Data-Efficient Multimodal Alignment for Histopathology-based Molecular Prediction
For computational pathology researchers, this provides a practical framework to infer molecular features from widely available H&E slides without costly sequencing, though it is incremental as it builds on existing foundation models and contrastive learning.
This work introduces a data-efficient method for predicting molecular pathway activity from H&E-stained whole-slide images by aligning frozen histopathology and RNA-seq foundation models via contrastive learning. On a multi-cancer cohort (N=1,720), it achieves a 25-fold improvement in retrieval over baselines and validates clinical utility on the POSEIDON trial, predicting immune and subtype signatures with R²>0.5 for morphologically grounded programs.
H&E-stained whole-slide images offer cohort-scale availability and rich spatial context but lack molecular specificity, whereas bulk RNA-seq provides transcriptome-wide resolution at high cost with limited archival availability. We show that training a lightweight alignment module atop frozen histopathology and RNA-Seq foundation models enables open-vocabulary molecular prompting -- querying H&E slides with gene-set signatures to predict pathway activity without sequencing or end-to-end retraining. Using contrastive learning on a multi-cancer cohort (N=1,720), we achieve a 25-fold improvement in retrieval over baseline methods. Systematic analysis reveals a graduated predictability spectrum: morphologically grounded programs (cell-cycle programs, immune-related) are most reliably predicted (R^2>0.5), while predicting pathways with no morphological footprint remains challenging as expected. We validate clinical utility on the POSEIDON clinical trial: H&E-predicted squamous cell carcinoma scores recapitulate NSCLC subtype identity and predicted IFN-gamma mirror PD-L1 tumor-cell expression groups. Furthermore, genesets describing immune activation and fibrosis predict known tumor microenvironment archetypes from histology alone. We further validate generalization of our approach across unseen cohorts and demonstrate data-efficient domain adaptation, establishing a slide-native framework for molecular analysis on H&E images.