Case study: PACpAInt
Authors
Case study: PACpAInt
Authors
About Owkin
Owkin is an AI biotechnology company that uses AI to find the right treatment for every patient. We combine the best of human and artificial intelligence to answer the research questions shared by biopharma and academic researchers. By closing the translational gap between complex biology and new treatments, we bring new diagnostics and drugs to patients sooner.
We use AI to identify new treatments, de-risk and accelerate clinical trials and build diagnostic tools. Using federated learning, a pioneering collaborative AI framework, Owkin enables partners to unlock valuable insights from siloed datasets while protecting patient privacy and securing proprietary data.
Owkin was co-founded by Thomas Clozel MD, a former assistant professor in clinical onco-hematology, and Gilles Wainrib, a pioneer in the field of machine learning in biology, in 2016. Owkin has raised over $300 million and became a unicorn through investments from leading biopharma companies (Sanofi and BMS) and venture funds (Fidelity, GV and BPI, among others).
Identifying pancreatic adenocarcinoma molecular subtypes from routine histology slides
Context
Pancreatic adenocarcinoma (PAC) is a very heterogeneous tumor with a high trial failure rate. Currently, molecular subtypes are defined by RNA profiling whose limitations prevents its application in routine care.
Methods
We used a multiple instance learning model with a self-attention mechanism called PACpAInt.
This multistep approach used deep learning models to detect PAC tumors from histology slides and predict molecular subtypes.
Results
Identified molecular subtypes in the three validation cohorts with independent prognostic value comparable to RNAseq.
Identified inter-slide heterogeneity in 39% of tumors that impacted survival. This helped us refine existing subgroups based on tumor heterogeneity.
Impact
Increased statistical power - Pharma can increase the statistical power of phase III trials by using this tool to select high-value subgroups with the greatest unmet need and that are most likely to benefit from the treatment.