Aignostics has formed a partnership with the Pancreatic Cancer Action Network (PanCAN) to generate and share spatial profiling readouts for one of the most comprehensive pancreatic cancer datasets assembled to date.
Through this collaboration, Aignostics will analyse pathology images in PanCAN’s Spark data platform, utilising Atlas H&E-TME, Aignostics’ AI-powered application for comprehensive spatial profiling. Powered by Aignostics pathology foundation model, Atlas H&E-TME performs tissue quality control, cell classification, and tissue segmentation, generating over 4,500 quantitative readouts per image. Outputs will be made broadly accessible: academic researchers can access them at no cost through Spark, along with molecular, clinical, and imaging data from >1400 pancreatic patients, while life sciences companies can license outputs for commercial use.
Pancreatic cancer is one of the most challenging cancers to treat. Outcomes have improved modestly over the past several decades, partly because the disease’s deeply immunosuppressive biology limits the success of new therapies. By layering Atlas H&E-TME onto PanCAN’s rich multimodal dataset, the partnership aims to provide researchers with new tools to decode that biology and improve patient outcomes.
Viktor Matyas, CEO of Aignostics, said, “PanCAN has built something genuinely rare in SPARK, and we’re proud to play a part in making it even more powerful for researchers. Expanding access to high-quality spatial profiling data, whether through partnerships like this one or initiatives like OpenTME, is core to our mission of turning complex pathology data into actionable insights for patients.”
Sudheer Doss, PhD, PanCAN’s chief business officer and head of Patient Health Data, said, “We are committed to making high-quality pancreatic cancer data more accessible and useful for researchers. This collaboration brings powerful AI tools to PanCAN’s Spark dataset, turning complex data into biological insights that can help improve patient outcomes. By adding spatial information to this multi-modal dataset, we can better understand how tumours develop and interact with their surrounding environment, which may lead to new discoveries about the disease.”
