancer related deaths and 11 454 deaths from other causes occurred. Restricting to 10 years maximum follow-up from breast cancer diagnosis, 20 367 breast cancer related deaths occurred during a total of 688 564.81 person years. The crude breast cancer mortality rate was 295.79 per 10 000 person years (95% confidence interval 291.75 to 299.88). Predictors varied for each regression model, but both Cox and competing risks models included age at diagnosis, body mass index, smoking status, route to diagnosis, hormone receptor status, cancer stage, and grade of breast cancer. The Cox model's random effects meta-analysis pooled estimate for Harrell's C index was the highest of any model at 0.858 (95% confidence interval 0.853 to 0.864, and 95% prediction interval 0.843 to 0.873). It appeared acceptably calibrated on calibration plots. The competing risks regression model had good discrimination: pooled Harrell's C index 0.849 (0.839 to 0.859, and 0.821 to 0.876, and evidence of systematic miscalibration on summary metrics was lacking. The machine learning models had acceptable discrimination overall (Harrell's C index: XGBoost 0.821 (0.813 to 0.828, and 0.805 to 0.837); neural network 0.847 (0.835 to 0.858, and 0.816 to 0.878)), but had more complex patterns of miscalibration and more variable regional and stage specific performance. Decision curve analysis suggested that the Cox and competing risks regression models tested may have higher clinical utility than the two machine learning approaches. CONCLUSION: In women with breast cancer of any stage, using the predictors available in this dataset, regression based methods had better and more consistent performance compared with machine learning approaches and may be worthy of further evaluation for potential clinical use, such as for stratified follow-up.

来源:BMJ,查看原文。版权归原作者所有。

引用

  1. Clift AK, Dodwell D, Lord S, et al. Development and internal-external validation of statistical and machine learning models for breast cancer prognostication: cohort study. BMJ. 2023. PMID: 37164379. doi:10.1136/bmj-2022-073800

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。