to the editor: I am glad that American Family Physician featured the important topic of assessing cancer risk, and breast cancer risk in particular. However, there are inaccuracies that need to be corrected in the section “Genetic Mutations and Family History.”

The article states, “The U.S. Preventive Services Task Force (USPSTF) has recommended that physicians offer genetic testing to several groups of women.” ¹ Readers are referred to Table 2, which details USPSTF recommendations for testing for BRCA mutations based on family history criteria. However, the specific recommendation of the USPSTF is that women who have not had breast or ovarian cancer be referred for genetic counseling (and consideration of testing) if their family history meets certain criteria.²

There are two serious problems with stating that people who meet these family history criteria “should be tested” for BRCA mutations that confer susceptibility to cancer. First, a substantial proportion of people with such family histories will have a low pretest probability of carrying a deleterious BRCA mutation.^3,4 Testing these persons might be unnecessary, but advising them about familial risk, cancer screening, and prevention might still be important. Furthermore, current testing does not identify all deleterious BRCA gene alterations, and the clinical significance of some gene variants is uncertain. Second, for most people whose family history suggests increased risk for hereditary breast and ovarian cancer, proceeding to genetic testing without first testing a family member who has had breast or ovarian cancer will be uninformative.^5,6 For these reasons, the USPSTF recommends referring some women for genetic counseling to determine if BRCA gene sequencing is indicated. Genetic testing for all persons who meet family history criteria could be a colossal waste of resources. In contrast, widespread testing has been advocated for women in population subgroups, such as Ashkenazi Jews who have a higher risk for three particular BRCA mutations⁷; however, pretest and post-test counseling would still be essential.

The footnote to Table 2 states, “Second-degree relatives include grandparents, aunts, uncles, and cousins.” Except in families with consanguinity, first cousins are actually third-degree relatives, with one eighth of their genetic material in common. Second-degree relatives have one fourth of their genetic material in common. Cousins are not included in the USPSTF recommendations nor in most other rule-based criteria for increased risk of inherited breast cancer, though they may sometimes be important in an analysis of a detailed pedigree. I heartily endorse the goal of recording the family cancer history of maternal and paternal relatives, including their ages at diagnosis.

in reply: We thank Dr. Acheson for her careful reading of our article. We agree completely with her statement about the USPSTF recommendations and regret that the statement was inadvertently shortened to its printed form during the editing process. We agree that, in general, cousins are not classified as second-degree relatives. However, our personal clinical experience and conversations with colleagues suggest that many patients do not know the health history of older relatives such as aunts and uncles, because that generation may have been less likely to disclose details of their personal health to younger relatives. In these cases, knowing the health history of cousins might provide clues or suggestions that are worth considering as part of the general decision about genetic counseling.