Berlin, Germany, and Limassol, Cyprus – February 2026 – StratifAI, the AI biomarker company for precision oncology, today announced a strategic partnership with German Medical Institute Cyprus (GMI) to support the analytical validation of Polaris™, the company’s multimodal AI platform, as the foundation for developing novel biomarkers that are predictive of treatment outcomes. This collaboration is part of StratifAI’s broader validation strategy for the Polaris™ platform, following a series of recent partnerships aimed at generating robust evidence for clinical use.
Advancing risk stratification in early breast cancer
Reliable assessment of recurrence risk is central to adjuvant treatment decision-making in early breast cancer. While clinicopathological factors and genomic assays provide valuable guidance, limitations remain in accessibility, scalability, and reproducibility across healthcare systems.
There is a growing need for validated diagnostic tools that integrate seamlessly into routine pathology workflows while delivering consistent clinical performance.
Polaris™ Breast: AI-based risk assessment from routine histology
Polaris™ Breast applies multimodal artificial intelligence to digitized histopathology slides to assess relapse risk directly from routinely collected tissue samples. By identifying complex morphological patterns within tumor tissue, the model aims to support more refined and individualized treatment planning in early-stage disease. The platform is designed to be robust across common variations in clinical practice, such as scanning hardware, and tissue preparation, conditions that routinely challenge AI systems in clinical deployment.
Validation with GMI
Through this partnership, StratifAI and GMI are conducting ananalytical validation study assessing the precision and reproducibility of Polaris™ Breast across variables related to histology slide preparation and digitization thereof. The study is structured to evaluate key analytical performance metrics, including repeatability, scanner-to-scanner consistency, and signal stability across tissue processing variables, in accordance with frameworks relevant to regulatory submissions in oncology diagnostics. Critically, the study includes both core needle biopsy specimens from the neoadjuvant setting and surgical resection specimens from the adjuvant setting, enabling a rigorous assessment of Polaris™ performance across fundamentally different tissue types, preparation conditions, and morphological contexts.
Biopsies and resections present distinct analytical challenges and demonstrating consistent, stable performance is crucial.
Building evidence for real-world deployment
The collaboration with GMI strengthens StratifAI’s growing network of clinical and validation partners supporting the Polaris™ platform. These efforts aim to establish a strong, multi-site evidence base demonstrating robustness, generalizability, and clinical utility, key requirements for adoption in routine oncology practice.
