Multimodal deep learning for cancer prognosis is commonly assumed to benefit from synergistic cross-modal interactions, yet this assumption has not been directly tested in survival prediction settings. This work adapts InterSHAP, a Shapley interaction index-based metric, from classification to Cox proportional hazards models and applies it to quantify cross-modal interactions in glioma survival prediction. Using TCGA-GBM and TCGA-LGG data (n=575), we evaluate four fusion architectures combining whole-slide image (WSI) and RNA-seq features. Our central finding is an inverse relationship between predictive performance and measured interaction: architectures achieving superior discrimination (C-index 0.64→0.82) exhibit equivalent or lower cross-modal interaction (4.8\%→3.0\%). Variance decomposition reveals stable additive contributions across all architectures (WSI≈40\%, RNA≈55\%, Interaction≈4\%), indicating that performance gains arise from complementary signal aggregation rather than learned synergy. These findings provide a practical model auditing tool for comparing fusion strategies, reframe the role of architectural complexity in multimodal fusion, and have implications for privacy-preserving federated deployment.
@article{arxiv.2603.29977,
title = {Quantifying Cross-Modal Interactions in Multimodal Glioma Survival Prediction via InterSHAP: Evidence for Additive Signal Integration},
author = {Iain Swift and JingHua Ye and Ruairi O'Reilly},
journal= {arXiv preprint arXiv:2603.29977},
year = {2026}
}