Hereditary ovarian cancer in Poland

Authors


Abstract

There is increasing evidence that hereditary factors play a greater role in ovarian cancer than in any of the other common cancers of adulthood. This is attributable, to a large extent, to a high frequency of mutations in the BRCA1 or BRCA2 genes. In Poland, 3 common founder mutations in BRCA1 account for the majority of families with identified BRCA mutations. Our study was conducted in order to estimate the prevalence of any of 3 founder BRCA1 mutations (5382insC, C61G and 4153delA) in 364 unselected women with ovarian cancer, and among 177 women with ovarian cancer and a family history of breast or ovarian cancer. A mutation was identified in 49 out of 364 unselected women with ovarian cancer (13.5%) and in 58 of 177 women with familial ovarian cancer (32.8%). The majority of women with ovarian cancer and a BRCA1 mutation have no family history of breast or ovarian cancer. The high frequency of BRCA1 mutations in Polish women with ovarian cancer supports the recommendation that all Polish women with ovarian cancer should be offered testing for genetic susceptibility, and that counseling services be made available to them and to their relatives. It is important that mutation surveys be conducted in other countries prior to the introduction of national genetic screening programs. © 2003 Wiley-Liss, Inc.

Ovarian cancer is the fourth most common cause of cancer among women in Poland and is the third leading cause of death from cancer.1 There are approximately 3,000 new cases of ovarian cancer and 2,000 deaths form ovarian cancer in Poland annually.1 Approximately 10% of cases of ovarian cancer in North America are hereditary and are due to deleterious mutations in the BRCA1 or BRCA2 genes.2 However, it is expected that the hereditary fraction of ovarian cancer will vary from country to country for several reasons, including the age distributions of the populations and the relative frequencies of BRCA1 and BRCA2 mutations. Three common founder mutations in BRCA1 (5382insC, 4153delA and C61G) are prevalent in several countries in eastern Europe, including Russia,3 Hungary,4, 5 Latvia,6 Austria7 and Poland.8, 9, 10 These 3 mutations account for approximately 80% of BRCA mutations in breast-ovarian cancer families in Poland,10 but it is not yet known to what extent these 3 mutations contribute to the overall burden of ovarian cancer in Poland. It is important that population surveys be conducted prior to the establishment of national screening programs for hereditary ovarian cancer. The purpose of our study is to establish the importance of BRCA1 founder mutations as a contributor to the ovarian cancer burden in Poland.

MATERIAL AND METHODS

Patient populations

Two groups of patients were studied. In the first group, 368 consecutive newly diagnosed patients with invasive ovarian cancer, who were treated at 1 of 2 institutions in Szczecin, Poland, between January 1999 and December 2001, were approached to participate. 364 women (99%) agreed to participate. Of these, 305 women were treated at the Szczecin Regional Oncological Hospital, and 59 cases were treated at the Clinic of Surgical Gynecology and Gynecological Oncology. Almost all patients with ovarian cancer diagnosed in Szczecin and more than 80% of the cases diagnosed in the Polish region of Western Pomerania are treated in these 2 hospitals.

The second study group consisted of women with familial ovarian cancer who were referred to 1 of 16 clinical cancer genetics centers throughout Poland between 1999 and 2001. These centers were established in 1998 as a national network with support from the Polish Ministry of Health for the purpose of coordinating cancer genetics services. Eligible women included those with invasive ovarian cancer and at least one first- or second-degree relative with ovarian cancer diagnosed at any age or with early-onset breast cancer (diagnosed at age 50 or below). The 177 patients came from 13 cities throughout Poland including Bydgoszcz (3 patients), Gdańsk (15), Gliwice (24), Kielce (34), Kraków (5), Lublin (7), Łódź (3), Olsztyn (30), Opole (7), Poznań (37), Rzeszów (3), Wrocław (6) and Zielona Góra (3) (Fig. 1, Table I).

Figure 1.

Map of Poland indicating the source of the familial ovarian cancer cases used in our study.

Table I. Prevalence of Mutations by Age of Diagnosis
Age groupNumber of casesNumber of mutations% positive
≤501122421.4
51–601021615.7
60+15096.0
Total3644913.5

A detailed pedigree was taken from all patients in both groups, which included the sites and ages-of-onset of all cancers in male and first-degree relatives. Patients in both groups were offered genetic testing for 3 founder BRCA1 mutations. Information was not available on specific ethnic groups, but the great majority of subjects were white and identified themselves to be of Polish ethnicity. The study protocol was approved by the Institutional Review Board of the Pomeranian Medical University and all patients provided written informed consent for genetic testing.

Mutation detection technique

Mutation analysis for 2 common Polish mutations (4153 delA and 5328 insC) was carried out by a multiplex specific polymerase chain reaction (PCR) assay. The third mutation (C61G) generates a novel restriction enzyme site in exon 5. This mutation is detected after digesting amplified DNA with Ava II (MBI Fermentas). To visualize the different BRCA1 alleles, the PCR products were subjected to electrophoresis in a 1.5% agarose gel, stained with ethidium bromide. Each sample run includes a positive and a negative control. When a mutation was identified, a second blood sample was taken from the patient to confirm the positive finding. Samples taken from patients in Gdansk, Poznan, Gliwice and Bydgoszcz were done locally, using identical methods with diagnostic kits sent from the study center in Szczecin. All other samples were tested in the Szczecin laboratory.

To evaluate the accuracy of the multiplex technique, the results of 30 samples representing each of the 3 mutations and normal DNA samples were compared. The results of the multiplex PCR assay and the direct DNA sequencing were 100% concordant.

RESULTS

Among 364 women with unselected ovarian cancer mutations there were 49 mutations identified (13.5%) including 5382insC (28 times), C61G (13 times) and 4153delA (8 times). Greater than half of the identified mutations were 5382insC and this single mutation was present in 7.7% of the cases.

The mean age of diagnosis of the 49 cases with BRCA1 mutations was 51.4 years; of the 315 cases without mutations, it was 56.2 years (p < 0.01). The prevalence of BRCA1 mutations declined with increasing age of diagnosis (Table II).

Table II. Prevalence of BRCA1 Mutations in Familial Ovarian Cancer Cases Throughout Poland
 Number testedNumber with mutations% positive
Bydgoscz33100
Gdansk151173.3
Gliwice24937.5
Kielce341029.4
Krakow5360
Lublin7457.1
Lodz33100
Olsztyn30620
Opole7114.3
Poznan37718.9
Rzeszow300
Szczecin462043.5
Wroclaw6116.7
Zielona Gora300
Total2237835.0

The family histories of the 364 unselected ovarian cancer cases from Szczecin were reviewed and these women were classified into those with and those without a significant family history. A significant family history is defined as a first- or second-degree relative with ovarian cancer (any age) or with breast cancer diagnosed before age 50. A total of 46 of the 364 patients (12.6%) were classified as familial. A mutation was found in 43.5% of the 46 patients with familial ovarian cancer and in 9.1% of the 318 patients with nonfamilial ovarian cancer. Twenty of the 49 mutation-positive patients (41%) were familial.

The prevalence of BRCA mutations among the 177 women with familial ovarian cancer from elsewhere in Poland was slightly lower than that of familial cases from Szczecin. A BRCA1 mutation was found in 58 of 177 women (32.8%) with familial cancer referred to 1 of 16 genetic centers throughout Poland, including 5382insC (43 patients), C61G (11 patients) and 4153delA (4 patients). There was considerable variation observed in the frequency of mutations in the different centers throughout Poland (Table II), but for most of the centers the samples sizes were small. The mutation prevalence among familial cases was 35% overall when the cases from Szczecin and elsewhere were combined.

DISCUSSION

We observed a high prevalence of BRCA1 mutations in familial and nonfamilial ovarian cancer cases in Poland. One of 3 founder mutations was identified in 13.6% of unselected cases of ovarian cancer. To our knowledge, this prevalence figure exceeds that for all other ethnic groups studied to date, with the exception of Ashkenazi Jews (Table III). Our estimate is similar to that reported in a previous smaller study from Hungary, where a BRCA1 mutation was found in 10 out of 90 unselected women with ovarian cancer (11.1%).5 It will be important to establish if the mutant frequency is high in other eastern European countries as well.

Table III. Frequency of BRCA1 Mutations in Series of Consecutive Ovarian Cancers from Different Populations
PopulationFrequency of mutationsReference
Ashkenazi Jewish35.5% (67/189)11
 27.5% (57/208)12
Polish13.5% (49/364)Our study
Pakistani13.3% (16/120)13
Hungarian11.1% (10/90)5
USA8.6% (10/116)14
Canadian7.6% (39/515)2
French Canadian5.1% (5/99)15
Finnish4.7% (11/233)16
Japanese4.0% (3/76)17
British3.4% (12/355)18
Norwegian3.3% (2/60)19

There are several possible reasons that may explain the differences in the mutation frequencies observed in the various populations. Several studies are based on only a small number of mutations, and variation may be due to sampling error. Second, different techniques with varying sensitivities have been employed. For example, the sensitivity of single-strand conformation polymorphism (SSCP) appears to be inferior to that of denaturing gradient gel electrophoresis (DGGE).20 Some studies restricted screening to certain regions of the BRCA1 gene,2, 14, 16, 17, 18 and in other studies only founder mutations5, 11, 12, 13, 15, 19 were evaluated.

There are also nontechnical reasons for the observed variation. Populations with high mutation frequencies are also those with prominent founder effects, and founder effects are the most likely reason for variation in mutation rates between countries. It is also possible that in countries with low background rates of ovarian cancer, the hereditary proportion is relatively high (even though the absolute incidence of hereditary ovarian cancer is not increased). Population demographic structure is also important. For example, in countries with high fertility and mortality rates, a relatively high fraction of the population will be young. There will be, as a result, proportionately more young women with ovarian cancer. A higher hereditary proportion is expected in these countries because of the association between age-of-onset and the prevalence of mutations.

In our study, we only tested for the 3 common founder mutations. It is possible that we missed mutations in BRCA2, or elsewhere in BRCA1. Therefore it is likely that the true hereditary fraction exceeds 13.6%. These 3 mutations accounted for 80% of the mutations found in a series of Polish breast-ovarian cancer families in a previous study.10 Adjusting for a sensitivity of 80%, then 17% of unselected cases of ovarian cancer should be due to BRCA mutations. Our study supports the recommendation that all women with ovarian cancer in Poland be tested for these 3 mutations. The mutation yield is high and the cost of testing and counseling a patient is below 100 dollars. It is possible that additional testing is warranted for selected patients, including a full screen of BRCA1 and BRCA2. Comprehensive testing is expensive, however, and the efficiency of this approach has not yet been evaluated. It is expected that a much greater yield of mutations will be achieved, and at a far lower cost, by screening all women with ovarian cancer for founder mutations than by doing a complete gene scan on the subset of familial cases.

There are several strengths to our study. The great majority of cases which were diagnosed in the geographic region of Szczecin were included in the unselected series and patient compliance was virtually 100%. We used hospital clinics to ascertain ovarian cancer patients and thereby avoided the selection biases which are inherent in registry-based studies. For example, in the Ontario population-based study of Risch et al.,2 only 63% of eligible patients were included. Nineteen percent of the patients had died prior to ascertainment and others were missed due to loss of follow-up, in addition to patient and physician refusal. In contrast, virtually all eligible patients were included in our study.

We also used a standardized form to record pedigree details of all first- and second-degree relatives that allowed us to classify probands into those with familial and nonfamilial cancer. A national program has been established in Poland that provides genetic counseling and testing services to the entire population at no cost to the participants. In contrast, in many of the clinic-based studies in North America,2, 14, 15 probands were self-selected according to their access to resources that allowed them to undergo genetic testing. There are no private hospitals in Poland and access to care should not be a determinant of genetic testing. North America is ethnically mixed, whereas the Polish population is relatively homogeneous.

There are several weaknesses in our study as well. As noted above, only 3 mutations were evaluated and our estimate of the mutation prevalence is likely to be low. Second, the number of families from each of the contributing centers was relatively low. Therefore, it is not possible to precisely compare the mutation frequencies to the specific mutation spectra for the different geographical regions.

In conclusion, there is a high prevalence of founder BRCA1 mutations in unselected cases of ovarian cancer in Poland. We recommend that all women with invasive ovarian cancer be offered genetic testing and that genetic counseling facilities be available to all healthy female relatives of mutation carriers, who can then be offered testing and information regarding prevention options. Preventive strategies offered to women with BRCA1 mutations include prophylactic oophorectomy21, 22 and tubal ligation.23 Oral contraceptive use of 5 years or more has been shown to reduce the risk of hereditary ovarian cancer by 60% and not to increase the risk of breast cancer if used after 25.24, 25 With the integration and dissemination of genetic screening and counseling services, it is reasonable to expect the hereditary proportion of ovarian cancer to decline in the near future.

Acknowledgements

The authors thank Mrs. Anita Giermakowska and Mrs. Lidia Jabłońska for performing DNA tests.

Ancillary