Mucin phenotype-based deep learning framework for intestinal metaplasia-carcinogenesis progression prediction
Xiaoyang Wu1,2, Fang Wang1,3, Weiyou Dai4
1Tumor Etiology and Screening Department of Cancer Institute, and Key Laboratory of Cancer Etiology and Prevention in Liaoning Education Department, the First Hospital of China Medical University, Shenyang, Liaoning, China.
Abstract:
This study decodes spatiotemporal mucin dynamics in gastric carcinogenesis, revealing gastric-type markers (MUC5AC/MUC6) decline progressively while intestinal-type markers (MUC2/CD10) peak in gastric intestinal metaplasia (GIM) before decreasing in gastric cancer (GC). We developed MPMR, a dual-function UNI-pretrained Vision Transformer (ViT) model, which directly predicts four mucin markers from H&E whole-slide images with near-perfect accuracy (AUC: 0.921-0.997) and generates interpretable simulated staining heatmaps via adversarial learning. Integrating these outputs with clinical variables, the MPMR-IMCP risk model significantly outperformed clinical-only models (ΔAUC = 0.050), enabling both phenotype analysis and GIM risk stratification without specialized staining. Validated in longitudinal cohorts from Chinese GC high-incidence regions, this framework offers an efficient solution for monitoring GIM malignant transformation.
