Article Text

Download PDFPDF

Unique de novo mutation of BRCA2 in a woman with early onset breast cancer
  1. M Robson1,
  2. L Scheuer1,
  3. K Nafa2,
  4. N Ellis2,
  5. K Offit1
  1. 1Clinical Genetics Service, Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
  2. 2Diagnostic Molecular Genetics Laboratory, Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY, USA
  1. Correspondence to:
 Dr Robson, Clinical Genetics Service, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10021, USA;

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.

Although it is widely stated that 5-10% of all breast cancers arise as the result of an inherited predisposition, the prevalence of mutations in BRCA1 or BRCA2 in unselected ascertainments of women with breast cancer is somewhat lower. In two population based series from the United States, presumably deleterious BRCA1 mutations were identified in only 14 female breast cancer patients out of a combined total of 884 women (1.6%).1,2 In two European series, BRCA1 or BRCA2 mutations were identified in 19/1035 (1.6%) Finnish breast cancer patients3 and 24/1220 (2.0%) breast cancer patients from the UK.4 Because of the low prevalence of detectable mutations, screening of unselected breast cancer patients has not been recommended. Mutation analysis is more often suggested for specific groups of breast cancer patients in whom mutations are more likely to be detected. The decision whether or not to offer genetic testing usually revolves around the presence or absence of a significant family history of breast or ovarian cancer, although age at diagnosis, bilaterality, and ethnicity may be important considerations. A negative family history, however, clearly does not exclude the presence of a germline mutation in BRCA1 or BRCA2. In a population based series from the United Kingdom, none of the 13 mutation carriers diagnosed with breast cancer before the age of 36 was reported to have had a family history of breast or ovarian cancer in first degree relatives.5 In contrast, in a series from the United States, most mutation carriers in a group of women with early onset breast cancer had a history of breast cancer in either first or second degree relatives.6 However, even in this series, one of the mutation carriers had no family history of breast cancer. The lack of a family history among mutation carriers may reflect small family size, non-penetrance, premature death of at risk women, or inadequacy of the history information itself.

Some affected mutation carriers without a family history may represent new mutations. The proportion of BRCA mutations that arise de novo is unknown. The two cases that have been reported to date appear to be recurrent mutations occurring at sites that may be predisposed to alteration.7,8 A woman with early onset breast cancer with no family history of breast or ovarian cancer was found to carry both the BRCA1 mutation 3888delGA and the BRCA2 mutation 6174delT.7 While the father of the proband was shown to carry the BRCA2 6174delT mutation, which is known to be a founder mutation in people of Ashkenazi descent, neither parent carried the BRCA1 3888delGA mutation, indicating that this alteration most likely arose de novo. Interestingly, an Ashkenazi woman with both early onset breast cancer and ovarian cancer had previously been reported to carry both the BRCA1 3888delGA mutation and the BRCA2 6174delT mutation,9 suggesting that the BRCA1 3888delGA mutation may, in fact, be a recurrent alteration developing at a mutational hot spot. Further evidence for such areas of predisposition to mutation comes from the recent report of a woman with early onset breast cancer who carried the BRCA2 mutation 3034del4.8 Although this particular mutation has been described many times in families of various ethnic origins,10 neither of the proband's parents carried the alteration, again suggesting a de novo origin at a genomic site prone to small deletions.

Both of the de novo BRCA1 or BRCA2 mutations that have been reported to date have been identified in other families. We now report a unique, previously undescribed de novo mutation, identified in a woman with early onset breast cancer.


The proband was a woman of Irish, Scots, and Welsh ancestry who was diagnosed at the age of 35 with bilateral infiltrating ductal carcinoma after investigation of abnormalities noted at the time of baseline mammography. She was treated with bilateral modified radical mastectomies and CMF chemotherapy. She had no family history of breast or ovarian cancer. Her mother was alive and without cancer at 71 years of age. She had five sisters, aged 32 to 47, none of whom had been diagnosed with either breast or ovarian cancer. Her father was diagnosed with colon cancer at the age of 57 and died of metastatic disease at 62. At 41 years of age, the proband enrolled in an Institutional Review Board approved study of genetic testing for hereditary breast cancer risk. After giving informed consent, she provided a sample of peripheral blood, from which DNA was extracted using standard techniques. Testing was performed at Myriad Genetic Laboratories (Salt Lake City, UT). Aliquots of DNA were subjected to polymerase chain reaction amplification and full sequence determination in both forward and reverse directions of the 23 exons of BRCA1, along with approximately 800 bp of intronic sequence surrounding the intron-exon boundaries. In addition, full sequence determination was performed of the 26 exons of BRCA2, along with approximately 900 bp of intronic sequence surrounding the intron-exon boundaries of this gene. Sequence analysis showed a single nucleotide insertion at nucleotide 7260 of BRCA2. The 7260insA insertion in exon 14 results in premature termination at codon 2359. Other than the proband, there are no reports of this alteration in the Breast Cancer Information Core database ( accessed 13 December 2001). The presence of the mutation was confirmed on a separate blood sample. Testing was then carried out on DNA samples obtained from the proband's mother and five sisters, none of whom was found to carry the mutation. Repeat testing of each of these family members was performed and confirmed the absence of the mutation in all.

To investigate the possibility of a de novo mutation, the paraffin embedded samples of the proband's father's colon cancer specimen were retrieved. DNA was extracted using standard techniques and analysed for the presence of the BRCA2 7260insA mutation. The mutation was not identified in the father's sample on initial or repeat analysis. To establish paternity, both parents and all sibs who had donated samples for DNA analysis were genotyped at three polymorphic loci (MYC-L1, D2S123, D17S250). The results of this genotyping are consistent with the reported paternity.


To our knowledge, this is the first report of a unique de novo mutation in either BRCA1 or BRCA2. Although this specific mutation has not previously been described, several other presumably deleterious alterations have been reported in its vicinity in exon 14. Examples include BRCA2 7253delAA (reported four times), 7297delCT (reported five times), and 7252C→T (Q2342X, reported once). This portion of the BRCA2 gene may therefore represent an area of susceptibility to mutation. The prevalence of de novo mutations in BRCA1 and BRCA2 is unknown. Although one would hypothesise that previously unreported mutations would be more likely to be the result of new mutational events, the recurrent nature of the two previously reported de novo alterations indicates that this criterion is not a reliable discriminator.

The identification of this mutation illustrates that analysis of BRCA1 and BRCA2 may be productive in the absence of a family history of breast or ovarian cancer. Empirical models estimating the likelihood of identifying germline mutations, such as BRCAPRO, are based upon the pattern of diagnoses within a family as a whole and have not been validated in women without a family history. Studies from the United Kingdom and United States have indicated that 5.9-9.4% of women diagnosed with breast cancer at the age of 35 or younger will have a detectable germline BRCA mutation.5,6 While many of these women will have a family history of breast or ovarian cancer, it is clear from the reported series that some will either have no such history5 or will only report breast cancer diagnoses in relatives of second or greater degree.6 Women with BRCA mutations are at increased risk of metachronous contralateral cancer, and it is therefore logical to assume that the presence of bilateral disease would increase the likelihood of detecting a germline mutation. However, the degree to which the presence of bilateral disease influences this probability has not been defined. Other factors, such as lack of hormone receptor expression, medullary histology, or high histological grade may also indicate an increased probability of germline mutation, although none of these are absolute discriminants.

The information derived from genetic testing of women with very early onset breast cancer may substantially influence clinical management. For example, after the mutation was identified, the proband underwent bilateral salpingo-oophorectomy in an attempt to reduce her ovarian cancer risk. Other women may elect to forego breast conservation therapy and undergo bilateral risk reducing mastectomy to address the substantial risk of metachronous contralateral disease. Because of the significant clinical implications both for the affected subject and for her family, as well as the less than complete sensitivity of a family history of breast cancer, consideration should be given to providing genetic counselling and discussing BRCA genotyping with all women with early onset breast cancer. The existence of de novo BRCA mutations should also be taken into account when generating penetrance estimates from genetic epidemiology studies. For instance, the kin-cohort model assumes that 50% of unaffected parents are obligate heterozygotes.11 The violation of this assumption by occurrence of de novo mutations may lead to underestimation of penetrance by this and similar models.