Abstract SY14-01: AI inferred spatial transcriptomics unlocks cancer biomarker discovery from histopathology

Abstract Path2Space: AI histopathology-inferred spatial transcriptomics as a driver of next-generation fast and cost effective precision oncology. I will start briefly reviewing my lab’s decade long work on developing a synthetic lethality approach for predicting patients’ response to therapy from tumor transcriptomics and whole slide pathology images [Cell 2021, MED 2023, Nature Cancer 2024]. The main part of my talk will focus on Path2Space, a new generic unpublished deep learning approach that predicts spatial gene expression directly from histopathology slides. Applied to study breast cancer as an example, Path2Space robustly predicts the spatial gene expression of more than 4000 genes and accurately infers cell-type abundances in the tumor microenvironment (TME) based on the inferred ST data. Path2Space characterizes thee TME of more than a 1000 breast cancer TCGA samples on an unprecedented scale and identifies three new spatially-grounded breast cancer patients subgroups with distinct survival rates. Closing a decade long circle, Path2Space- enables more accurate predictions of patients’ response to chemotherapy and trastuzumab directly from H&E slides than those obtained by existing established sequencing-based transcriptomics biomarkers. Path2Space thus offers a transformative approach to robustly delineate the tumors microenvironment directly from their histopathology slides. It facilitates the development of a new type of spatially-grounded biomarkers and lays the basis for a new generation of fast and low cost precision oncology biomarkers. Citation Format: Eytan Ruppin. AI inferred spatial transcriptomics unlocks cancer biomarker discovery from histopathology [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr SY14-01.