PAST: A multimodal single-cell foundation model for histopathology and spatial transcriptomics in cancer

TL;DR

PAST is a novel multimodal foundation model that integrates histopathology images and single-cell transcriptomics to enhance cancer analysis, enabling accurate predictions and insights at cellular resolution.

Contribution

It introduces PAST, the first pan-cancer single-cell foundation model trained on extensive paired data, unifying cellular morphology and gene expression for advanced cancer research.

Findings

Outperforms existing methods across multiple cancer types.

Accurately predicts gene expression and molecular features from pathology images.

Demonstrates robustness and scalability in diverse applications.

Abstract

While pathology foundation models have transformed cancer image analysis, they often lack integration with molecular data at single-cell resolution, limiting their utility for precision oncology. Here, we present PAST, a pan-cancer single-cell foundation model trained on 20 million paired histopathology images and single-cell transcriptomes spanning multiple tumor types and tissue contexts. By jointly encoding cellular morphology and gene expression, PAST learns unified cross-modal representations that capture both spatial and molecular heterogeneity at the cellular level. This approach enables accurate prediction of single-cell gene expression, virtual molecular staining, and multimodal survival analysis directly from routine pathology slides. Across diverse cancers and downstream tasks, PAST consistently exceeds the performance of existing approaches, demonstrating robust generalizability and scalability. Our work establishes a new paradigm for pathology foundation models, providing a versatile tool for high-resolution spatial omics, mechanistic discovery, and precision cancer research.