Approved research

Professor Montserrat Garcia-Closas

Lay summary

Aims of research: Increasing knowledge of breast cancer (BC) risk factors can be translated into improvements in prevention and early detection of BC by achieving better risk stratification in the population and tailoring interventions according to risk. Paramount to this is the validation of risk stratification models in prospective cohort studies, and the discovery of novel biomarkers of risk. Aim. To validate risk prediction models for breast cancer: a) Evaluate and compare the performance of established BC risk prediction models (e.g. BCRAT, IBIS, BOADICEA). b) Evaluate the performance of new risk prediction models under development including environmental, biomarker and genetic information. This aim will be addressed using data both from the UK Biobank and from the Breakthrough Generations Study (BGS), a prospective cohort of > 110,000 women in the UK led by our group. How the research meets UK Biobank?s stated purpose: The proposed research will result in validated models for prediction of BC that can be used in risk-stratified strategies for prevention and early detection. The addition of appropriate data from BGS participants will improve the ability to achieve these aims. We are requesting an extract of raw data on BC risk factors and outcomes for women in the UK Biobank to derive variables for our analyses. Comparable variables with BGS will be harmonised for pooled analyses, effectively increasing the number of women with data suitable for analysis by almost 50%. Estimates of the risk of developing breast cancer will be derived for study participants using risk models. The model performance will then be evaluated within each cohort as well in the pooled dataset, e.g. by comparing the numbers of cases observed and predicted by the models within risk categories. Female members of any age participating in the UK Biobank cohort (currently n=273,474).