Real-World Evidence for Breast Cancer Risk Stratification: Insights from the P.I.N.K. Study

Background/Objectives: Early detection of breast cancer (BC) substantially improves outcomes, with a 5-year survival rate of 99% for early-stage disease. Risk-based screening, which tailors recommendations according to individual clinical, personal, and lifestyle factors, is a promising strategy to enhance early detection. However, previous studies have reported inconsistent evidence of association with BC risk for several modifiable lifestyle factors including premenopausal adiposity, sedentary behaviour, oral contraceptive use. To address these uncertainties, we investigated the association between clinical, personal, and modifiable lifestyle factors and BC risk in the Prevention, Imaging, Network, Knowledge (P.I.N.K.) project. P.I.N.K. generated real-world evidence by integrating clinical data with information collected through a structured questionnaire evaluating the associations between BC and clinical, personal, and lifestyle factors in 27,123 Italian women aged ≥ 40 years, who underwent integrated breast imaging diagnostics. Methods: Characteristics of women with and without BC were compared. Adjusted logistic regression, Least Absolute Shrinkage and Selection Operator (LASSO), and Random Forest models were applied to identify relevant predictors. Variables selected by at least two methods were included in a multivariate logistic regression model, with BC diagnosis as the outcome. Results: Accordingly with previous studies a consistent association was found for high breast density, late menopause, overweight, abdominal adiposity, cardiovascular-cerebrovascular and oncological comorbidities and smoking. Greater consumption of plant-based foods, lower intake of red and processed meat, and regular physical activity were associated with lower odds of BC as well. Conversely, an unexpected inverse association was assessed for menstrual irregularities, hyperlipidaemia, and oral contraceptive use. Conclusions: The integration of multiple sources of information both clinical and self-reported data collected directly from women adds to the growing evidence on the role of modifiable lifestyle factors in BC risk in supporting more targeted prevention strategies. Future prospective studies should further investigate the long-term effects of lifetime physical activity trajectories, different oral contraceptive formulations, and detailed reproductive history, including menstrual cycle characteristics and menopausal timing, to refine individualized BC risk prediction.