Abstract 6310: Bridging spatial proteomics and classical histopathology: advanced spatial analysis through H&E-enhanced multiplexed immunoprofiling

Background: Multiplex immunofluorescence (mIF) is increasingly adopted in translational research due to its ability to provide rich spatial and molecular information. Nevertheless, H&E staining remains the gold standard for clinical diagnosis, revealing essential histological context and morphological features not visible in mIF images. While H&E is routinely performed on serial sections in analytical mIF workflows, biological variability between sections limits its utility. Co-registration of H&E and mIF images from the same slide enables precise colocalization of cells across channels, combining the histological detail of H&E with the spatial insights of mIF to enhance both biological assessment and image analysis. Overview: This study explored the integration of same-section H&E staining and imaging into mIF workflows using formalin-fixed paraffin-embedded (FFPE) tissue samples from mixed cancer indications. After mIF staining and imaging with InSituPlex™ technology, same-section H&E images were co-registered with the mIF images before incorporation into a standard image analysis pipeline. Co-registration and analysis were performed through the STARVUE™ image data science pipeline, using UltiStacker.AI™ and UltiAnalyzer.AI™. The integration was evaluated across several key applications. --Pathology Assessment: H&E image data was overlaid onto the mIF channels, allowing for the histological context to be directly referenced when identifying the tumor and excluding non-target tissues from the analysis. --mIF-Enhanced H&E: Marker expression from mIF channels was overlaid onto the H&E images, allowing biomarker profiles to be visualized in the histological context. This colocalization enriched the understanding of immune activity beyond what H&E alone can provide. --Tissue Detection: mIF-based tissue detection often struggles with non-stained elements. Co-registered H&E images addressed this limitation by offering additional features for refined tissue-background separation, improving accuracy in measurements such as cell density. --Tumor Segmentation: Cytokeratin (CK) is commonly used to identify tumor epithelia in both brightfield and mIF assays. Tumors lacking CK expression or using non-CK assays pose challenges for segmentation. By leveraging co-registered H&E images, a deep learning model trained on H&E data successfully segmented tumor compartments, enabling accurate mIF-based cell analysis where CK-based methods failed. Conclusions: The results of this study highlight the enhanced utility of the gold standard H&E stain through integration of co-registered same-section H&E images into mIF workflows. This approach not only enriches tissue analysis but also lays the groundwork for developing novel methods that further broaden the applications of H&E in spatial biology. Citation Format: Michael Downing, Ruben Cardenes, Douglas Wood, Gourab Chatterjee, Lorenz Rognoni. Bridging spatial proteomics and classical histopathology: advanced spatial analysis through H&E-enhanced multiplexed immunoprofiling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 6310.