KOHBRA BRCA risk calculator (KOHCal): a model for predicting BRCA1 and BRCA2 mutations in Korean breast cancer patients.

ABSTRACT

The widely used Western BRCA mutation prediction models underestimated the risk of having a BRCA mutation in Korean breast cancer patients. This study aimed to identify predictive factors for BRCA1/2 mutations and to develop a Korean BRCA risk calculator. The model was constructed by logistic regression model, and it was based on the Korean Hereditary Breast Cancer study, in which 1669 female patients were enrolled between May 2007 and December 2010. A separate data set of 402 patients, who were enrolled from Jan 2011 to August 2012, was used to test the performance of our model. In total, 264 (15.8%) and 67 (16.7%) BRCA mutation carriers were identified in the model and validation set, respectively. Multivariate analysis showed that age at breast cancer diagnosis, bilateral breast cancer, triple-negative breast cancer (TNBC) and the number of relatives with breast or ovarian cancer within third-degree relatives were independent predictors of the BRCA mutation among familial breast cancer patients. An age <35 years at diagnosis, bilateral breast cancer, both breast and ovarian cancer and TNBC remained significant predictors in non-familial breast cancer cases. Our model was developed based on logistic regression models. The validation results showed no differences between the observed and expected carrier probabilities. This model will be a useful tool for providing genetic risk assessments in Korean populations.