Yuan Wei, Chen Yijiang, Zhu Biyue, Yang Sen, Zhang Jiayu, Mao Ning, Xiang Jinxi, Li Yuchen, Ji Yuanfeng, Luo Xiangde, Zhang Kangning, Xing Xiaohan, Kang Shuo, Xiao Dongyuan, Wang Fang, Wu Jinkun, Zhang Haiyan, Tang Hongping, Maurya Himanshu, Corredor German, Barrera Cristian, Zhou Yufei, Pandav Krunal, Zhao Junhan, Jain Prantesh, Delasos Luke, Huang Junzhou, Yang Kailin, Teknos Theodoros N, Lewis James, Koyfman Shlomo, Pennell Nathan A, Yu Kun-Hsing, Han Xiao, Zhang Jing, Wang Xiyue, Madabhushi Anant
College of Biomedical Engineering, Sichuan University, Chengdu, Sichuan, China.
Department of Radiation Oncology, Stanford University School of Medicine, Palo Alto, CA, USA.
Signal Transduct Target Ther. 2025 Sep 3;10(1):285. doi: 10.1038/s41392-025-02374-w.
Accurate prognosis prediction is essential for guiding cancer treatment and improving patient outcomes. While recent studies have demonstrated the potential of histopathological images in survival analysis, existing models are typically developed in a cancer-specific manner, lack extensive external validation, and often rely on molecular data that are not routinely available in clinical practice. To address these limitations, we present PROGPATH, a unified model capable of integrating histopathological image features with routinely collected clinical variables to achieve pancancer prognosis prediction. PROGPATH employs a weakly supervised deep learning architecture built upon the foundation model for image encoding. Morphological features are aggregated through an attention-guided multiple instance learning module and fused with clinical information via a cross-attention transformer. A router-based classification strategy further refines the prediction performance. PROGPATH was trained on 7999 whole-slide images (WSIs) from 6,670 patients across 15 cancer types, and extensively validated on 17 external cohorts with a total of 7374 WSIs from 4441 patients, covering 12 cancer types from 8 consortia and institutions across three continents. PROGPATH achieved consistently superior performance compared with state-of-the-art multimodal prognosis prediction models. It demonstrated strong generalizability across cancer types and robustness in stratified subgroups, including early- and advanced-stage patients, treatment cohorts (radiotherapy and pharmaceutical therapy), and biomarker-defined subsets. We further provide model interpretability by identifying pathological patterns critical to PROGPATH's risk predictions, such as the degree of cell differentiation and extent of necrosis. Together, these results highlight the potential of PROGPATH to support pancancer outcome prediction and inform personalized cancer management strategies.
准确的预后预测对于指导癌症治疗和改善患者预后至关重要。虽然最近的研究已经证明了组织病理学图像在生存分析中的潜力，但现有的模型通常是以癌症特异性的方式开发的，缺乏广泛的外部验证，并且常常依赖于临床实践中不常用的分子数据。为了解决这些局限性，我们提出了PROGPATH，这是一个统一的模型，能够将组织病理学图像特征与常规收集的临床变量相结合，以实现泛癌预后预测。PROGPATH采用了一种基于图像编码基础模型构建的弱监督深度学习架构。形态学特征通过注意力引导的多实例学习模块进行聚合，并通过交叉注意力变换器与临床信息融合。基于路由器的分类策略进一步优化了预测性能。PROGPATH在来自15种癌症类型的6670名患者的7999张全切片图像（WSIs）上进行了训练，并在17个外部队列中进行了广泛验证，这些队列共有来自4441名患者的7374张WSIs，涵盖了来自三大洲8个联盟和机构的12种癌症类型。与最先进的多模态预后预测模型相比，PROGPATH始终表现出卓越的性能。它在不同癌症类型中表现出强大的通用性，在分层亚组中具有稳健性，包括早期和晚期患者、治疗队列（放疗和药物治疗）以及生物标志物定义的亚组。我们通过识别对PROGPATH风险预测至关重要的病理模式，如细胞分化程度和坏死程度，进一步提供了模型的可解释性。总之，这些结果凸显了PROGPATH在支持泛癌预后预测和为个性化癌症管理策略提供信息方面的潜力。
