The authors declare no conflict of interest.
Screening for BRCA1 and BRCA2 mutations among French-Canadian breast cancer cases attending an outpatient clinic in Montreal
Article first published online: 5 JUN 2013
© 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Special Issue: BRCA1 and BRCA2
Volume 85, Issue 1, pages 31–35, January 2014
How to Cite
Screening for BRCA1 and BRCA2 mutations among French-Canadian breast cancer cases attending an outpatient clinic in Montreal., , , , , , , , , , , .
- Issue published online: 12 DEC 2013
- Article first published online: 5 JUN 2013
- Accepted manuscript online: 26 APR 2013 12:20PM EST
- Manuscript Revised: 19 APR 2013
- Manuscript Accepted: 19 APR 2013
- Manuscript Received: 24 DEC 2012
- Canadian Breast-Cancer Research Alliance (CBCRA)
- Fonds de la recherche en santé du Québec (FRSQ)
- breast cancer;
Study subjects were French-Canadian women with ductal carcinoma in situ (DCIS) or invasive breast cancer (incident or prevalent) who were treated and followed at a single breast cancer clinic affiliated with the Research Center of University of Montreal (CRCHUM), who were either aged less than 50 years at diagnosis or who were 50 years or older and with at least two affected first- or second-degree relatives. Subjects were tested for six founder mutations (three in BRCA1 and three in BRCA2); 1093 eligible cases were tested. Of these, 56 women (5.1%) were mutation carriers, including 43 BRCA2 carriers and 13 BRCA1 carriers. The prevalence of mutations was 5.3% for unselected women aged 50 and less and was 4.6% for familial cases over age 50. The prevalence of mutations was 3.3% for women with DCIS and was 5.3% for women with invasive cancer. It is rational to offer genetic testing to all French-Canadian women diagnosed recently or in the past with either DCIS or invasive breast cancer before age 50 or with familial breast cancer above age 50.
Breast cancer is the most frequently diagnosed cancer among women in Canada. In 2011, the annual estimated rate was 102 per 100,000 women in Canada and 110 per 100,000 women in Quebec . Among gene carriers, the cumulative risk of breast cancer by 70 years of age has been estimated to be 65% for BRCA1 carriers and 45% for BRCA2 carriers , but penetrance depends on the location of the mutation and the country of residence . In Quebec, genetic testing is offered to women with breast cancer and a positive family history of breast or ovarian cancer, but not to women based purely on their age of onset or pathologic features. The main objective of this study was to estimate the yield of genetic testing when offered routinely to women who attend a large breast-cancer outpatient facility.
We have previously shown that 4.4% of 564 French-Canadian women with early-onset breast cancer carry one of six founder mutations in BRCA1 or BRCA2 . We wished to extend this earlier study to encompass a larger study population and to include patients with a past history of breast cancer and with ductal carcinoma in situ (DCIS) as well. Women were eligible for this study, if they were French-Canadian and had either invasive breast cancer or DCIS diagnosed at age 50 or below in the past 10 years and were under active follow-up. Women with invasive breast cancer or DCIS diagnosed between 51 and 70 years of age with a family history of two or more cases of breast or ovarian cancer were also included.
Materials and methods
In the breast cancer clinic at Hôtel-Dieu Hospital, more than 500 women with breast cancer are under active follow-up. Patients were eligible for this study if they were French-Canadian and had been diagnosed with invasive breast cancer or DCIS at or before age 70. French-Canadian was self-defined or with reference to three or more French-Canadian grandparents. Women aged 51–70 years were eligible if they had two or more first- or second-degree relatives with breast or ovarian cancer. All women aged 50 or below were eligible, regardless of family history. Women were ineligible if the diagnosis occurred more than 10 years in the past. Potential study subjects were identified through medical chart review and were approached to participate in the study. Participating subjects signed a consent form and completed a questionnaire containing information on known and suspected risk factors for breast-cancer as well as socio-demographic information. A blood sample was collected for genetic testing. A detailed family history was taken from each participating subject. For each patient the pathology report was reviewed and the cancer was classified as invasive only, invasive with DCIS component and DCIS alone. The estrogen receptor (ER) status was classified according to the pathology report (positive, negative, equivocal and missing).
Blood samples for BRCA genetic testing were sent to the laboratory of Dr Steven Narod (Toronto). Genomic DNA was extracted from peripheral blood leukocytes using the Gentra Puregene Blood Kit (Qiagen, Venlo, the Netherlands). Six founder mutations that are frequent among French-Canadians were tested Table 2. A multiplex polymerase chain reaction (PCR) was used to amplify and subsequently screen two mutations in BRCA1 (2953delGTA/insC and 3875delGTCT) and two in BRCA2 (3398delAAAAG and 8765delAG), previously described . A tetra-primer amplification refractory mutation system (ARMS) PCR  was used to detect the remaining BRCA1 C4446T and BRCA2 8085G>T mutations. All mutations found through the ARMS and multiplex PCR assays were confirmed by direct sequencing [BigDye Terminator v3.1 Cycle Sequencing Kit, and 3130XL Genetic Analyzer (Applied BioSystems, Canada)] according to manufacturer's instructions. Primers used for all assays and sequences are available upon request.
From 2004 to 2011, a total of 3250 women below age 70 were treated and followed for breast cancer at the Breast Cancer Centre of the CRCHUM. Of these, 1310 (43%) patients were diagnosed at age 50 or below and 1940 patients (57%) were diagnosed after age 50; 1154 of these women were eligible for the study and were invited to participate. Overall, 88% of all eligible patients agreed to participate, including 89% of the early-onset patients and 85% of the later-onset familial patients; 121 of the patients had DCIS and 972 patients had invasive cancer. Of the 972 invasive cancers, 64% had a DCIS component.
Genetic testing was completed on 1093 patients. In total, 56 mutations (5.1%) were detected, 43 in BRCA2 and 13 in BRCA1 (Table 1). A mutation was found in 5.3% of the early-onset cases and in 4.6% of the later-onset familial cases. Among the early-onset cases, the mutation prevalences were 14% in women diagnosed from 20 to 29, 9.1% in women diagnosed from 30 to 39 and 4.2% for women diagnosed from 40 to 49.
Two founder mutations accounted for 66% of the total number of mutations. The BRCA2 exon 20 8765delAG mutation was seen 27 times and the BRCA1 exon13 4446C>T mutation was seen 10 times (Table 2); 42 of 56 mutations (75%) were in patients diagnosed before 50 years of age and 14 of 56 mutations were found in patients diagnosed older than age 50.
The mutation prevalence was 3.3% for women with DCIS, 7.5% for women with invasive cancer only and 4.1% for women with mixed invasive/DCIS. DCIS (alone or in association with invasive cancer) was associated with the presence of a BRCA2 mutation [odds ratio (OR) = 3.3], but this did not reach statistical significance (p = 0.11; Table 3). The mutation prevalence was 5.2% for women with ER-positive breast cancer and was 5.0% for women with ER-negative breast cancer. 64% of the BRCA1 mutation carriers had ER-negative breast cancers and 90% of the BRCA2 carriers had ER-positive breast cancers (Table 4). Only 4 of 42 mutations found among women with an ER-positive breast cancer were in BRCA1 (9.5%).
|Mutation status||Invasive breast only||DCIS only||Invasive/DCIS mix|
|Total||BRCA1+||% positive||BRCA2+||% positive|
Among the 791 patients diagnosed at 50 years of age or below, 323 had a family history of two or more relatives with breast or ovarian cancer, 218 had a family history of a single relative with breast or ovarian cancer and 250 had no affected relative. The mutation prevalences in the three groups were 10.8% (35/323), 1.4% (3/218) and 2.0% (5/250), respectively.
We estimated the prevalence of mutations in French-Canadian women with invasive and in situ breast cancer. We restricted this study to women who were considered to be at elevated risk of carrying a mutation by virtue of early-onset disease or a positive family history. We included 121 cases of DCIS as well as invasive cancers. On the basis of the results of this study, we conclude that it is reasonable to offer genetic testing for the six founder mutations to all incident and prevalent cases of breast cancer for women in these risk groups. We studied prevalent and incident cases; we believe these results are relevant for women who have had a diagnosis of breast cancer in the past, as these women may be at increased risk for second primary breast cancer and for ovarian cancer. In addition, if they are found to be carriers, their sisters and daughters will also be at high risk of carrying a mutation. The unaffected relatives may be offered predictive testing with the goal of offering interventions to prevent cancers. Therefore, we included women with a past history of breast cancer in the last 10 years. Most other studies have concentrated on incident cases in order to avoid survivorship bias, but we are concerned that this focus may be too narrow in clinical terms and will result in the omission of many mutation carriers in the population who have survived breast cancer and who are under clinical follow-up.
The rationale for offering testing to all French-Canadian breast cancer patients, but not to all Quebecers, is not a consequence of the high frequency of mutations in the population, but rather is because of the low cost of testing for six founder mutations. The frequency of mutations in our sample (5.1%) is similar to the frequency of mutations in a recent study of 1628 unselected breast cancer patients of mixed ethnicities in Seattle (4.7%) . However, in the Seattle study, 96 women carried 1 of 66 different mutations and in our study 56 women carried 1 of 6 different mutations. Should the cost of full gene sequencing fall to below 100 dollars, it may be cost-effective to extend this recommendation to breast cancer patients of all ethnic groups. This is similar to the situation in Poland, where the mutations are not unusually common, but they are restricted to a few founder mutations  but different from the situation in Ashkenazi Jews  or in the Bahamas , where founder mutations are very common. In French-Canadian women, BRCA2 mutations outnumber BRCA1 mutations by a ratio of approximately 3:1; there is also a preponderance of BRCA2 over BRCA1 in Iceland  and in China , but in Poland  and the Bahamas , BRCA1 mutations predominate.
The mutation yield of women with breast cancer with one or no affected relative was 1.7% (8/468) and among women with two or more relatives was 10.8%. Similarly, the prevalence of BRCA1 mutations was also very low among women with ER-negative breast cancers (0.5%). Nevertheless, we do not recommend that testing be restricted to women with a strong family history or that women with ER + breast cancers be tested for BRCA2 mutations only, doing so would result in missing approximately 20% of the carriers in this population. Testing based on family history would require the surgeon or medical oncologist to obtain the relevant pedigree information and arrange the appropriate referral. We do not recommend selective testing based on the invasiveness or the ER status of the tumor because doing so would add a level of complexity to the testing process. Given the low cost of testing the founder panel (approximately 100 dollars per person) we believe that it is justifiable to perform 60 tests in order to identify one mutation carrier. We also believe that the total yield of mutation carriers in Quebec would be higher if the restrictions on testing were minimized in this clinical situation.
The current report extends our previous work from 564 tested cases to 1093 cases, with a mean age of diagnosis at 47.2. Using the selection criteria, the expected mutation prevalence is 5.1%. In many jurisdictions, such as Ontario, the minimum prevalence for testing is 10%, but in the French-Canadian population of Quebec, we recommend testing be offered at a much lower prevalence because the cost of testing is much lower. Similar considerations have been made for Jewish women  and for women from the Bahamas .
The majority of carriers (37/56; 66%) had one of two mutations. The BRCA2 8765delAG mutation accounted for 27 (48%) mutations and the BRCA1 4446C>T accounted for 10 (18%) mutations. These mutations have also accounted for a high percentage of mutations in other studies of French-Canadian families [12-14]. In a recent paper, these six mutations accounted for 28 of 37 mutations (76%) found in a panel of French-Canadian families with multiple cases of breast cancer . It is possible that other mutations may have been missed in the analysis as we have restricted our study to the most common mutations found in the French-Canadian population. Large genomic insertions or deletions were not included due to the rarity of these rearrangements in this population .
We included 121 cases of DCIS in our study. We confirm results from Myriad Genetics  that this the mutation prevalence rate is higher in this population (3.3%) than in the general population (0.25%), but the rate was lower than for women with invasive cancer (5.3%). However, it is notable that in the context of invasive breast cancer the presence of DCIS was associated with a lower prevalence rate (4.1% vs 7.5%). It is also notable that among cases of DCIS, or invasive cancer with a DCIS component, the proportion of mutations that were in BRCA2 was 86%, compared with 65% for cases of invasive cancer without DCIS. These differences might be due to random fluctuation (p = 0.11), but may also support a model whereby early neoplastic changes in breasts of BRCA1 carriers (but not BRCA2 carriers) progress rapidly to invasive cancer without an extensive in situ intermediary.
In Quebec, the family sizes traditionally have been large and as a result the mutation-positive families generally have several affected female relatives. The mean number of breast cancer cases in the 42 mutation-positive early-onset breast cancer families was 5.0 and in the 14 late-onset breast cancer families it was 5.0. This highlights the potential for cancer prevention if the mutation status of these women could be known beforehand. Our ability to prevent hereditary breast cancer would be enhanced if the first case of breast cancer in each family were to be tested and then mutation testing offered to all unaffected relatives. Those women found to be carriers could be offered effective preventive options.
This study was supported by the Canadian Breast-Cancer Research Alliance (CBCRA) and the Fonds de la recherche en santé du Québec (FRSQ). We would like to thank all our study subjects for their contribution, as well as Simone Montpetit for typing the manuscript and providing technical support.
- 1Canadian Cancer Society's Steering Committee on Cancer Statistics. Canadian cancer statistics 2011. Toronto, ON: Canadian Cancer Society; 2011.