The Pulse
Lunit and Summit Put AI Pathology to Work on Ivonescimab
Lunit and Summit Therapeutics have signed a research agreement to analyze tumor tissue images with Lunit’s SCOPE IO platform. The work will study whether features in routine pathology slides can inform biomarker research for Summit’s invest
AI.info Team · Written with AI · fact-checked ·
Lunit and Summit Therapeutics are betting that routine pathology slides can reveal patterns useful in developing an immunotherapy, but their new agreement offers no evidence yet that those patterns predict treatment outcomes. Announced October 23, 2026, the research partnership will use Lunit’s SCOPE IO platform to examine tumor tissue in support of Summit’s ivonescimab programs. The companies say the analysis may help identify biomarkers and guide patient selection; they also state that they are announcing no clinical outcome data.
The distinction matters: the agreement tests an analytical approach, not a diagnostic tool or a treatment benefit. Lunit’s release says SCOPE products are for research use only and are not for diagnostic procedures or decisions.
What SCOPE IO will look for in routine slides
The research will apply SCOPE IO to whole-slide images stained with hematoxylin and eosin, commonly abbreviated H&E. The companies aim to characterize features of the tumor microenvironment—the cells, structures and spatial relationships within and around a tumor—and explore whether those features could support biomarker discovery, patient stratification or selection, and a deeper understanding of response to ivonescimab.
Lunit says its platform can assess H&E slides without requiring additional stains or assays. The company describes the intended output as quantitative, spatially resolved information on immune and tumor features, including tumor architecture, immune infiltration, stromal organization and interactions between tumor and immune cells. The announcement does not specify the number of samples, the tumors to be analyzed, a study timeline or a validation plan.
The drug-development question behind the deal
Summit is developing ivonescimab, an investigational bispecific antibody that targets PD-1 and VEGF, across several solid tumors. The drug is designed to combine PD-1 blockade, an immunotherapy mechanism, with VEGF inhibition, which acts on blood-vessel growth. Summit’s release identifies the molecule as a potential first-in-class candidate, but it is not approved in the company’s licensed territories, including the United States and Europe.
For Summit, the research question is whether measurements extracted from standard tissue images can help explain which tumor features are associated with response to this particular therapy. “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,” said Allen Yang, Summit’s chief research and development strategy officer.
Promised insights are not clinical evidence
Lunit CEO Brandon Suh framed the platform’s potential in terms of understanding which patients may benefit from immuno-oncology therapies and why. That is the intended research value, not a finding from this agreement: the companies report no results from the work and no clinical outcome data. Their announcement says any insights may inform translational research, exploratory biomarker strategies and clinical development planning.
The partnership therefore sets out a method and a drug-development use case, while leaving its practical value to be tested. For now, SCOPE IO’s role in this work is limited to research analysis of H&E slide images; the announcement does not establish a biomarker, a patient-selection rule or a change to ivonescimab’s clinical development.