Incremental impact of breast cancer SNP panel on risk classification in a screening population of white and African American women.

Breast cancer risk prediction remains imperfect, particularly among non-white populations. This study examines the impact of including single-nucleotide polymorphism (SNP) alleles in risk prediction for white and African American women undergoing screening mammogram. Using a prospective cohort study, standard risk information and buccal swabs were collected at the time of screening mammography. A 12 SNP panel was performed by deCODE genetics. Five-year and lifetime risks incorporating SNPs were calculated by multiplying estimated Breast Cancer Risk Assessment Tool (BCRAT) risk by the total genetic risk ratio. Concordance between the BCRAT and the combined model (BCRAT + SNPs) in identifying high-risk women was measured using the kappa statistic. SNP data were available for 810 women (39 % African American, 55 % white). The mean BCRAT 5-year risk was 1.71 % for whites and 1.18 % for African Americans. Mean genetic risk ratios were 1.09 in whites and 1.29 in African Americans. Among whites, three SNPs had higher frequencies, and among African Americans, seven SNPs had higher and four had lower high-risk allele frequencies than previously reported. Agreement between the BCRAT and the combined model was relatively low for identifying high-risk women (5-year κ = 0.54, lifetime κ = 0.36). Addition of SNPs had the greatest effect among African Americans, with 12.4 % identified as having high-5-year risk by BCRAT, but 33 % by the combined model. A greater proportion of African Americans were reclassified as having high-5-year risk than whites using the combined model (21 vs. 10 %). The addition of SNPs to the BCRAT reclassifies the high-risk status of some women undergoing screening mammography, particularly African Americans. Further research is needed to determine the clinical validity and utility of the SNP panel for use in breast cancer risk prediction, particularly among African Americans for whom these risk alleles have generally not been validated.

The influence of health care policies and health care system distrust on willingness to undergo genetic testing.

PURPOSE: As the potential role of genetic testing in disease prevention and management grows, so does concern about differences in uptake of genetic testing across social and racial groups. Characteristics of how genetic tests are delivered may influence willingness to undergo testing and, if they affect population subgroups differently, alter disparities in testing. METHODS: Conjoint analysis study of the effect of 3 characteristics of genetic test delivery (ie, attributes) on willingness to undergo genetic testing for cancer risk. Data were collected using a random digit dialing survey of 128 African American and 209 white individuals living in the United States. Measures included conjoint scenarios, the Revised Health Care System Distrust Scale (including the values and competence subscales), health insurance coverage, and sociodemographic characteristics. The 3 attributes studied were disclosure of test results to the health insurer, provision of the test by a specialist or primary care doctor, and race-specific or race-neutral marketing. RESULTS: In adjusted analyses, disclosure of test results to insurers, having to get the test from a specialist, and race-specific marketing were all inversely associated with willingness to undergo the genetic test, with the greatest effect for the disclosure attribute. Racial differences in willingness to undergo testing were not statistically significant (P=0.07) and the effect of the attributes on willingness to undergo testing did not vary by patient race. However, the decrease in willingness to undergo testing with insurance disclosure was greater among individuals with high values distrust (P=0.03), and the decrease in willingness to undergo testing from specialist access was smaller among individuals with high competence distrust (P=0.03). CONCLUSIONS: Several potentially modifiable characteristics of how genetic tests are delivered are associated with willingness to undergo testing. The effect of 2 of these characteristics vary according to the level of health care system distrust, suggesting that policy decisions about delivery of genetic testing may influence differences in uptake across patient subgroups defined by levels of distrust rather than by race.

Age at diagnosis may trump family history in driving BRCA testing in a population of breast cancer patients.

BACKGROUND: Standard BRCA genetic testing criteria include young age of diagnosis, family history, and Jewish ancestry. The purpose of this study was to assess the effect of these criteria on BRCA test utilization in breast cancer patients. METHODS: Breast cancer patients aged 18 to 64 years living in Pennsylvania in 2007 completed a survey on family history of breast and ovarian cancer and BRCA testing (N = 2,213). Multivariate logistic regression was used to estimate odds of BRCA testing by patient characteristics, and predicted probabilities of testing were calculated for several clinical scenarios. RESULTS: Young age at diagnosis (<50 years) was strongly associated with BRCA testing, with women diagnosed before age 50 years having nearly five times the odds of receiving BRCA testing compared to women diagnosed at age 50 or older (OR = 4.81; 95% CI, 3.85-6.00; P < 0.001). Despite a similar BRCA mutation prevalence estimate (8-10%), a young Jewish patient <50 years with no family history had markedly higher predicted probability of testing (63%) compared with an older, non-Jewish breast cancer patient with more than one first-degree relative (43%). CONCLUSION: Age at diagnosis, Jewish ancestry, and both maternal and paternal family history are strongly predictive of BRCA testing. However, among women diagnosed at age 50 or older, family history may be an underused criterion that may benefit from targeted intervention. IMPACT: Robust methods specific to ascertaining detailed family history, such as through electronic medical records, are needed to accurately identify patients for BRCA testing.