Waiv, Daiichi Sankyo Partner on AI Biomarker Discovery

Waiv, formerly known as Owkin Dx, has entered into a collaboration with Daiichi Sankyo to support digital pathology biomarker discovery for an antibody-drug conjugate (ADC) development program, the companies announced.

The Paris-based biotechnology company will use its artificial intelligence-powered computational pathology platform to analyze early-phase clinical data and identify biomarkers that may predict patient response to treatment. The collaboration is aimed at improving precision medicine strategies in oncology, particularly for ADC therapies, which combine targeted antibodies with cancer-killing drugs.

Under the agreement, Waiv will conduct tumor microenvironment (TME) analysis using both hematoxylin and eosin (H&E) and immunohistochemistry (IHC) stained pathology samples. The company will also apply its biomarker discovery and outcome prediction technologies to identify histopathological indicators associated with treatment response ahead of future clinical trial phases.

The partnership highlights the growing role of artificial intelligence in cancer drug development, especially in areas where clinical datasets are limited. Biomarker discovery in small patient populations remains one of the pharmaceutical industry’s most complex challenges, particularly for advanced therapies such as ADC medicines.

Waiv said its computational pathology platform was specifically designed for low-data environments, enabling researchers to extract clinically meaningful insights even when fewer than 100 patient samples are available. The platform is powered by foundation models trained on hundreds of thousands of pathology images collected through the company’s international data network, which includes academic institutions, hospitals and laboratories worldwide.

The company’s AI approach combines several capabilities, including the development of customized machine learning models optimized for limited datasets, extraction of predictive signals directly from whole-slide pathology images, and identification of novel biomarkers beyond predefined pathological features. Waiv added that its system is designed to generate reproducible and interpretable outputs suitable for clinical decision-making and future diagnostic test development.

Meriem Sefta said the collaboration reflects the company’s strength in translating complex pathology data into clinically actionable biomarker solutions.

“Identifying which patients will respond to a therapy directly from the pathology slide is simultaneously one of the hardest problems and one of the most important opportunities in oncology drug development,” Sefta said. “This collaboration with Daiichi Sankyo reflects our ability to engage early and take biomarkers all the way through to clinically validated, deployable tests.”

The agreement further expands the use of AI-driven pathology tools in oncology research as pharmaceutical companies increasingly seek more efficient ways to match therapies with patients most likely to benefit.