Artificial intelligence developer Owkin scored a major collaboration with Sanofi to help bring its digital clinical research platform to bear on the drugmaker’s core oncology efforts in four different cancers.
The two-part deal includes a $180 million equity investment in the company alongside a $90 million discovery and development partnership that spans the next three years, plus the chance for additional milestone-based payments based on the pair’s success.
Owkin’s research network sits between data scientists and international medical researchers and helps build and train deep-learning AI models on large, decentralized data sets without requiring all participants to pool their resources.
Using that federated system, the two French companies aim to preserve personal privacy while tapping into data collated from potentially hundreds of thousands of patients.
“Owkin’s unique methodology, which applies AI on patient data from partnerships with multiple academic medical centers, supports our ambition to leverage data in innovative ways in R&D,” Sanofi’s chief digital officer, Arnaud Robert, said in a statement.
That includes supporting the Big Pharma’s precision medicine efforts—including the development of biomarkers capable of predicting patient responses to different therapies—in non-small cell lung cancer, triple-negative breast cancer, mesothelioma and multiple myeloma. Owkin’s platform will also be employed to uncover new therapeutic targets and optimize clinical trial recruitment.
“We believe that the future of precision medicine lies in technologies that can unlock insights from the vast amount of patient data in hospitals and research centers in a privacy-preserving and secure way,” said Owkin co-founder and CEO Thomas Clozel.
This past September, the AI startup showed off a deep learning model capable of classifying a breast cancer patient’s risk of a metastatic relapse over the next five years.
In a retrospective study conducted with French cancer research institute Gustave Roussy, Owkin’s program analyzed digital tumor slides alongside the patient’s medical history and other clinical data, including age at surgery and surgery type. The results, showing accuracy of about 81%, were presented at the annual meeting of the European Society of Medical Oncology.
Before that, Owkin demonstrated it could help predict treatment responses in mesothelioma based on scans of tumor biopsy samples. That program, dubbed MesoNet, could also visually highlight the areas of the tissue slide that drove its predictions to help pathologists better classify patients.