Innani Shubham, Bell W Robert, Harmsen Hannah, Nasrallah MacLean P, Baheti Bhakti, Bakas Spyridon
Department of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA.
Indiana University Melvin and Bren Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.
Neurooncol Adv. 2025 Jul 11;7(1):vdaf140. doi: 10.1093/noajnl/vdaf140. eCollection 2025 Jan-Dec.
Isocitrate dehydrogenase (IDH) mutation status is a diagnostic requirement for glioma with associated prognostic and therapeutic implications. Clinical routine visual assessment of tissue is insufficient to determine IDH status conclusively, mandating molecular workup that is unavailable everywhere.
We developed an interpretable Artificial Intelligence (AI)-based approach for determining IDH status directly from H&E-stained glioma slides. Our study is based on 2442 multi-institutional whole slide images (WSIs) from 3 independent retrospective glioma collections, following their reclassification according to the WHO 2021 criteria: (1) TCGA-GBM/TCGA-LGG ( = 1534, = 799), (2) University of Pennsylvania Health System collection (UPHS, = = 114), and (3) EBRAINS ( = = 794). Method development is based on TCGA, whereas UPHS and EBRAINS are independent hold-out datasets for model validation. Six pathology-specific foundation AI models and an ImageNet-pretrained AI model facilitate robust feature extraction. Features are aggregated into slide-level representations via an interpretable multiple-instance learning (MIL) mechanism to differentiate IDH-wildtype from IDH-mutant cases and generate attention heatmaps correlating with identifiable morphologic characteristics.
Our approach yields AUC = 0.96 over a 10-fold cross-validation schema and generalizable performance on independent validation (AUC = 0.97, AUC = 0.95). Interpretability analysis reveals high attention regions in IDH-wildtype tumors exhibiting significant pleomorphism and microvascular proliferation, while IDH-mutant tumors show dense nodular cell concentrations, microcysts, and gemistocytic cells.
Accurate H&E-based determination of glioma IDH mutation status can expedite conclusive diagnosis and clinical decision-making and even facilitate it in underserved regions. Finally, interpretability analysis of distilled human-identifiable features can further improve our understanding of the disease.
异柠檬酸脱氢酶（IDH）突变状态是胶质瘤诊断的必要条件，具有相关的预后和治疗意义。临床常规对组织的视觉评估不足以最终确定IDH状态，因此需要进行分子检查，但这并非在所有地方都可行。
我们开发了一种基于人工智能（AI）的可解释方法，可直接从苏木精-伊红（H&E）染色的胶质瘤切片中确定IDH状态。我们的研究基于来自3个独立回顾性胶质瘤数据集的2442张多机构全切片图像（WSIs），这些图像根据世界卫生组织2021年标准重新分类：（1）癌症基因组图谱-胶质母细胞瘤/癌症基因组图谱-低级别胶质瘤（TCGA-GBM/TCGA-LGG，n = 1534，n = 799），（2）宾夕法尼亚大学医疗系统数据集（UPHS，n = n = 114），以及（3）EBRAINS（n = n = 794）。方法开发基于TCGA，而UPHS和EBRAINS是用于模型验证的独立保留数据集。六个病理学特定的基础AI模型和一个在ImageNet上预训练的AI模型有助于进行强大的特征提取。通过可解释的多实例学习（MIL）机制将特征聚合为玻片级表示，以区分IDH野生型和IDH突变型病例，并生成与可识别形态特征相关的注意力热图。
我们的方法在10倍交叉验证模式下的曲线下面积（AUC）为0.96，在独立验证中具有可推广的性能（AUC = 0.97，AUC = 0.95）。可解释性分析显示，IDH野生型肿瘤中的高注意力区域表现出明显的多形性和微血管增殖，而IDH突变型肿瘤则显示出密集的结节状细胞聚集、微囊肿和肥胖型星形细胞。
基于H&E准确确定胶质瘤IDH突变状态可以加快最终诊断和临床决策，甚至在医疗服务不足的地区也能提供便利。最后，对提炼出的人类可识别特征进行可解释性分析可以进一步加深我们对该疾病的理解。
