MGP_HNSC32: A H&E Morphology-Derived 32-Gene Prognostic Signature in Head and Neck Squamous Cell Carcinoma
In brief
In 701 patients, an H&E-derived score marked 40% higher progression hazard
A 32-gene score inferred from routine H&E slides separated high- and low-risk patients in an independent head and neck cancer cohort, with the high-risk group showing a 40% higher progression hazard at the prespecified cutoff. The continuous score was not significantly associated with progression, so this proof-of-concept needs further validation before clinical use.
- Journal
- Cancer medicine (Q1)
- Published
- 1 November 2026
- Study design
- Prospective / inception cohort
- Evidence level
- Level 2, Moderate (CEBM 2b)
- Authors
- Di Zhao, Zhiheng Huang, Jing Xu, Zheng Zhang, Jochen Hess, Bohai Feng
- PMID
- 42928519
- DOI
- 10.1002/cam4.72427
Why clinicians should know about it
- Picked for Histology (top studies of the week, 1 November 2026): Connects H&E morphology with tissue features and gene-expression predictions
Abstract
Head and neck squamous cell carcinoma (HNSC) exhibits substantial molecular and morphological heterogeneity, yet current morphology-based prognostic models often lack biological interpretability. We developed a transformer-based computational framework (CTransPath) to virtually predict gene expression directly from hematoxylin and eosin (H&E) whole-slide images. Morphology-associated genes were identified from the TCGA-HNSC cohort, and a 32-gene morphology-guided prognostic signature (MGP_HNSC32) based on H&E-derived predicted gene expression was developed using Random Survival Forest and multivariable Cox regression modeling. In TCGA-HNSC, the MGP_HNSC32 risk score significantly stratified patients by progression-free interval (PFI; HR = 4.13, 95% CI: 2.84-6.02), overall survival (OS; HR = 2.42, 95% CI: 1.78-3.29), and disease-specific survival (DSS; HR = 4.88, 95% CI: 2.95-8.06; all p < 0.001), and remained independently associated with prognosis after multivariable adjustment. High-risk tumors were enriched for cell cycle, E2F, MYC, and epithelial-mesenchymal transition pathways and displayed poorly differentiated histology with nuclear atypia and stromal desmoplasia. In contrast, low-risk tumors showed keratinizing morphology and immune-rich stroma, corresponding to differentiation- and metabolism-related transcriptional programs. In the independent HANCOCK cohort (n = 701), risk scores calculated entirely from H&E-derived predicted gene expression using the TCGA-derived coefficients and prespecified cutoff were associated with progression-free survival (HR = 1.40, 95% CI: 1.03-1.90, p = 0.034). However, the continuous risk score was not significantly associated with progression-free survival (HR = 1.001, 95% CI: 0.999-1.003, p = 0.328). MGP_HNSC32 links histomorphology to transcriptional programs and demonstrates preliminary prognostic associations using routine H&E images. The H&E-based gene-expression prediction component should be considered proof-of-concept, and further independent and prospective validation is required before clinical implementation.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.