Revealing intra-group immunotherapy response heterogeneity in metastatic urothelial carcinoma through interpretable feature extraction and spectral clustering
Our framework, which combines SHAP-based interpretable feature extraction with spectral clustering, revealed subclass-level heterogeneity in ICI response, highlighting biologically distinct immune subclasses. This approach may facilitate the development of subclass-specific therapeutic strategies.
Authors
- Tianyi Shi (ORCID: https://orcid.org/0000-0002-9204-9949)
- Bryan J. Mathis (ORCID: https://orcid.org/0000-0002-9879-8289)
- Yoshiyuki Nagumo (ORCID: https://orcid.org/0000-0003-0332-4791)
- Tetsuya Sakurai (ORCID: https://orcid.org/0000-0002-5789-7547)
- Xiucai Ye
Institutions
- University of Tsukuba (JP)
- Tsukuba University of Technology (JP)
- Cyberdyne (Japan) (JP)
- University of Tsukuba Hospital (JP)
- Institute of Applied Biochemistry (JP)
Publication Details
- Journal
- Frontiers in Immunology
- Published
- 2026-01-06
- DOI
- https://doi.org/10.3389/fimmu.2025.1629001
- Primary Topic
- Bladder and Urothelial Cancer Treatments
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Japan Agency for Medical Research and Development
- University of Tsukuba
- Japan Society for the Promotion of Science