- 从 H&E 全切片图像直接预测 857 个基因的表达；最佳模型为两种病理基础模型特征的双重集成。
- 776/857 个基因可预测，平均相关系数 0.43（交叉验证）/0.45（留出集）/0.40（外部 Frankfurt 队列）；预测表达与 NECTIN4 等临床相关标志物高度相关。
- AI 预测的基底型/管腔型分型与总生存相关，与真实分型一致（准确率 >0.8）。
收录范围：PubMed 英文摘要及中文翻译（abstract-only）。原文为订阅制（Unpaywall：closed），出版社页对本站仅显示摘要，未采集全文、图表。
摘要
肌层浸润性膀胱癌（MIBC）在临床结局和治疗反应方面存在显著异质性。尽管基因表达谱分析已提供了重要的生物学和预后见解，但部分由于成本原因，转录组检测尚未常规纳入标准临床实践。已有研究证明，AI 模型可从 H&E 染色全切片图像（WSI）预测基因表达，有望为患者分层和生物标志物发现提供一种可规模化的转录组信息获取途径。我们开发并测试了直接从 H&E 染色 WSI 预测 857 个基因表达的模型。表现最佳的最终模型为双重集成模型，整合了基于不同基础模型提取特征所训练模型的预测结果。该双重集成模型在 857 个基因中有 776 个可预测，平均相关系数为 0.43（交叉验证）、0.45（留出集）和 0.40（外部 Frankfurt 队列）。预测表达与具有临床意义的标志物（如 NECTIN4）的表达高度相关。预测表达估计的风险比与实测表达的风险比高度相关，并在两个队列中均识别出与总生存相关的基因（无论基于预测表达还是实测表达）。AI 预测的基底型与管腔型亚型与总生存相关，并与真实分型一致（准确率 >0.8）。研究结果表明，从常规染色推断的基因表达可支持 MIBC 的生物标志物发现与患者分层。
- Expression of 857 genes predicted directly from H&E whole-slide images; the best model was a dual ensemble of two pathology foundation-model feature sets.
- 776/857 genes predictable, mean correlation 0.43 (cross-validation) / 0.45 (holdout) / 0.40 (external Frankfurt cohort); predicted expression correlated strongly with clinically relevant markers such as NECTIN4.
- AI-predicted basal vs luminal subtypes were associated with overall survival and matched ground truth (accuracy >0.8).
Scope: PubMed abstract only. The article is subscription-only (Unpaywall: closed); the publisher page showed only the abstract to this site, so full text, figures and tables were not collected.
Abstract
Muscle-invasive bladder cancer (MIBC) exhibits significant heterogeneity in clinical outcomes and treatment responses. While gene expression profiling has provided important biological and prognostic insight, transcriptomic assays are not routinely incorporated into standard clinical practice, in part due to cost. AI models have been demonstrated to predict gene expression from H&E-stain whole-slide images (WSIs) potentially offering a scalable approach for gathering transcriptomic information for patient stratification and biomarker discovery. We developed and tested models to predict expression of 857 genes directly from H&E-stained whole-slide images (WSI). The best performing final model was a dual-ensemble combining predictions from models trained on features extracted with different foundation models. The dual-ensemble reached 776/857 predictable genes with an average correlation of 0.43 (cross-validation), 0.45 (holdout) and 0.40 (external Frankfurt cohort). Predicted expression strongly correlates with expression of clinically relevant markers (like NECTIN4). Estimated hazard ratios of predicted expression correlated strongly with those of measured expression and genes associated with overall survival for both predicted and measured expression in both cohorts were identified. AI-predicted basal vs. luminal subtypes were associated with overall survival and aligned with the ground-truth (accuracies > 0.8). Findings demonstrate that gene expressions inferred from routine staining can support biomarker discovery and patient stratification in MIBC.