|
|
Background Some features of breast cancer in women with a BRCA1 mutation suggest that hereditary breast cancer has a poor outcome. We conducted a national population-based study of Israeli women to determine the influence, if any, of a BRCA1 or a BRCA2 mutation on the prognosis in breast cancer.
Methods We obtained data on all incident cases of invasive breast cancer that were diagnosed from January 1, 1987, to December 31, 1988, and recorded in the Israel National Cancer Registry. We requested a paraffin-embedded tumor block or an unstained slide and the corresponding pathological and clinical records for all such cases. DNA extracted from the tumor specimens was analyzed for the three founder mutations in BRCA1 and BRCA2. For each subject, available pathological and oncologic records were reviewed.
Results We were able to retrieve a pathological sample from 1794 of 2514 subjects (71%). Among those women, we obtained medical records for 1545 (86%). A BRCA1 or BRCA2 mutation was identified in 10% of the women who were of Ashkenazi Jewish ancestry. The adjusted hazard ratios for death from breast cancer were not significantly different among mutation carriers and noncarriers (hazard ratio among BRCA1 carriers, 0.76; 95% confidence interval [CI], 0.45 to 1.30; P=0.31; hazard ratio among BRCA2 carriers, 1.31; 95% CI, 0.80 to 2.15; P=0.28). Among women who were treated with chemotherapy, the hazard ratio for death among BRCA1 carriers was 0.48 (95% CI, 0.19 to 1.21; P=0.12).
Conclusions Breast cancer–specific rates of death among Israeli women are similar for carriers of a BRCA founder mutation and noncarriers.
Approximately 60% of Israeli Jews are Ashkenazi, and others are Sephardi, Iraqi, or of mixed ancestry; there are also Muslim, Christian, and Druze residents. Two mutations in BRCA1 (185delAG and 5382insC) and one in BRCA2 (6174delT) are common in Ashkenazi women. Approximately 2% of all Ashkenazi women2,3 and 12% of Ashkenazi women with breast cancer4 carry a mutation in one of these two genes. Much of the information that correlates mutational status with breast-cancer type has come from studies of the Ashkenazi population, but little is known about the influence of a BRCA1 or BRCA2 mutation on the natural history of breast cancer or the response to treatment. BRCA1-associated breast cancers often occur in younger women, and such tumors are high grade and lack estrogen receptors.5,6 All these features are associated with a poor prognosis. However, the evidence concerning the effect of a BRCA1 or BRCA2 mutation on the prognosis is inconsistent.7,8,9,10,11,12,13,14 To clarify the influence of these mutations on the outcome in breast cancer, we evaluated 10-year survival rates in a national cohort of Israeli women whose breast cancer had been diagnosed in 1987 or 1988.
Methods
Study Patients
We identified women who had received the diagnosis of invasive breast cancer in Israel between January 1, 1987, and December 31, 1988, through the records of the Israel National Cancer Registry. This registry records the woman's name, birth date, and national identification number and information on the treating hospital and selected tumor characteristics (morphologic type and stage). It records information for more than 95% of women with breast tumors diagnosed in Israel.15 Women identified through the registry were linked to the Israel Population Register to ascertain their probable ethnic origin. Information in the register includes whether the woman is Jewish, the country of birth for women born abroad, and the places of birth of both parents for women born in Israel. Probable ethnic group was assigned for each woman with breast cancer. Jewish women born in Europe, North America, South America, and South Africa were defined as Ashkenazi unless a different ethnic background was stated in the medical records. Jews born in Asia (mostly in the Arab countries of the Middle East, Iran, and India) and North Africa were considered to be Sephardi. Iraqi Jews were assigned their own category. Jews born in Israel were classified according to the country of birth of their parents, when data were available. If parental origin was not available or if the parents had different ethnic backgrounds, the ancestry was assigned as Ashkenazi.
For each eligible subject, we requested one or more paraffin-embedded tumor blocks or unstained slides and the corresponding pathological and oncologic records from the treating hospital. The records were reviewed for pathological stage (including the size of the tumor and the number of lymph nodes that were evaluated and were malignant), histological analysis of the tumor, and estrogen-receptor status. Information on grade was not routinely recorded during the study years, and estrogen-receptor data were available for only 55% of subjects. We extracted information from the medical records for clinical stage, treatment, and outcome. The main outcomes of interest were overall survival and death from breast cancer. Other outcomes, such as local recurrence, contralateral breast cancer, and other primary cancers, were also recorded (see the table in the Supplementary Appendix, available with the full text of this article at www.nejm.org). The stage at diagnosis was assigned to each subject on the basis of all the information available to the investigators from the pathological reports and the clinical-oncology records. We extracted data on the cause of death from the official records of the Central Bureau of Statistics of Israel, which was coded on the basis of nomenclature from the International Classification of Diseases, 9th Revision.
The study was approved by the institutional review board at Carmel Medical Center in Haifa, Israel, and by the Israeli Ministry of Health. These investigations were conducted according to guidelines for the protection of human subjects issued by the Carmel Medical Center, the Israel National Cancer Registry, the National Cancer Institute in Bethesda, Maryland, and the University of Toronto.
We identified 2729 subjects with invasive breast cancer who were treated at 24 hospitals. Two hospitals had discarded their tumor samples before the study began, and the 215 subjects at these hospitals were therefore excluded. Among the remaining 2514 subjects, we retrieved a pathological sample from 1794 (71%). DNA was extracted from the paraffin blocks and was analyzed for the three founder mutations in the BRCA1 and BRCA2 genes. We used a method described previously16 to assign mutational status to all 1794 women for whom samples were available. We were able to retrieve medical records for 1545 of these women (86%).
Statistical Analysis
The prevalence of BRCA1 and BRCA2 mutations was determined for each ethnic subgroup as the proportion of mutation-positive subjects among all subjects. We compared demographic and clinical features between BRCA carriers and noncarriers with the use of the chi-square test, Fisher's exact test, Mann–Whitney test, or t-test, as appropriate. All women in the study were followed for 16 years for survival and incidence of other cancers. Information on cause of death was available for deaths that occurred during the first 10 years of follow-up. The 10-year overall survival rates were calculated as the number of women who were alive after 10 years divided by the number of women in the cohort. These proportions were compared between groups with the use of either the chi-square test or Fisher's exact test. Both the time to death and the time to death from a breast-cancer–specific cause were compared with the use of Kaplan–Meier curves, and statistical significance was evaluated with the use of the log-rank test. The Cox proportional-hazards model was used to estimate the hazard ratio for carriers of the mutation, as compared with noncarriers, after adjustment for age, tumor size, lymph-node status, and metastases.
Results
Mutation Rates According to Ethnic Group
Tumor samples were obtained from 1794 subjects who had received the diagnosis of breast cancer in Israel in 1987 or 1988. All but 46 of the subjects were Jewish women. Among these 1748 Jewish women, 1102 were Ashkenazi, 418 were Sephardi, 87 were Iraqi, and 141 were of other, unknown, or mixed ancestry. Table 1 shows the characteristics of the tumors in the various ethnic subgroups. All tumor samples were tested for the presence of the three founder mutations. No mutation was detected in any of the 46 women who were not Jewish. A mutation was detected in 135 samples from 1748 Jewish women (8%); these included 105 of 1102 samples from Ashkenazi women (10%), 6 of 87 from Iraqi women (7%), and 4 of 418 from Sephardic women (1%). The ethnic origin of an additional 20 of 131 Jewish carriers of the mutations (15%) who were born in Israel could not be assigned (and were listed as unclassified). For 10 Jewish subjects, there was no information on either country of birth or ethnic group.
|
In the newly defined Ashkenazi group, 131 women were carriers; 76 had BRCA1 mutations, 52 had BRCA2 mutations, and 3 had mutations in both genes and thus were excluded from all analyses. Among the 336 Ashkenazi women whose breast cancer had been diagnosed before the age of 50 years, the prevalence of the mutation was 18%; among the 984 women who were 50 years of age or older when they received the diagnosis, the prevalence was 7%.
Characteristics of Breast Cancers
We used pathological reports from all participating pathology institutes to compare the tumors in the 131 Ashkenazi women who had a mutation with the tumors in the 1189 Ashkenazi women for whom no mutation was identified (Table 2). The mean age at diagnosis was lower in BRCA1 carriers than in noncarriers (52 vs. 61 years, P<0.001). In contrast to the other two groups, most of the BRCA1 carriers received the diagnosis of breast cancer before the age of 50 years. Tumor size and lymph-node status were similar in the three groups. BRCA1 carriers were less likely to have estrogen-receptor–positive tumors than were noncarriers (24% vs. 65%, P<0.001). The difference between carriers and noncarriers was marked for tumors that were reported to be strongly estrogen-receptor–positive (0% of BRCA1 carriers vs. 31% of noncarriers, P<0.001).
|
All subjects were followed for a minimum of 10 years. The crude 10-year survival rates were 51% for Jewish women and 52% for non-Jewish women, but there were only 46 women in the latter category (Table 1). Among Jewish women, 10-year survival rates were similar for Ashkenazi women (50%) and Sephardi women (51%). Among the Ashkenazi subjects, the presence of a BRCA mutation did not significantly influence the overall 10-year survival rate: 49% for BRCA1 carriers, 48% for BRCA2 carriers, and 51% for noncarriers (Table 3). Among the mutation carriers, 89% of deaths were due to cancer, as compared with 72% of the deaths among noncarriers (P=0.002). The adjusted hazard ratio for death from any cause did not differ significantly between carriers of BRCA1 mutations and noncarriers (hazard ratio, 1.13; 95% confidence interval [CI], 0.78 to 1.66; P=0.52) or between carriers of BRCA2 mutations and noncarriers (hazard ratio, 1.20; 95% CI, 0.77 to 1.86; P=0.42) (Table 3).
|
The rate of survival at 10 years was 67% for BRCA1 carriers, 56% for BRCA2 carriers, and 67% for noncarriers (P=0.25 for the comparison between carriers of either mutation and noncarriers by the log-rank test with 2 degrees of freedom). Of 24 deaths from breast cancer among BRCA1 carriers, 21 (88%) occurred within 5 years after diagnosis, as compared with 17 of 22 such deaths (77%) in BRCA2 carriers and 244 of 361 (68%) in noncarriers (P=0.04 for the comparison between carriers of either mutation and noncarriers). Among women who died from breast cancer, the median time to death was 46 months for noncarriers, 37 months for BRCA1 carriers, and 48 months for BRCA2 carriers. The adjusted hazard ratios for death from breast cancer did not differ significantly for carriers of BRCA mutations as compared with noncarriers (hazard ratio for BRCA1, 0.76; 95% CI, 0.45 to 1.30; P=0.31; hazard ratio for BRCA2, 1.31; 95% CI, 0.80 to 2.15; P=0.28) (Table 3).
Among women who did not receive chemotherapy after surgery, the 10-year survival rates were 76% for BRCA1 carriers and 74% for noncarriers (hazard ratio, 0.93; 95% CI, 0.43 to 2.02; P=0.86). Among women who received chemotherapy, the 10-year survival rates were 71% for BRCA1 carriers and 46% for noncarriers (hazard ratio, 0.48; 95% CI, 0.19 to 1.21; P=0.12). The interaction between BRCA1 mutation status and chemotherapy was significant only for overall survival (P=0.02). The survival of BRCA2 carriers was similar to that for noncarriers, with or without chemotherapy (Table 4).
|
|
To determine whether the subjects we studied were representative of all Israeli women with breast cancer, we reviewed registry information for the 935 subjects for whom a tumor specimen was not obtained. The mean age at diagnosis for nonparticipants in the study was slightly older than that for participants (60 vs. 59 years, P=0.03). Nonparticipants were more likely than participants to be deceased (69% vs. 60%, P<0.001). Ashkenazi origin was more frequent among participants than among nonparticipants (62% vs. 53%), whereas Israeli-born Jews of uncertain ethnic background were more frequent among nonparticipants than among participants (17% vs. 7%, P<0.001 for comparisons regarding uncertain ethnic background).
Discussion
By examining registry information, we identified the majority of women who had received the diagnosis of breast cancer in Israel in 1987 or 1988 and identified all deaths attributable to breast cancer in the 10-year period after diagnosis. The study was designed to estimate survival rates for women with a BRCA1 or BRCA2 mutation, as compared with women without a detected mutation. These estimates are important to women with a BRCA mutation who face a decision between preventive surgery and intensive surveillance.
In our study, the size of the tumor was not a predictor of the probability of death, and lymph-node status was a predictor of borderline significance. Survival rates were similar for women who had a tumor that was 2 cm or less and those with a tumor of more than 2 cm. Of 20 women with a BRCA1 mutation and a node-negative tumor of 2 cm or less, 3 died of breast cancer within 10 years after diagnosis. The influence of a BRCA1 or BRCA2 mutation on the outcome of chemotherapy was not statistically significant.
Among several previous studies of breast-cancer survivorship among BRCA carriers, three are notable for robust study designs and relatively large samples. Two of these studies involved Jewish women,12,13 and the third involved Dutch women.14 Robson and colleagues12 used a design similar to ours. They typed tumor blocks from 496 Jewish women with breast cancer who had undergone breast-conserving surgery between 1980 and 1995. Of these women, 56 were found to carry a founder mutation (11%). After a median survival of 10 years, the rate of breast-cancer–specific survival was worse for carriers of BRCA1 mutations than for noncarriers (62% vs. 86%, P<0.001). This difference was noted regardless of whether women had undergone chemotherapy. Our study is similar to that of Robson et al. except that breast-conserving surgery was not a criterion for selection of our subjects. The 10-year survival rates for the women in our study were lower than those reported by Robson et al.,12 whose subjects received the diagnosis of breast cancer in New York or Montreal. Less than half of the Israeli women in our study had tumors of 2 cm or less, as compared with 76% of women in the study by Robson et al., yet we observed a survival disadvantage for carriers of BRCA1 mutations in women with small tumors.
In a similar study, El-Tamer et al.13 tested tumor blocks from 487 Jewish women who were treated at the New York Presbyterian Hospital. They found that 51 of the women were mutation carriers (10%), but there were no significant differences in either overall or breast-cancer–specific survival between carriers and noncarriers. The average follow-up period was short (50 months), and the cohort was not divided according to tumor size or the use of chemotherapy. Furthermore, 10 of 31 subjects who carried a BRCA mutation had ductal carcinoma in situ.
In a report from a clinic in Rotterdam for patients at high risk for breast cancer,14 the breast-cancer–specific survival times for carriers of BRCA1 mutations and noncarriers were similar. The average age at diagnosis in the Dutch study was 39 years; in our study, we observed effects related to the age at diagnosis. Furthermore, hazard ratios in the Dutch study were adjusted for tumor grade and estrogen-receptor status. Such adjustment may be appropriate for testing whether the effect of BRCA1 mutational status on prognosis is independent of other prognostic factors but can be misleading for women with BRCA1 mutations who are trying to choose between prophylactic surgery and surveillance.
In our study, subjects were not selected for age, family history, tumor stage, or treatment; these factors may influence survival rates but obscure important underlying differences. There are several other strengths of our study. We were successful in obtaining tumor specimens from 22 of 24 treating hospitals (71% of eligible subjects), and our cohort appeared to be representative of all Israeli women with breast cancer. We were able to genotype all the specimens we had obtained. All subjects had received the diagnosis of breast cancer within a 2-year period and were followed for a minimum of 10 years. Treatment regimens were relatively homogeneous and were chosen without knowledge of mutational status. Breast cancer was diagnosed in all women in our study before the introduction of mammography17,18 or genetic testing in Israel, and special surveillance was not offered to women with a family history of cancer. No study subject was aware of her genetic status.
Our study has a number of limitations. Tumor grade and estrogen-receptor status were not routinely recorded. We tested for only the three founder mutations. It is possible that we misclassified some hereditary cases, but the great majority of Jewish women with BRCA1 and BRCA2 mutations have one of these three mutations.19,20 Since we identified only 135 mutation carriers, the subgroup analyses relied on a small number of subjects. The Israel National Cancer Registry captures outcome data on almost all patients with cancer who are treated in Israel, but recurrences and deaths in women after they have emigrated from Israel are not recorded. We had 16-year follow-up data on mortality and incident cancers, but information on the cause of death was available from the Central Bureau of Statistics only for deaths that occurred before 2000. Inherent to the type of study we conducted are the lack of central verification of the pathological diagnosis and incomplete data concerning estrogen-receptor status, information that was retrieved from medical records when it was available. Although we did not see any clear survival differences in association with the BRCA mutations we sought, potentially important differences emerged when we analyzed the data according to tumor size and the use or nonuse of chemotherapy.
Supported by a grant (CA8097) from the National Cancer Institute.
No potential conflict of interest relevant to this article was reported.
We thank the following oncology and pathology units for their help in providing samples from and records of study participants: Ziv Hospital (G. Zaidan), Western Galilee Hospital (Y. Cohen, M. Ben Shahar), Rambam Hospital (O. Ben Izhak, A. Kuten), Bnei Zion Hospital (I. Mislevich, L. Lev), Carmel Hospital (Y. Sova, M. Steiner), Hillel Yafe Hospital (G. Groisman), Meir Hospital (J. Bernhaim, Y. Shapira), Beilinson Hospital (E. Okun, A. Sulkes), Hasharon Hospital (B. Klein), Ichilov Hospital (B. Lifshits, M. Inbar), Shiba Hospital (Y. Kopelovich, R. Katan, B. Kaufman), Wolfson Hospital (I. Avinoach, Y. Brenner), Kaplan Hospital (M. Husser, A. Shani), Soroka Hospital (J Goldstien, Y. Cohen), Shaare Zedek Hospital (E. Rosenman, A. Gabizon), Bikur Holim Hospital (G. Lachovizki), Hadassah Hospital (T. Peretz), Haemek Hospital (Y. Honigman), and Dr. M. Bar-Chana of the Israel National Cancer Registry.
Source Information
From Clalit Health Services, National Cancer Control Center and Department of Community Medicine and Epidemiology, Carmel Medical Center and B. Rappaport Faculty of Medicine, Technion, Haifa, Israel (G.R., S.B.-N., O.B.-G., N.B.-J., H.S.R.); and the Women's College Research Institute, University of Toronto, Toronto (S.Z., S.A.N.).
Address reprint requests to Dr. Rennert at the Clalit Health Services, National Cancer Control Center, Carmel Medical Center, Haifa 34362, Israel, or at rennert{at}tx.technion.ac.il.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Prognosis of Breast Cancer in Carriers of BRCA1 and BRCA2 Mutations
Goodwin P. J., Phillips K.-A., West D. W., Roukos D. H., Rennert G., Narod S.
Extract |
Full Text |
PDF
N Engl J Med 2007;
357:1555-1556, Oct 11, 2007.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
