Token-guided multimodal prognosis in hepatocellular carcinoma: a framework steered by tumour–stroma ratio
Menée à l'aide de données d'images de lames histologiques ainsi que de données cliniques portant sur 392 patients atteints d'un carcinome hépatocellulaire et menée à l'aide de données du projet "The Cancer Genome Atlas" portant sur 168 patients supplémentaires, cette étude met en évidence une relation non linéaire entre le ratio tumeur-stroma et la mortalité spécifique puis propose un nouveau modèle d'intelligence artificielle capable de quantifier précisément ce ratio
Background : The tumour–stroma ratio (TSR) is a potential prognostic indicator, yet hindered by quantification challenges and conflicting reports.
Objective : To determine whether TSR follows a non-linear prognostic pattern and to develop an artificial intelligence (AI)-powered framework for standardised TSR assessment and prognosis prediction in hepatocellular carcinoma (HCC).
Design : We integrated whole-slide image (WSI) data with clinical variables across a retrospective cohort (n=392) and The Cancer Genome Atlas dataset (n=168). Restricted cubic splines were used to interrogate non-linear hazard dynamics, with biological validation via transcriptomics and immunohistochemistry. An AI-driven foundation model framework was developed for TSR quantification and multimodal prognostic modelling.
Results : Our analysis unveiled an inverted U-shaped non-linear relationship between TSR and mortality, identifying a risk initiation threshold at 0.188 and a peak at 0.268. Transcriptomics analysis indicated that this high-risk phenotype is characterised by active tumour proliferation, stromal activation and tumour microenvironment crosstalk. Technically, AI-derived TSR showed strong correlation with expert assessment (R² >0.9). Furthermore, we developed a novel ‘Token-Guided Multimodal Fusion’ architecture to integrate WSI, TSR and clinical variables as high-dimensional tokens directly into the computational logic. Consequently, our multimodal framework demonstrated prognostic accuracy (area under the curve >0.80) compared with unimodal baselines.
Conclusion : This study redefines TSR assessment, shifting from manual estimation to high-dimensional semantic reasoning. By identifying the non-linear prognostic mechanics of the stroma, our token-guided framework offers a biologically interpretable solution for HCC. We suggest that the future of computational pathology may lie not in simple quantification, but in the semantic fusion of human domain knowledge with AI reasoning.
Gut , article en libre accès, 2026