Computational prognosis of therapies for solid tumors

This project aims to establish a transnational, data-driven framework for the computational prognosis of breast cancer therapies, supporting precision oncology through digital twin technology. Building on the CancerMate platform (TRL 7), the project integrates mechanistic mathematical modeling of tumor growth and therapy response with machine-learning techniques to predict patient-specific outcomes across multiple breast cancer subtypes and neoadjuvant treatment regimens.
Central to the approach is the development and validation of virtual biomarkers—computational counterparts of clinical biomarkers—that capture underlying biological processes governing tumor progression under therapy. These virtual biomarkers enable robust stratification of retrospective patient cohorts and support prospective predictions of treatment efficacy, toxicity, and resistance patterns. By combining mechanistic insight with data-driven optimization, the project enhances the interpretability, robustness, and clinical relevance of predictive oncology models.
The cooperation seeks partners with complementary expertise in clinical oncology, biomedical imaging, data science, and human–computer interaction to implement an integrated, automated workflow. Particular emphasis is placed on the development of advanced graphical user interfaces and immersive extended reality (XR) environments to facilitate model interpretability, clinician training, and seamless adoption within hospital infrastructures.
The project will deliver a decentralized decision-support ecosystem that improves care quality, reduces clinical burden, and optimizes healthcare costs. Patients, clinicians, funders, and technology providers will be actively engaged through participatory workshops and training activities. All generated data and computational tools will adhere to FAIR principles and Open Science practices, contributing to sustainable innovation and alignment with the objectives of the European Cancer Mission.