Owkin to Bring Agentic AI into Vienna’s Oncology Ecosystem
Owkin and the Comprehensive Cancer Center of the Medical University of Vienna have entered a collaboration to deploy agentic AI tools in oncology research and care. The partnership will introduce Owkin’s AI copilot, Owkin K, into Vienna’s oncology ecosystem, with a first focus on advanced bladder cancer. The system will be refined through feedback from clinicians and researchers, aiming to support biological reasoning, precision medicine, and therapeutic strategy development.
Owkin is an AI biotech company founded in 2016 by Thomas Clozel, MD, a clinical research doctor and former assistant professor in onco-hematology, and Gilles Wainrib, PhD, a specialist in machine learning in biology. The company develops AI technologies to analyze multimodal patient data, uncover new drug targets, predict patient outcomes, and support precision medicine strategies. Its work spans drug discovery, clinical trial optimization, and diagnostic solutions.
Owkin operates through a group of specialized entities: Bioptimus, focused on building a universal reference foundation model for biology; Owkin Dx, developing clinically validated digital pathology tools and biomarker screening solutions; Epkin, advancing OKN4395, a first-in-class selective triple inhibitor of the prostaglandin receptors EP2, EP4, and DP1, which play a role in inflammation and tumor immune evasion. OKN4395 is currently in Phase 1 clinical testing (INVOKE trial) and represents Owkin’s first drug program fully developed using its AI platform. Owkin Core (Owkin K) is the company’s software foundation, integrating multimodal data and “agentic AI” to power research and clinical applications.
See also: The Rise of AI Agents in Biotech, Where Are We Now?
The agreement centers on integrating Owkin K into clinical and research workflows at MedUni Vienna and University Hospital Vienna.
Owkin positions its agentic AI system under two related software environments: K Navigator and K Pro.
Owkin’s K Navigator is an AI co-pilot tailored for biomedical researchers, distinct from general-purpose language models. The platform is built to accelerate hypothesis generation and exploration of multimodal datasets, including spatial omics and gene expression data.
Key reported functions include:
- Data analysis and visualization – Researchers can analyze gene expression across different cell types, stratify results by variables such as sex, and generate a range of visualizations within seconds. Interactive visualization features allow users to explore datasets dynamically.
- Scientific reasoning – The system is designed to support hypothesis validation and gap analysis, combining statistical data with reasoning capabilities.
- Scientific writing – K Navigator can produce publication-style text, helping researchers draft review paragraphs or manuscript sections in academic tone.
- Knowledge sourcing – Outputs are restricted to credible scientific data sources, with functions for literature review and trend analysis to identify research opportunities.
Owkin positions K Navigator as an entry point into its broader “agentic AI” ecosystem, complementary to its forthcoming K Pro platform for decision-support in clinical and translational research.
The platform is currently in use by researchers at multiple academic and clinical institutions, including the University of Chicago, Johns Hopkins University, UCSF, Harvard Medical School, among others.
Initial efforts will concentrate on urologic cancers, beginning with advanced bladder cancer, a major clinical challenge worldwide. Vienna’s researchers will combine their proprietary patient datasets with Owkin’s platform to query high-resolution biological and clinical data. The system is designed to help identify clinical features linked to molecular signatures or therapy responses, and to examine immune cell infiltration and gene expression relationships across tumor environments.
According to the institutions, this process could accelerate hypothesis generation and validation, highlighting novel therapeutic targets and patient stratification approaches. By incorporating University’s of Vienna experts’ feedback, Owkin K’s underlying large language model is expected to be further fine-tuned for oncology applications.
The Medical University of Vienna, with more than 6,500 employees and 30 departments, is one of Europe’s largest biomedical research centers. Its Comprehensive Cancer Center connects clinical and research teams across cancer treatment, education, and translational research.
Topics: AI & Digital
