The contribution of founder mutations in BRCA1 to breast and ovarian cancer in Lithuania
Steven A Narod, Womens College Research Institute, Toronto, Ontario, Canada.
Tel.: +1 416- 351- 3765;
fax: +1 416- 351- 3767;
Elsakov P, Kurtinaitis J, Petraitis S, Ostapenko V, Razumas M, Razumas T, Meskauskas R, Petrulis K, Luksite A, Lubiński J, Górski B, Narod SA, Gronwald J. The contribution of founder mutations in BRCA1 to breast and ovarian cancer in Lithuania.
We evaluated the prevalence of BRCA1 founder mutations in unselected cases of breast, ovarian and colon cancer from Lithuania. We identified a founder mutation (4153delA, 5382insC or C61G) in 6% of 235 unselected cases of breast cancer and in 19% of 43 unselected cases of ovarian cancer. Only one patient with a mutation was identified among 178 cases of colon cancer. No mutation was identified among 422 newborn controls. This data indicates that the genetic burden of breast and ovarian cancer attributable to BRCA1 mutations in Lithuania is very high and supports the recommendation that all cases of breast and ovarian cancer in Lithuania be offered genetic testing.
The Baltic states comprise three countries (Lithuania, Latvia and Estonia) with different languages and ethnic compositions. In two previous studies of families from Latvia with the hereditary breast-ovarian cancer syndrome, three recurrent mutations in BRCA1 were identified (4153delA, 5382insC and C61G) (1, 2). Each of these three mutations is also a founder mutation in the Polish population (3). These three mutations account for approximately 80–90% of all detectable BRCA1 and BRCA2 gene mutations in Latvia and in Poland (1, 2, 4). The contribution of BRCA1 mutations to the burden of colon cancer in Eastern Europe has not been well studied.
Previous studies of the Lithuanian population have concentrated on breast-ovarian cancer families. Gronwald et al. found a founder mutation in 9 of 13 familial breast/ovarian cancer families (5). Janavičius et al. studied 28 Lithuanian patients who were referred for familial breast/ovarian cancer (6); in total, eight mutations were found; a founder mutation was found in seven families. In these two studies, families were selected for family history; the contribution of these three founder mutations to the overall burden of breast cancer and ovarian cancer in Lithuania has not been studied. The aim of this study was to estimate the frequency of BRCA1 founder mutations in unselected series of breast, ovarian and colorectal cancer patients from Lithuania.
Material and methods
Breast cancer, ovarian and colorectal cancers cases were identified from patients treated at the Vilnius University Oncology Institute (VUOI) between 2007 and 2009. The VUOI treats patients from all regions of Lithuania, and treats approximately 45% of all patients treated for primary breast cancer in the country, as well as 30% of the ovarian cancer patients and 22% of the colorectal cancer patients. The patients were consecutive, newly diagnosed cases of cancer, unselected for age, sex, or family history. The patient study group included 235 breast cancer patients, 43 ovarian cancer patients and 178 colorectal cancer patients (91 male and 87 female). Each patient provided written informed consent to take part in the study. The reference pathologist reviewed a representative slide from each cancer to confirm the diagnosis. A family history was obtained through patient interview. Patient characteristics are presented in Table 1.
Table 1. Study group patients characteristics
|Age (median, range)||56 (28–90)||58 (30–76)||70 (30–86)|
To estimate the frequencies of these mutations in the Lithuanian population at large, we used 442 anonymous blood samples from newborn children (222 girls and 220 boys) collected in Vilnius Maternity Hospital between 2007 and 2009.
DNA was extracted from peripheral blood lymphocytes for cases and from cord blood from controls. Two BRCA1 mutations (4153delA and 5382 insC) were studied using ASA-PCR and one mutation (C61G) was detected with RFLP-PCR, as described elsewhere (3).
A BRCA1 mutation was detected in 14 of 235 (6%) unselected breast cancer cases, in 8 of 43 (18.6%) unselected ovarian cancer cases, and in 1 of 178 (0.6%) unselected colorectal cancer cases. No mutation was found among the 442 controls. The distribution of mutations is presented in Table 2. The 4153delA mutation accounted for seven of eight cases of hereditary ovarian cancer, but for only one-half of the cases of hereditary breast cancer. The median age of diagnosis of the 14 hereditary breast cancer cases was 44 years (range 31–60 years), compared to a median age of diagnosis of 54 years (range 27–86 years) for the cases without mutation. A mutation was found in 5 of 28 women (17.9%) diagnosed with breast cancer under the age of 40.
Table 2. The frequency of founder mutations in unselected series of cancer patients
|C61G||1||0.4||–|| ||–|| |
The median age of diagnosis of the eight hereditary ovarian cancer cases was 48 years (40–60 years), compared to median age of diagnosis of 57 years (28–74 years) for the group of cases without a mutation. The frequency distributions of the breast and ovarian cancer cases, according to mutation status and age of diagnosis are shown in Table 3.
Table 3. The results of distribution of breast, ovarian cancer cases according to age of affected by cancer
| ≤ 50 years||95||11 (11.6%)|
| ≤ 40 years||28||5 (17.9%)|
| ≤ 50 years||19||5 (26.3%)|
| ≤ 40 years||5||1 (20%)|
The family histories of the mutation-positive cases were reviewed. Among the 14 women with breast cancer and a BRCA1 mutation, two had a first-degree relative with breast cancer and two had a second-degree relative affected with breast cancer. Two of the breast cancer cases reported a first-degree relative with cancer of the female genital tract. In the group of eight ovarian cancer patients with a BRCA1 mutation, one woman had a first-degree relative with ovarian cancer and none had a relative affected with breast cancer.
We found that 6% of unselected cases of breast cancer and 19% of unselected cases of ovarian cancer in Lithuania carried a founder mutation in the BRCA1 gene. The majority of mutation-positive patients did not have a significant family history of breast or ovarian cancer; 71.4% of the breast cancer patients with a mutation and 87.5% of the ovarian cancer patients with a mutation had no first- or second-degree relative with breast or ovarian cancer.
The current study provides estimates of the total heritable proportion of breast and ovarian cancer in Lithuania and the distribution of the individual mutations in patients with breast or ovarian cancer. This distribution may not reflect the underlying distribution of mutations in the Lithuanian population at large. To the extent that the penetrance of one mutation exceeds that of another for a particular site of cancer, the mutation will be over-represented when patients with cancer at that site are ascertained. The data from our study suggest that the relative penetrance of breast versus ovarian cancer is higher for the 5382insC mutation than for the 4153delA mutation. This observation is in accordance with our earlier observation that the 4153delA mutation is a low-penetrance BRCA1 mutation in Poland (7). To estimate the relative frequencies of the different mutations in the underlying population, it is better to study newborn children (or male controls); however, because the mutations are individually rare, a very large control group is required. In our study, no mutation was seen in a control group of 442 individuals.
A high frequency of BRCA1 or BRCA2 founder mutations has been reported in unselected ovarian and breast cancer patients from Ashkenazi Jewish and Polish populations (3, 8, 9). Recently, very high mutation frequencies have been reported for two small isolated island populations, Greenland (10, 11) and the Bahamas (data not shown).
In this small series of unselected ovarian cancer patients, we found that 18.6% carried a BRCA1 founder mutation. In a similar study from Poland, one of the three mutations was present in 13.5% of ovarian cancer patients (12). This proportion is higher than the frequency of 12.6% for all BRCA1 and BRCA2 mutations in the ethnically mixed population of Ontario (13).
Our study does not provide support for the hypothesis that BRCA1 mutations contribute to the burden of colon cancer in Lithuania. If there is an effect on colon cancer risk, it is modest. It is possible that BRCA1 mutations other than these are pathogenic for colon cancer, or that in other ethnic groups with different genetic backgrounds, the penetrance of colon cancer is higher. A study of consecutive Israeli Jewish patients with colorectal cancer documented a slightly elevated risk of colorectal cancer in carriers of BRCA1/2 mutations (14). However, a study of another series of unselected Jewish patients with colorectal cancer did not confirm this association (15).
Lithuania is one of three Baltic countries; to our knowledge no BRCA1 mutation surveys have been conducted in Estonia but two papers have been published from Latvia. In 1999, Csokay et al. found that 8 of 23 patients (15 familial; 8 non-familial) carried one of the founder mutations studied here (1). In a larger study from 2005, Tikhomirova et al. found a mutation in 13 of 50 (26%) Latvian breast cancer patients diagnosed below the age of 50 (2). The results of the studies of the Latvian and Lithuanian patients are similar in terms of mutation distribution and suggest that the two populations are ethnically similar. It will be of interest to extend these results to Estonia.
In other Slavic countries, particularly in Russia and in Poland, the 5382insC mutation is very common (4, 16). Sokolenko et al. found this single mutation to be present in 4% of unilateral breast cancer patients, in 10% of bilateral cancer patients and in 6% of early-onset cases (diagnosed at or less than age 40) (16). 5382insC is also a founder mutation in Greece (17), Belarus (18), the Czech Republic (19) and Slovenia (20). It is the second most common BRCA1 mutation in the Ashkenazi Jewish population (8, 9). To our knowledge, the frequency of founder mutations of 6% of unselected breast cancer patients in Belarus exceeds the frequencies reported in the earlier studies.
There are several limitations to our study. The number of cases is relatively small and the results are based on 14 mutation-positive breast cancer cases and 8 mutation-positive ovarian cancer cases. We screened for only the three founder mutations and it is possible that other non-founder mutations were missed. The control group was too small (n = 422) to provide a precise estimate of the prevalence of these mutations in the underlying population, and as a result, it was not possible to generate relative risks or lifetime risks of cancer associated with each of the three mutations.
In populations with a small number of recurrent founder BRCA mutations, such as Lithuania, it is reasonable to offer to test all breast and ovarian cancer patients for the founder mutations. A high proportion of carriers of BRCA1 and BRCA2 mutations do not have a significant family history of cancer and it is unreliable to rely on family history in order to decide upon whom to test. There is evolving evidence that the type of treatment offered to a women with breast cancer and a BRCA1 mutation may differ from that for sporadic cancers (21).
Conflict of interest
None of the authors of this paper declares a conflict of interest.