文献速递膀胱癌与上尿路尿路上皮癌

人工智能预测非肌层浸润性膀胱癌BCG反应的系统综述

Artificial intelligence for predicting BCG response in non-muscle-invasive bladder cancer: a systematic review.
PubMed 文献 · BJU Int · 2026-08-17

数字病理CHAI平台是唯一提供预测价值初步信号的模型,显示BCG与吉西他滨/多西他赛显著交互作用(P=0.029);整合模型PROGRxN-BCa和DeepSurv优于标准计算器但增益有限。仅40%研究进行外部验证且无前瞻性验证。

要点

  • 数字病理CHAI平台分层高级别复发(HR 2.08)、进展(HR 3.87)和BCG无应答疾病(HR 2.31)。
  • CHAI显示BCG vs 吉西他滨/多西他赛显著交互作用(P=0.029),为首个预测价值信号。
  • PROGRxN-BCa(C-index 0.79)和DeepSurv(C-index 0.881)优于标准计算器,ΔC-index 0.05-0.10。

核心结论

  • 感知型AI尤其是数字病理具有最高外部验证水平,并提供唯一对BCG vs 替代方案有预测价值初步信号的生物标志物。
  • 整合模型如PROGRxN-BCa和DeepSurv优于标准计算器但增益有限,且可免费获取。
  • 前瞻性验证、系统校准报告和治疗-生物标志物交互分析是临床采用前的优先事项。

研究设计与样本

  • 系统综述,遵循PRISMA 2020指南,PROSPERO注册号CRD420261376808。
  • 检索PubMed、EMBASE和Web of Science截至2026年4月。
  • 纳入开发或验证AI/ML模型预测BCG相关结局且有定量性能指标的研究。
  • 使用PROBAST评估偏倚风险。

主要发现

  • 按输入数据:6项使用数字病理,2项放射组学,2项基因组学/转录组学,4项临床病理标志物,1项尿液生物标志物。
  • 数字病理CHAI平台在12个国际中心验证,分层高级别复发(HR 2.08)、进展(HR 3.87)和BCG无应答疾病(HR 2.31)。
  • CHAI是唯一提供预测价值初步信号的模型,显示BCG vs 吉西他滨/多西他赛显著交互作用(P=0.029)。
  • 整合模型PROGRxN-BCa(C-index 0.79)和DeepSurv(C-index 0.881)优于标准计算器但增益有限(ΔC-index 0.05-0.10)。
  • 仅40%研究进行外部验证,无前瞻性验证;5项研究存在高偏倚风险。

关键数值与推荐

  • 数字病理CHAI平台:高级别复发HR 2.08,进展HR 3.87,BCG无应答疾病HR 2.31。
  • CHAI显示BCG vs 吉西他滨/多西他赛交互作用P=0.029。
  • PROGRxN-BCa的C-index为0.79,DeepSurv的C-index为0.881,ΔC-index 0.05-0.10。
  • 仅40%研究进行外部验证,无前瞻性验证;5项研究高偏倚风险。

临床怎么用

  • 数字病理AI可作为感知工具从组织学提取信号,辅助识别BCG无应答者。
  • 整合模型可重新加权临床病理、分子或尿液变量,优于标准计算器但增益有限。
  • 在BCG短缺背景下,预测BCG vs替代方案反应的生物标志物日益重要。
  • 临床采用前需前瞻性验证和系统校准报告,理想情况下嵌入BRIDGE试验(NCT05538663)。

局限性

  • 仅40%研究进行外部验证,无前瞻性验证。
  • 5项研究存在高偏倚风险。
  • 整合模型相对标准计算器的增益有限(ΔC-index 0.05-0.10)。

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

OBJECTIVES: To systematically identify, appraise and synthesise artificial intelligence (AI) and machine-learning (ML) models that predict treatment response and clinical outcomes after intravesical bacillus Calmette-Guérin (BCG) in non-muscle-invasive bladder cancer (NMIBC), a setting in which current risk calculators underperform, and identifying non-responders has become urgent as alternatives to BCG enter practice. METHODS: PubMed, EMBASE and Web of Science were searched through April 2026 following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines (International Prospective Register of Systematic Reviews [PROSPERO] number CRD420261376808). Studies developing or validating AI/ML models for BCG-associated outcomes with a quantitative performance metric were included. Risk of bias was assessed with the Prediction model Risk Of Bias ASsessment Tool (PROBAST). Evidence was synthesised along a clinically oriented framework: AI as a perceptual tool (extracting signal from histology or imaging) or integrative tool (re-weighting clinicopathological, molecular, or urinary variables). RESULTS: A total of 15 studies (>24 900 patients) were included: seven perceptual, eight integrative. By input data, six used digital pathology, two radiomics, two genomics/transcriptomics, four clinicopathological markers, and one urinary biomarkers. The digital-pathology Computational Histology Artificial Intelligence (CHAI) platform, validated across 12 international centres, stratified high-grade recurrence (hazard ratio [HR] 2.08), progression (HR 3.87) and BCG-unresponsive disease (HR 2.31), and was the only model providing a first signal of predictive value, demonstrating a significant BCG vs gemcitabine/docetaxel interaction (P = 0.029). Integrative models PROGRxN-BCa (concordance index [C-index] 0.79) and DeepSurv (C-index 0.881) outperformed standard calculators but with modest gains (ΔC-index 0.05-0.10). Only 40% of studies performed external validation, none prospectively; five were at high risk of bias. CONCLUSION: Perceptual AI, particularly digital pathology, has the highest external validation and provides the only biomarker with a first signal of predictive value for BCG vs alternatives, increasingly relevant in the context of the global BCG shortage. Integrative models such as PROGRxN-BCa and DeepSurv outperform standard calculators with incremental gains and are freely accessible. Prospective validation, systematic calibration reporting and treatment-by-biomarker interaction analyses, ideally embedded in trials such as the BRIDGE trial (ClinicalTrials.gov identifier: NCT05538663), remain priorities before clinical adoption.

来源:BJU Int,查看原文。版权归原作者所有。

引用

  1. Cella L, Contieri R, Paciotti M, et al. Artificial intelligence for predicting BCG response in non-muscle-invasive bladder cancer: a systematic review. BJU Int. 2026. PMID: 42605566. doi:10.1111/bju.70404

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。