The effort will focus on applying Lunit SCOPE IO to hematoxylin and eosin-stained (H&E) whole-slide images to characterize tumor microenvironment features. The companies aim to explore how AI-powered pathology analysis can support biomarker discovery, patient stratification or selection, and a deeper biological understanding of response to ivonescimab therapies.
Summit is developing ivonescimab, a novel, potential first-in-class investigational PD-1 / VEGF bispecific antibody, in several solid tumors. Lunit SCOPE IO is designed to support immuno-oncology research and biomarker development through AI-powered analysis of H&E slides, enabling scalable assessment of tumor microenvironment features without requiring additional stains or assays.
“Understanding the tumor microenvironment is increasingly important as immuno-oncology moves beyond single-axis checkpoint inhibition toward more complex mechanisms and combinations,” said Allen Yang, MD, PhD, Chief Research and Development Strategy Officer of Summit. “We are pleased to work with Lunit to explore how AI-powered pathology can help generate deeper, more scalable insights from routine tissue samples and inform the continued development of ivonescimab.”
“Summit is at the forefront of next-generation immuno-oncology development, and we are excited to support its efforts with Lunit SCOPE IO,” said Brandon Suh, Lunit CEO. “By extracting quantitative, spatially resolved immune and tumor microenvironment features from standard H&E slides, we believe SCOPE IO can help biopharma partners better understand which patients may benefit from immuno-oncology therapies and why.”
This agreement reflects the growing importance of computational pathology in oncology drug development. Routine H&E slides contain rich information about tumor architecture, immune infiltration, stromal organization, and spatial interactions within the tumor microenvironment. However, traditional manual assessment can be difficult to scale and may not fully capture the quantitative and spatial complexity of the tumor-immune interface. AI-powered analysis offers the potential to convert these routinely generated pathology images into reproducible, high-dimensional biomarker data.
The companies expect the work to generate tumor microenvironment insights that may inform future translational research, exploratory biomarker strategies, and clinical development planning. No clinical outcome data are being announced as part of this release.
