Article Text

Download PDFPDF

A prospective study of breast cancer risk in relatives of BRCA1/BRCA2 mutation carriers
  1. Erica Rowan,
  2. Aletta Poll,
  3. Steven A Narod
  1. Women’s College Research Institute, University of Toronto, Canada
  1. Correspondence to:
 Dr Steven Narod
 Women’s College Research Institute, 790 Bay Street, 7th Floor, Toronto, Ontario, M5G 1N8, Canada; steven.narod{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

The BRCA1 and BRCA2 genes were discovered in 1994 and 1995.1,2 Studies have shown that around 5–10% of breast cancer cases are due to a mutation in one of these genes.3 Once a mutation has been identified in a family, it is possible to unambiguously assign mutation status to unaffected female relatives; this is the basis for predictive genetic counselling. The risk of developing breast cancer up to the age of 70 years among women with a positive BRCA1 test is estimated to be from 45–87%.4,5 In contrast, women from a family with a BRCA1 or BRCA2 mutation, but who themselves test negative for the familial mutation, are considered to have the normal population risk for breast and ovarian cancer, which in Canada is around 7% up to 70 years of age. Women with a negative test result are generally advised to undergo the regular breast cancer screening practices, typically consisting of an annual mammograms and a clinical breast examination.

There is some question as to whether the risk of breast cancer in these mutation-negative women is in fact raised. Two recent studies reported an increase in breast and ovarian cancer risk for these women. Smith et al reported that 24% of 118 relatives with breast cancer or ovarian cancer cases in 277 families with a BRCA mutation were phenocopies, and they estimated the risk of breast cancer in mutation-negative women to be five times greater than expected.6 Gronwald et al studied unselected cases of breast cancer in Poland, 7 and found that only one of 17 affected sisters was a phenocopy, and that one of 72 non-carrier relatives had developed breast cancer. The risk of breast cancer was about double that expected.

In the current study, we followed up patients who were seen at the hereditary breast cancer clinic of the Women’s College Hospital; each of these women was negative for BRCA1 and BRCA2, but had a relative who was mutation-positive.


The patient population was recruited from women seen at the Familial Breast Cancer Center of Women’s College Hospital, Toronto, Ontario, between 2 January 1996 and 30 August 2006. The women were resident in Ontario, Canada and were aged 30–70 years. Inclusion criteria were: (1) a first-degree or second-degree relative with a documented BRCA1 or BRCA2 mutation, (2) the participant being negative for this mutation; and (3) no history of breast, ovarian or other cancer at the date of disclosure of the participant’s genetic test result. All women provided consent to be contacted by the research team. We contacted patients by telephone and conducted a short survey. Information was collected on personal cancer history, on current breast and ovarian cancer screening practices, and methods of prevention.

Statistical analysis

Each patient was considered to be at risk for breast cancer from the date of disclosure of the negative test result until the date of the follow-up interview. The expected number of breast cancers in the cohort was estimated as the product of the person-years of follow-up (categorised into 5-year periods, starting at the age of 30 and ending at the age of 70) and the age-specific breast cancer rates for the Ontario population for the period 1993–1997 (Cancer incidences in five continents).8 The standardised incidence ratio (SIR) is the ratio of the observed to expected number of cases.


In total, 104 subjects from 64 families (38 BRCA1 and 26 BRCA2) met the eligibility criteria. Three subjects were lost to follow-up. The median time of follow-up was 8 years (range 1 to 10). Three incident cases of invasive breast cancer were reported in the 101 women (2.9%). Based on the Ontario breast cancer rates, the expected number was 1.0 (SIR = 2.9; 95% CI 1.0 to 8.6) and the increase in risk was not statistically significant (table 1). No case of ovarian cancer was reported, but two cancers at other sites were reported (one thyroid and one hairy cell leukaemia) compared with an expected number of 1.7.

Table 1

 Risk of breast cancer in women who have tested negative for a familial BRCA1/BRCA2 mutation

The three breast cancers were diagnosed in women at ages 41, 43 and 48 years. The women were diagnosed with cancer 7, 8 and 2 years after the negative result, respectively. Two of the three women were considered to be at average risk of breast cancer, and one woman had been advised that she had a moderately raised risk, despite being mutation-negative (figure 1). She was diagnosed with breast cancer at the age of 41. Her maternal aunt was diagnosed with bilateral breast cancer (at the ages of the ages of 50 and 64) and was found to be positive for a BRCA1 mutation. The subject’s mother was also diagnosed with bilateral breast cancer (at the ages of 48 and 59); however, she was found to be mutation-negative. Therefore, owing to the presence of a case of bilateral breast cancer in the family, which was not accounted for by the familial mutation, this subject was considered to be at greater risk, despite the negative result. At the time of the test result disclosure, she had been advised that her risk was roughly double that of women in the general population. The other two women who developed breast cancer had been advised that they were not at greater risk. In these two families, there were no unaccountable cases of breast cancer.

Figure 1

 Pedigree of the patient deemed to be at increased risk of breast cancer despite negative testing for the known family mutations. Squares denote males; circles denote females. All symbols with a diagonal slash denote deceased members. Family members with breast cancer are shown by black symbols, and the age of diagnosis (if known) is shown. Family members with other cancers are in grey and the type of cancer is shown. Current age (or the age of death) is listed below each symbol. +, Presence of a BRCA mutation; −, tested negative for the family mutation; PSU, primary site unknown.

For two of the affected women, the complete sequence of the BRCA1 and BRCA2 genes were screened for mutations using a combination of techniques (protein truncation test, denaturing gradient gel electrophoresis and denaturing high-performance liquid chromatography). The third woman was tested for the Ashkenazi Jewish panel only. No BRCA1 or BRCA2 mutation was found.

Key points

  • Women from families with a BRCA mutations but who test negative for the mutation may be at greater risk of breast cancer.

  • The evidence is insufficient to recommend increased surveillance for mutation-negative women.

  • Hereditary breast cancer possibly has a polygenic component.


The case described here illustrates one of the difficulties encountered when conducting epidemiology studies using families that were selected for testing because of multiple case of breast cancer in the study. Had there not been two cases of bilateral breast cancer diagnosed in the sisters, this family probably would not have been identified as a candidate for genetic testing. This point was made by Gronwald et al—that is, that the presence of a sporadic case of breast cancer in the family is likely to increase the likelihood that the family will be ascertained and will therefore receive testing.7 Furthermore, when a negative result is communicated to a woman, the residual risk of breast cancer should be calculated after taking into consideration any breast cancers in the family that are not accounted for by the familial mutation.

In summary, we observed three cases of breast cancer in this small cohort of mutation-negative women, rather than the single case expected. This is consistent with two earlier reports.6,7 It is possible that these women face a risk of breast cancer that is greater than that of women in the general population, but further data are needed before this claim can be made with confidence.



  • Competing interests: None declared.

    Ethics approval granted by the ethics committee of the Sunnybrook and Womens Hospital Research Institute.