Waiv joins Daiichi Sankyo’s push into AI biomarker discovery: full analysis
Waiv, the Paris-based AI precision testing company formerly known as Owkin Dx, said May 6 that it has partnered with Daiichi Sankyo to lead digital pathology biomarker discovery for an ADC program in oncology. The companies said Waiv will analyze early-phase trial material using AI models applied to H&E and IHC slides, with the goal of finding biomarkers linked to treatment response before the program moves into later clinical stages. (via.tt.se)
The collaboration fits a clear pattern in Daiichi Sankyo’s oncology strategy. The company has been steadily building out AI-enabled biomarker partnerships around its ADC pipeline, including a December 2025 deal with Lunit focused on translational oncology research, a March 2026 collaboration with Tempus aimed at biomarker discovery and clinical differentiation in an ADC clinical program, and a separate April 2026 agreement with Imagene AI centered on multimodal biomarker discovery and response prediction. Taken together, those partnerships suggest Daiichi Sankyo is investing not just in ADC molecules themselves, but in the data infrastructure needed to identify which patients are most likely to benefit. (lunit.io)
In the Waiv deal, the company said it will apply its end-to-end computational pathology platform to early-phase data, including tumor microenvironment analysis across both H&E and IHC-stained samples. The stated aim is to generate biomarkers and outcome-prediction signals that can inform the next clinical trial phases. Waiv framed the effort around one of the harder problems in oncology development: discovering useful biomarkers when datasets are small, sometimes under 100 patients. The company says its models are supported by foundation models trained on hundreds of thousands of images from an international data network spanning academic institutions, hospitals, and laboratories. (via.tt.se)
Waiv’s positioning also matters. On its site, the company describes itself as focused on clinical-grade, AI-powered tests for biomarker discovery, outcome prediction, treatment-response assessment, and precision testing across oncology. The company recently disclosed backing from OTB Ventures, Alpha Intelligence Capital, Serena Data Ventures, Karista, SistaFund, and CRB Health Tech, with initial financing of $35 million to expand its testing solutions and commercial footprint. That helps explain why a pharma partner like Daiichi Sankyo may view Waiv as more than a discovery vendor, and potentially as a route toward deployable biomarker assays later in development. (wearewaiv.com)
The announcement itself did not name the ADC asset involved, and no outside expert commentary tied specifically to the Waiv deal was readily available in primary sources reviewed. Still, industry messaging around Daiichi Sankyo’s other recent AI collaborations has been consistent: use multimodal and pathology-based AI earlier in development to improve biomarker hypotheses, strengthen patient stratification, and support companion diagnostic strategy. In that sense, the Waiv agreement looks less like a one-off and more like another building block in a broader translational oncology playbook. That is an inference based on the timing and similarity of Daiichi Sankyo’s recent partnerships. (imagene-ai.com)
Why it matters: For veterinary professionals, especially those tracking comparative oncology, this is a useful signal about where cancer drug development is heading. AI pathology platforms are increasingly being asked to work in exactly the kinds of environments veterinary medicine knows well: limited case counts, heterogeneous samples, and a need to extract more predictive value from routine histopathology. Although this collaboration is in human oncology, the operational questions, how to validate image-derived biomarkers, how to standardize slide-based prediction, and how to turn exploratory signals into clinically usable tests, are highly relevant to veterinary diagnostics and oncology research. (via.tt.se)
There is also a practical industry lesson here. ADC developers are under pressure to show not only efficacy, but also which subgroups are most likely to respond and how to position therapies in increasingly crowded treatment landscapes. If AI-derived pathology biomarkers can help enrich trials earlier, they could reduce development risk and improve the odds of successful companion diagnostic strategies. For veterinary oncology companies and academic groups, that raises the bar for how pathology, imaging, and outcomes data may need to be collected and linked in future translational studies. (finance.yahoo.com)
What to watch: The next key signals will be whether Daiichi Sankyo identifies the ADC program publicly, whether Waiv’s work advances into a named biomarker assay or companion diagnostic path, and whether future trial updates show AI-derived pathology markers being used for enrollment, stratification, or response analysis. (via.tt.se)