Owkin's K Pro Autopilot agent proposed a KRAS G12D drug combination that widened eligible patients from about 10% to 60%.
Over 40% of pancreatic cancers carry KRAS G12D, but KRAS inhibitors lack durability and only about 10% of patients were eligible for combination therapy. K Pro's agentic Autopilot Mode ran lab-in-the-loop work to identify a new synthetic lethal vulnerability and a predicted combination. The page reports eligibility expanding from 10% to 60% and synergy confirmed in preclinical cell lines.
Owkin is a French AI company building biology models and the K Pro AI Scientist, working with pharma on drug positioning, trial design and pathology diagnostics.
This sits in the Build stage of the AI use-case lifecycle: Named examples from the foundation-model labs and platforms pharma teams use to actually construct, train and generate the model or application.
Published by Owkin on its own site — read the original case study on owkin.com ↗
Related use cases
- 01Boehringer Ingelheim used Owkin's K Pro multi-omic reports to prioritize targets and find spatially informed patient-selection biomarkers.
- 02Epkin used Owkin's K Pro to map prostaglandin biology and design the Phase 1 strategy for OKN4395.
- 03An oncology pharma used Owkin's K Pro and MOSAIC data to triage bispecific ADC candidates in silico.
- 04Cleveland Clinic Abu Dhabi cut cohort assembly from days or weeks to seconds with Owkin's data transformation AI and K Pro.
- 05Owkin's MesoNet model predicts mesothelioma survival from pathology images and could shrink Phase 3 trial sizes.
- 06Compugen added Owkin's K Pro alongside its in-house target discovery AI platform.