BACKGROUND: With the identification of genes predisposing to hereditary breast cancer, the accurate and consistent estimation of a woman's risk of developing breast cancer based on her family history is of paramount importance if national service guidelines are to be developed. PATIENTS AND METHODS: The residual lifetime risk of developing breast cancer was estimated for 200 women attending a breast cancer genetic assessment clinic by three different methods currently in use in the UK. Risks were computed on the basis of the Cancer and Steroid Hormone (CASH) study data and were classified as 'low/moderate' (<20%) or 'high' (>20%). These risk categories are representative of those currently used to allocate surveillance and genetic testing. Risks were then compared to estimates derived by other methods used in current clinical practice, including those of Houlston and Murday. RESULTS: The CASH data-based method ascribed 27% to the high risk category, as compared to 53% for the combined Houlston and Murday methods. A method based on the number of affected relatives alone ascribed only 14% to the high risk category. Overall, 108 (54%) women were placed in the same risk category by all three methods. CONCLUSIONS: This study demonstrates that there is a significant degree of variability between methods currently used to estimate breast cancer risk which has serious implications for individual patient management, service provision and multicentre studies evaluating the benefits of genetic testing for breast cancer susceptibility.
BACKGROUND: With the identification of genes predisposing to hereditary breast cancer, the accurate and consistent estimation of a woman's risk of developing breast cancer based on her family history is of paramount importance if national service guidelines are to be developed. PATIENTS AND METHODS: The residual lifetime risk of developing breast cancer was estimated for 200 women attending a breast cancer genetic assessment clinic by three different methods currently in use in the UK. Risks were computed on the basis of the Cancer and Steroid Hormone (CASH) study data and were classified as 'low/moderate' (<20%) or 'high' (>20%). These risk categories are representative of those currently used to allocate surveillance and genetic testing. Risks were then compared to estimates derived by other methods used in current clinical practice, including those of Houlston and Murday. RESULTS: The CASH data-based method ascribed 27% to the high risk category, as compared to 53% for the combined Houlston and Murday methods. A method based on the number of affected relatives alone ascribed only 14% to the high risk category. Overall, 108 (54%) women were placed in the same risk category by all three methods. CONCLUSIONS: This study demonstrates that there is a significant degree of variability between methods currently used to estimate breast cancer risk which has serious implications for individual patient management, service provision and multicentre studies evaluating the benefits of genetic testing for breast cancer susceptibility.