Transcriptomic Models for Immunotherapy Response Prediction Show Limited Cross-cohort Generalisability
Organizations: UNSW BioMedical Machine Learning Lab (BML), School of Biomedical Engineering, UNSW Sydney, Sydney, NSW, 2052, Australia · Garvan Institute of Meidical Research, Sydney, NSW, 2000, Australia · School of Biomedical Engineering, UNSW, Sydney, NSW, 2052, Australia · Department of Computer Science, Zhejiang University, Hangzhou, Zhejiang, 310027, China · Institute for Intelligent Systems Research and Innovation, Deakin University, Geelong, Victoria, Australia · School of Engineering & Technology, Duy Tan University, Da Nang, Vietnam · Shenzhen Key Laboratory for High Performance Data Mining, Shenzhen Institute of Advanced Technology, Shenzhen, China · Center of Excellence in Precision Medicine and Digital Health, Faculty of Dentistry, Chulalongkorn University, Bangkok, Thailand · Shanghai Institute of Immunology, Shanghai Jiao Tong University School of Medicine, Shanghai 200025, China · Visiting Scholar (Collaborative Projects), Center of Excellence in Precision Medicine and Digital Health, Chulalongkorn University, Bangkok, Thailand
Abstract
Immune checkpoint inhibitors (ICIs) have transformed cancer therapy; yet substantial proportion of patients exhibit intrinsic or acquired resistance, making accurate pre-treatment response prediction a critical unmet need. Transcriptomics-based biomarkers derived from bulk and single-cell RNA sequencing (scRNA-seq) offer a promising avenue for capturing tumour-immune interactions, yet the cross-cohort generalisability of existing prediction models remains unclear.We systematically benchmark nine state-of-the-art transcriptomic ICI response predictors, five bulk RNA-seq-based models (COMPASS, IRNet, NetBio, IKCScore, and TNBC-ICI) and four scRNA-seq-based models (PRECISE, DeepGeneX, Tres and scCURE), using publicly available independent datasets unseen during model development. Overall, predictive performance was modest: bulk RNA-seq models performed at or near chance level across most cohorts, while scRNA-seq models showed only marginal improvements. Pathway-level analyses revealed sparse and inconsistent biomarker signals across models. Although scRNA-seq-based predictors converged on immune-related programs such as allograft rejection, bulk RNA-seq-based models exhibited little reproducible overlap. PRECISE and NetBio identified the most coherent immune-related themes, whereas IRNet predominantly captured metabolic pathways weakly aligned with ICI biology. Together, these findings demonstrate the limited cross-cohort robustness and biological consistency of current transcriptomic ICI prediction models, underscoring the need for improved domain adaptation, standardised preprocessing, and biologically grounded model design.