Accurate breast cancer risk assessment is essential for selecting appropriate surveillance and preventive interventions. Current clinical assessments rely primarily on family history and rare pathogenic variants, although common genetic variation explains a large and clinically relevant proportion of inherited risk. This project will evaluate the clinical implementation of breast cancer polygenic risk scores (PRS) among women managed by the Hereditary Cancer Unit at Karolinska University Hospital. We will initially recruit 100-200 families with at least two female relatives enrolled in breast surveillance and evaluate how adding a validated 313-variant breast cancer PRS to CanRisk-based assessments affects estimated risk, surveillance recommendations, and consideration of risk-reducing surgery. We will quantify variation between relatives from the same family and examine the additional contribution of mammographic breast density. We will also evaluate larger genome-wide PRS models and their performance in clinically relevant patient groups. The project has ethical approval from the Swedish Ethical Review Authority, Dnr 2025-00772-01. The project forms part of a substantive collaboration with the programme “Early Detection of Breast Cancer through AI-guided Screening, Risk Identification and Multi-Modal Surveillance,” which received SEK 22 million from Cancerfonden for 2026–2029. Data from our PRS project will be incorporated into this collaborative programme, enabling joint analyses of genetic, clinical, and image-derived risk factors. The collaboration includes AI-supported analysis of mammography and MRI, a prospective screening study of up to 200,000 examinations, a retrospective study of approximately 2,600 women with hereditary or familial risk and 12,400 population controls, and multimodal analyses in high-risk cohorts. Within this framework, our primary contribution is the generation and clinical evaluation of PRS and the integration of polygenic risk with family history, monogenic findings, breast density, and AI-derived imaging measures. NAISS resources are required for genotype quality control and imputation, genome-wide PRS calculation, multimodal modelling, and repeated statistical validation. All data will be pseudonymized and processed in accordance with the ethical approval, GDPR, and institutional data-governance requirements. The project will provide evidence on whether polygenic risk can improve individualized breast cancer prevention and how genetic and imaging-derived risk information can be combined in a clinically useful manner.