Imagene AI, a leader in multimodal AI for precision oncology, has announced a new collaboration with global pharma innovator Daiichi Sankyo, aiming to accelerate biomarker discovery and improve how patients are matched to therapies.
Under the deal, Daiichi Sankyo will tap Imagene AI’s OI Suite—a multimodal platform powered by the CanvOI foundation model—to extract deeper biological insights from tumor imaging data.
By combining Hematoxylin and Eosin (H&E) and Immunohistochemistry (IHC) whole-slide images with molecular profiles and long-term clinical outcomes, the partnership seeks to identify critical biomarkers earlier in the drug development process.
The objective is clear: stronger patient stratification, earlier validation of biomarker hypotheses, and more data-driven decisions from early research through late-stage clinical trials.
“Collaborating with Daiichi Sankyo reflects a shared commitment to advancing biomarker discovery as a key driver of development success,” said Dean Bitan, Co-founder and CEO of Imagene AI.
“By working together, we are integrating multimodal discovery and quantitative IHC scoring to move from biomarker hypothesis to patient stratification with greater confidence, and to generate quantitative signals to enhance companion diagnostic strategy and improve how patients are matched to the therapies most likely to benefit them.”
At the heart of the collaboration is a focused effort to pinpoint biomarkers linked to treatment response—particularly for Daiichi Sankyo’s antibody drug conjugate (ADC) programs.
Using large-scale real-world multimodal data, Imagene AI will deploy its AI-driven pipelines to build response prediction models, mapping biological pathways and key histological features tied to therapeutic outcomes.
The company will also roll out its proprietary Composite Continuous Scoring system, an AI-powered method designed to quantify target expression from IHC data. By integrating multiple variables into a single continuous score, the approach promises a more precise and biologically grounded way to evaluate treatment targets.
Backing the initiative is Imagene AI’s expansive real-world data infrastructure, which includes more than 3.5 million tissue samples alongside rich omics and clinical outcomes data. This massive dataset is expected to give the collaboration an edge in navigating complex and data-limited drug development environments.
Together, the two companies are betting that multimodal AI—and smarter biomarker discovery—can unlock faster, more precise oncology breakthroughs.
