免疫治疗正在重塑肿瘤治疗格局，但其疗效仍存在显著的个体差异。大多数患者无法从治疗中获益，部分患者甚至会出现严重的免疫相关毒性反应。这些挑战的根源在于，抗肿瘤免疫应答并非由单一信号通路决定，而是源于肿瘤、免疫系统与宿主微环境之间复杂的多尺度动态相互作用。精准解析并预测这种系统性行为，是推动下一代免疫治疗发展的关键科学问题。课题组致力于通过临床驱动的人工智能与机制性系统建模，解析肿瘤-免疫系统的复杂互作网络。通过整合多层级数据资源——包括电子病历、常规实验室检查与医学影像（临床数据层），以及单细胞测序与空间组学等多组学数据（基础实验层）——构建能够同时实现疗效预测、机制解析与疗法优化的综合计算框架。课题组核心研究方向为计算免疫肿瘤学，并逐步拓展至自身免疫病等更广泛的免疫相关疾病领域。

具体而言，围绕"为何免疫治疗仅对部分患者有效？如何使更多患者获益？"这一核心科学问题，课题组开展三个相互关联的研究方向，形成从临床预测、机制解析到治疗设计的系统性研究体系：

1. 精准免疫治疗的预测性人工智能 —— 从数据整合到临床决策

课题组致力于开发面向真实临床场景的预测模型，在治疗启动前预判患者的治疗响应概率与毒性风险谱，为个体化临床决策提供循证依据。课题组强调精准医学的普惠化路径：相较于依赖昂贵专项检测的模型架构，优先利用常规临床数据（如常规血液学检测、电子病历、常规病理切片等）构建预测框架，让精准肿瘤学在资源受限的医疗场景中得以应用，推动医疗公平性与技术可及性。同时，坚持可解释的人工智能范式：通过多模态数据融合与算法创新，构建兼具预测精度与生物学可解释性的计算模型，从而增强模型的临床可信度与转化应用价值。

2. 系统免疫学解析耐药机制 —— 从相关性分析到因果性推断

仅预测疗效尚不足以充分指导治疗优化，对免疫治疗耐药背后生物学规律的深层理解同样不可或缺。课题组整合单细胞组学、空间组学与多尺度系统建模方法，系统解析肿瘤微环境的细胞异质性、状态可塑性及其空间组织结构特征。重点解析肿瘤免疫逃逸的分子与细胞机制，探索代谢应激（如肿瘤缺氧微环境）与细胞间通讯网络如何协同作用，驱动免疫抑制与耐药形成。通过构建肿瘤免疫微环境的时空动态图谱，模拟细胞状态转变轨迹及其空间组织规律，识别可干预的耐药节点，为开发新型免疫治疗策略奠定理论基础。

3. 肿瘤免疫数字孪生平台（eOnco-Immune）—— 从机制解析到理性设计

基于预测模型与机制研究的积累，课题组的长期目标是推动免疫治疗从经验性试错走向理性化工程化设计。课题组计划构建 肿瘤免疫数字孪生平台，通过整合患者特异性数据、免疫系统先验知识与多尺度计算模型，在计算环境中模拟肿瘤-免疫系统的时空动态演化，以及不同肿瘤对各类治疗策略的响应，从而支持虚拟临床试验与计算辅助的可编程免疫治疗设计，在最大化抗肿瘤效应的同时最小化毒副作用。通过这一整合性计算框架，课题组致力于将免疫治疗转化为可预测、可模拟、可精准设计的工程化医学体系。

^#, first author(s), ^*, corresponding author(s).

2026

1. Danh-Tai Hoang^#*, Tian-Gen Chang^#*, Cristina R. Ferrone, et al. AI-driven pathology and blood-based biomarkers: a golden opportunity to democratize precision oncology. Cancer Discovery, (in press)

2025

2. Tian-Gen Chang^*, Jay Friedman, Paul E. Clavijo et al. Hypoxia-Induced EGR1 Remodels Neutrophils to Suppress Anti-Tumor Immunity. Science Immunology, 2025; 10(114): eadz2273. [Link]

3. Xiaojuan Fan^#*, Tian-Gen Chang^#, Chuyun Chen et al. Analysis of RNA translation with a deep learning architecture provides new insight into translation control. Nucleic Acids Research, 2025; 53(7): gkaf277. [Link]

4. Tian-Gen Chang^*, Seongyong Park, Alejandro A. Schäffer et al. Hallmarks of artificial intelligence contributions to precision oncology. Nature Cancer, 2025; 6, 417–431. [Link]

5. Tian-Gen Chang^#, Aris Spathis^#, Alejandro A. Schäffer et al. Tumor and blood B cell abundance outperforms established immune checkpoint blockade response prediction signatures in head and neck cancer. Annals of Oncology, 2025; 36(3): 309-320. [Link]

·       Highlighted in Annals of Oncology Editorial: Is the abundance of B cells the best biomarker to predict immune checkpoint inhibitor response in head and neck squamous cell cancers? 2025; 36(3): 238-239 [Link]

6. Sanna Madan^#, Tian-Gen Chang^#, et al. Single-cell-guided identification of logic-gated antigen combinations for designing effective and safe CAR therapy. BioRxiv, 2025. [Link]

7. Chen Weller^#, Osnat Bartok^#, Christopher S. McGinnis^#, Heyilimu Palashati, Tian-Gen Chang, et al. Translation dysregulation in cancer as a source for targetable antigens. Cancer Cell, 2025; 43(5): 823-840. [Link]

2024

8. Yingying Cao^#, Tian-Gen Chang^#, Fiorella Schischlik^# et al. Inferring characteristics of the tumor immune microenvironment of patients with HNSCC from single-cell transcriptomics of peripheral blood. Cancer Research Communications, 2024; 4(9): 2335-2348. [Link]

9. Tian-Gen Chang^#, Yingying Cao^#, Hannah J. Sfreddo^# et al. LORIS robustly predicts patient outcomes with immune checkpoint blockade therapy using common clinical, pathologic and genomic features. Nature Cancer, 2024; 5: 1158–1175. [Link]

·       Highlighted in Nature Cancer “News & Views”: Enhanced precision in immunotherapy, 2024; 5: 1136–1138 [Link]

·       Featured in Clinical Chemistry: An AI Model (LORIS) to Predict Immune Checkpoint Blockade Response in Cancer: A Clinical Data Science Perspective, 2025; 71(3): 345–347 [Link]

·       Featured by NCI Center for Biomedical Informatics & Information Technology, NCI Media Advisory, NCI Podcast "Inside Cancer Careers", and Podcast “Cancer HealthCast”

10. Yingying Cao^#, Tian-Gen Chang^#, Sahil Sahni et al. Reusability report: Leveraging supervised learning to uncover phenotype-relevant biology from single-cell RNA sequencing data. Nature Machine Intelligence, 2024; 6: 307–314. [Link]

11. Marica Rosaria Ippolito^#, Johanna Zerbib^#, Yonatan Eliezer, Eli Reuveni, Sonia Viganò, Giuseppina De Feudis, Eldad D Shulman, Anouk Savir Kadmon, Rachel Slutsky, Tian-Gen Chang, et al. Increased RNA and protein degradation is required for counteracting transcriptional burden and proteotoxic stress in human aneuploid cells. Cancer Discovery, 2024; 14(12): 2532-2553. [Link]

12. Johanna Zerbib^#, Marica Rosaria Ippolito^#, Yonatan Eliezer, Giuseppina De Feudis, Eli Reuveni, Anouk Savir Kadmon, Sara Martin, Sonia Viganò, Gil Leor, James Berstler, Julia Muenzner, Michael Mülleder, Emma M Campagnolo, Eldad D Shulman, Tian-Gen Chang, et al. Human aneuploid cells depend on the RAF/MEK/ERK pathway for overcoming increased DNA damage. Nature Communications, 2024; 15(1): 7772. [Link]

13. Parth Desai^#, Nobuyuki Takahashi, Rajesh Kumar, Samantha Nichols, Justin Malin, Allison Hunt, Christopher Schultz, Yingying Cao, Desiree Tillo, Darryl Nousome, Lakshya Chauhan, Linda Sciuto, Kimberly Jordan, Vinodh Rajapakse, Mayank Tandon, Delphine Lissa, Yang Zhang, Suresh Kumar, Lorinc Pongor, Abhay Singh, Brett Schroder, Ajit Kumar Sharma, Tian-Gen Chang, et al. Microenvironment shapes small-cell lung cancer neuroendocrine states and presents therapeutic opportunities. Cell Reports Medicine, 2024; 5(6): 101610. [Link]

2023

14. Tian-Gen Chang^*, Yingying Cao, Eldad D. Shulman, et al. Optimizing cancer immunotherapy response prediction by tumor aneuploidy score and fraction of copy number alterations. npj Precision Oncology, 2023; 7: 54. [Link]

15. Sanna Madan^#, Sanju Sinha, Tian-Gen Chang, et al. Pan-cancer analysis of patient tumor single-cell transcriptomes identifies promising selective and safe chimeric antigen receptor targets in head and neck cancer. Cancers, 2023; 15(19):4885. [Link]