Contrastive multimodal deep learning for survival prediction in grade 2/3 gliomas
JNCI Cancer Spectrum · 09.05.2026 · 10 (3)
DOI: 10.1093/jncics/pkag039 ↗Kurz gefasst
Fachartikel in der Zeitschrift JNCI Cancer Spectrum, online erschienen am 15.04.2026. Verfasst von Peiying Hua und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Data Science, Geisel School of Medicine at Dartmouth, Hanover, NH, United States. Hauptthemen laut MEDLINE: Glioma, Deep Learning, Brain Neoplasms. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIGMS NIH HHS, NLM NIH HHS, US National Library of Medicine.
Abstract (Originalsprache)
Background
Accurate survival prediction for grade 2/3 glioma patients remains challenging due to tumor biological heterogeneity and limitations of current prognostic methods that rely on single-modality data.
Methods
We developed a multimodal deep learning framework integrating histopathology whole-slide images, somatic mutations, and clinical-demographic data. A 3-stage training pipeline combined contrastive learning with survival-specific optimization to align cross-modal representations. The framework was trained on 498 grade 2/3 glioma patients from The Cancer Genome Atlas and evaluated using 5-fold cross-validation and an independent Dartmouth-Hitchcock Medical Center (DHMC) cohort (n = 61).
Results
The contrastive multimodal model achieved a c-index of 0.91 (95% confidence interval [CI] = 0.84 to 0.96), significantly outperforming the unimodal models (image-only = 0.76; non-image-only = 0.87) and showing an improvement over the noncontrastive multimodal model (c-index = 0.89), although this difference was not statistically significant. Kaplan-Meier analysis demonstrated clear survival separation across risk strata (log-rank P = 4.4 × 10−5). Contrastive learning improved representation clustering quality, with silhouette scores increasing from 0.20 to 0.24 (P = .05). External evaluation on the DHMC cohort achieved a c-index of 0.87 (95% CI = 0.77 to 0.95) after domain adaptation.
Conclusion
Contrastive multimodal learning significantly enhances survival prediction in grade 2/3 gliomas by effectively integrating histopathology, genomics, and clinical data. This annotation-free approach enables early risk stratification using routinely collected data and shows promise for informing personalized treatment decisions and clinical trial stratification.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Biomedical Data Science, Geisel School of Medicine at Dartmouth, Hanover, NH, United States.
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, United States.
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, United States.
- Department of Computer Science, Dartmouth College, Hanover, NH, United States.
ORCID-Profile: Peiying Hua, Chun-Chieh Lin, Saeed Hassanpour
Förderung
- NIGMS NIH HHS
- NLM NIH HHS
- US National Library of Medicine
Zitieren
APA
Hua, P., Lin, C. C., Fenlon, T., Liu, X., & Hassanpour, S. (2026). Contrastive multimodal deep learning for survival prediction in grade 2/3 gliomas. JNCI Cancer Spectrum, 10(3). https://doi.org/10.1093/jncics/pkag039
Vancouver
Hua P, Lin CC, Fenlon T, Liu X, Hassanpour S. Contrastive multimodal deep learning for survival prediction in grade 2/3 gliomas. JNCI Cancer Spectrum. 2026;10(3). doi:10.1093/jncics/pkag039
BibTeX
@article{Hua2026,
author = {Hua, Peiying and Lin, Chun-Chieh and Fenlon, Travis and Liu, Xiaoying and Hassanpour, Saeed},
title = {Contrastive multimodal deep learning for survival prediction in grade 2/3 gliomas},
journal = {JNCI Cancer Spectrum},
year = {2026},
volume = {10},
number = {3},
doi = {10.1093/jncics/pkag039}
}
EUVIMED