Abstract B8: Predictors of BRCA1/2 genetic testing in a population of newly diagnosed breast cancer patients

Abstract

Abstract B8 Genetic testing for mutations in BRCA1 and BRCA2 (BRCA1/2) can influence surgical decisions for newly diagnosed breast cancer patients. The goal of this study was to examine predictors of BRCA1/2 testing in a population of newly diagnosed breast cancer patients, prior to definitive surgery. In this prospective trial participants were randomized in a 2:1 ratio to receive Rapid Genetic Counseling (RGC) or Usual Care (UC) following a baseline telephone interview. Those randomized to RGC were asked to complete genetic counseling as quickly as possible with the option of pursuing BRCA1/2 testing. Participants in the UC group were not formally referred for genetic counseling but could self-refer. Eligibility criteria included women diagnosed with breast cancer <50 years of age or diagnosed with breast cancer ≥50 years of age with a family history of breast cancer <50 years or ovarian cancer at any age. All newly diagnosed women who met our eligibility criteria were directly recruited from breast surgery clinics. Of the 126 women that have been randomized to the RGC group, 60 (47.6%) underwent BRCA1/2 testing. In bivariate analyses, predictors of BRCA1/2 testing included higher a priori risk of BRCA1/2 mutation calculated by BRCAPRO score, number of first degree relatives with breast and/or ovarian cancer, age less than 45 years, marital status, Jewish ethnicity, greater genetic testing knowledge, stronger preference for bilateral mastectomy, higher perceived risk of a second breast cancer, and higher perceived risk of a breast cancer recurrence. Participants who did not receive a specific surgical recommendation were more likely to undergo genetic testing as were participants who discussed genetic counseling with their physicians. Finally, women who were particularly concerned about the effect of surgery on body image were more likely to pursue genetic testing. To identify independent predictors of pre-surgical genetic testing, we entered all variables with significant bivariate associations to genetic testing into a backward logistic regression. The final logistic model included the following five variables: BRCAPRO score (OR=1.07, 95% CI=1.02, 1.12), marital status (OR=3.71, 95% CI=1.19, 11.54), perceived risk of breast cancer recurrence (OR=1.03, 95% CI=1.00-1.06), lack of surgical recommendation (OR=1.16, 95% CI=1.00-1.35), discussion of genetic counseling with a physician (OR=9.70, 95% CI=3.11, 30.24) and concerns regarding body image (OR=3.20, 95% CI=1.17, 8.74). The two variables most strongly related to genetic testing were BRCAPRO score (p=.003) and discussion of genetic counseling with a physician (p<.001). Surprisingly, perceived risk of recurrence was independently associated with testing while perceived risk of a contralateral breast cancer was not, raising the possibility that participants might not fully understand the distinction between a recurrence and a new primary cancer. These results suggest that newly diagnosed patients who are not provided with a specific surgical recommendation may be interested in BRCA1/2 testing to assist in their decision-making. The data also confirm prior reports that physician recommendation is a critical predictor of interest and utilization of genetic counseling and testing. Citation Information: Cancer Prev Res 2008;1(7 Suppl):B8.