Interpretable multi-modal artificial intelligence model for predicting gastric cancer response to neoadjuvant chemotherapy
Neoadjuvant chemotherapy assessment is imperative for prognostication and clinical management of locally advanced gastric cancer. We propose an incremental supervised contrastive learning model (iSCLM), an interpretable artificial intelligence framework integrating pretreatment CT scans and H&E-stained biopsy images, for improved decision-making regarding neoadjuvant chemotherapy. We have constructed and tested iSCLM using retrospective data from 2,387 patients across 10 medical centers and evaluated its discriminative ability in a prospective cohort (132 patients; ChiCTR2300068917). iSCLM achieves areas under receiver operating characteristic curves of 0.846-0.876 across different test cohorts. Computed tomography (CT) and pathological attention heatmaps from Shapley additive explanations and global sort pooling illustrate additional benefits for capturing morphological features through supervised contrastive learning. Specifically, pathological top-ranked tiles exhibit decreased distances to tumor-invasive borders and increased inflammatory cell infiltration in responders compared with non-responders. Moreover, CD11c expression is elevated in responders. The developed interpretable model at the molecular pathology level accurately predicts chemotherapy efficacy.
Authors
- Dinggang G. Shen (ORCID: https://orcid.org/0000-0002-7934-5698)
- Z. Wang (ORCID: https://orcid.org/0009-0004-2587-9653)
- Tao Xu (ORCID: https://orcid.org/0000-0001-8782-5010)
- Jingao Xu (ORCID: https://orcid.org/0000-0002-8347-2657)
- Qiong Xiao (ORCID: https://orcid.org/0000-0001-9470-3927)
- Yi Ba
- Zitong Wang (ORCID: https://orcid.org/0009-0009-2782-5858)
- Peng Gao (ORCID: https://orcid.org/0000-0002-6586-4307)
- Lijie He (ORCID: https://orcid.org/0000-0001-7559-6249)
- Guan Wang (ORCID: https://orcid.org/0000-0003-1973-2044)
- Jing Yi (ORCID: https://orcid.org/0000-0002-8554-1855)
- Ting Deng (ORCID: https://orcid.org/0000-0003-4552-8039)
- Xianchun Gao (ORCID: https://orcid.org/0000-0001-6162-2434)
- Yu Fu (ORCID: https://orcid.org/0000-0002-2413-3858)
- Yingying Feng (ORCID: https://orcid.org/0000-0002-9854-692X)
- Yong Zhang (ORCID: https://orcid.org/0000-0001-6650-6790)
- Erdong Yao (ORCID: https://orcid.org/0000-0001-5466-1347)
- Jiangdian Song (ORCID: https://orcid.org/0000-0002-3355-9930)
- Ya Long (ORCID: https://orcid.org/0000-0001-8358-7772)
- Xue Jiang (ORCID: https://orcid.org/0000-0001-9977-3916)
- Liying Zhang (ORCID: https://orcid.org/0000-0003-4517-0751)
- Yingjie Zhang (ORCID: https://orcid.org/0009-0000-7284-2424)
- Xiaoyan Li (ORCID: https://orcid.org/0000-0002-1144-0648)
- Yan Zhao (ORCID: https://orcid.org/0000-0002-7760-916X)
- Jinghua Sun (ORCID: https://orcid.org/0009-0009-5600-8849)
- Minwen Zheng (ORCID: https://orcid.org/0000-0002-0862-3208)
- Miao Yuan
- Hanyu Chen (ORCID: https://orcid.org/0000-0002-5648-9939)
- Hui Juan Jennifer Tan
- Yanhong Deng (ORCID: https://orcid.org/0000-0002-4333-468X)
- Junhua Zhao
- Yuchong Yang
- Jipeng Mei
- Zhicheng Wang (ORCID: https://orcid.org/0000-0002-2232-4298)
- Bin Wang (ORCID: https://orcid.org/0000-0001-7665-6290)
Institutions
- Army Medical University (CN)
- Sun Yat-sen University (CN)
- Dalian Medical University (CN)
- Chinese Academy of Medical Sciences & Peking Union Medical College (CN)
- Peking Union Medical College Hospital (CN)
- Tianjin Medical University Cancer Institute and Hospital (CN)
- National Clinical Research Center for Digestive Diseases (CN)
- Sixth Affiliated Hospital of Sun Yat-sen University (CN)
- Second Affiliated Hospital of Dalian Medical University (CN)
- Liaoning Provincial People's Hospital (CN)
- Shenyang Fifth People Hospital (CN)
- Southwest Hospital (CN)
- Neusoft (China) (CN)
- United Imaging Healthcare (China) (CN)
- First Hospital of China Medical University (CN)
- First Affiliated Hospital of Zhengzhou University (CN)
- Xijing Hospital (CN)
- Liaoning Cancer Hospital & Institute (CN)
- China Medical University (CN)
- Northeastern University (CN)
- Air Force Medical University (CN)
Publication Details
- Journal
- Cell Reports Medicine
- Published
- 2024-12-01
- DOI
- https://doi.org/10.1016/j.xcrm.2024.101848
- Citations
- 21
- Primary Topic
- Gastric Cancer Management and Outcomes
- Type
- article
- Field-Weighted Citation Impact
- 8.80
Funders
- National Natural Science Foundation of China
- China Medical University
- Peking Union Medical College Hospital
- National Science and Technology Major Project
- National Key Research and Development Program of China