Development and validation of risk prediction model for breast and ovarian cancers
Principal Investigator:
Professor Kenneth Muir
Approved Research ID:
5791
Approval date:
September 1st 2014
Lay summary
Ovarian and breast cancer are hormonally dependent cancers. Breast cancer is the most common female cancer in the UK. Although ovarian cancer has a lower incidence, its 5 year survival rate is half of that for breast cancer. Prevention and/or early detection are important in both diseases. Risk prediction models can be used to assess individual risk. So far, most risk prediction models for these cancers often include clinical indices with only very limited basic epidemiological factors such as family history, age etc. This proposal therefore aims to build a risk calculator for breast and ovarian cancers. We will utilize data on lifestyle/environmental factors, biomarkers and genetic markers. Data (not samples) from the full female cohort is required for women (ovarian and breast cancers and controls). In each cancer type, we will build risk models that predict individual risk. This research fits UK Biobank?s stated purpose in that it is in the public interest. For models developed using non-UK Biobank data the entire female cohort will also be used in order to perform prospective model validation. We will build one or more optimised risk prediction models fit for predicting risk in both sporadic and familial cases (including genetic markers, biomarkers, lifestyle/environmental factors collected within the female cohort). Familial cases are defined as breast and/or ovarian cancer cases with first degree relatives affected with breast and/or ovarian cancer. We will also explore whether ?any cancer? in first degree relatives also add to the prediction models. Both cohort and nested case control methods will be used. We will require data only (not samples) from the whole female cohort including full genetic data (not samples)and biomarker data (not samples) once available. We would like to receive lifestyle/environmental data first and in due course the genetic and biomarker data (not samples) when they become available.
