Redefining breast cancer screening: advances in risk stratification and personalisation

Abstract Background Conventional breast cancer (BC) screening approaches, primarily based on age criteria, often fail to consider individual variability in risk. This results in challenges such as overdiagnosis and false-positive results. Aim This review aimed to explore strategies for risk-stratified breast cancer screening and to evaluate how personalised tools improved early detection and screening efficiency. Main text Risk-stratified screening tailors the screening protocol, such as starting age, frequency, and imaging modality, according to individual risk profiles. These profiles are assessed using an evolving array of demographic, clinical, genetic, imaging, and artificial intelligence (AI)-based tools. Current risk models integrate family history, breast density, and genetic mutations; however, they vary in accuracy and applicability across populations. Advances such as polygenic risk scores and AI-driven image analysis are poised to refine stratification and reduce unnecessary interventions. Despite promising developments, widespread clinical implementation remains limited due to insufficient prospective evidence and standardisation. Conclusion Transitioning towards personalised screening is essential to reduce mortality while minimising harm. Integrating genetic testing, imaging biomarkers, and AI tools into routine screening protocols may enable more precise, individualised recommendations, thereby aligning practice with the principles of personalised medicine.