BACKGROUND: Germline mutations in the BRCA1 and BRCA2 genes are associated with an increased risk of breast cancer. Whether women with breast cancer who have inherited mutations in these genes have a different outcome after breast conservation therapy than women with “ sporadic” cancer is unresolved. Consequently, we compared the outcomes after breast conservation therapy in Ashkenazi women with or without germline mutations in BRCA1 and/or BRCA2 (hereafter called BRCA). METHODS: We studied 305 women of Ashkenazi Jewish descent undergoing breast-conserving treatment for 329 invasive breast cancers. We reviewed their clinical records, retrieved their archival tissue samples, and tested those samples for the founder mutations BRCA1 185delAG, BRCA1 5382insC, and BRCA2 6174delT. Genetic results were linked to clinical data and outcomes by univariate and multivariate analyses. All Pvalues are two-sided. RESULTS: We detected mutations in BRCA genes in 28 of 305 women. Women with BRCA mutations were more likely to be diagnosed with cancer before the age of 50 years (P<.001) and to have lymph node involvement (P = .04). Ipsilateral breast tumor recurrence was more common in women with BRCA mutations, although this did not reach statistical significance (relative risk [RR] = 1.79; 95% confidence interval [CI] = 0.64-5.03). Women with mutations were more likely to develop contralateral breast cancer (RR = 3.50; 95% CI = 1.78-8.74; P = .001). Distant disease-free survival was shorter in women with mutations (66.2% versus 84.3% at 10 years; P = .05), as was breast cancer-specific survival (71.9% versus 87.2% at 10 years; P = .02). Tumor stage and nodal status, but not mutation status, were predictive of distant disease-free and breast cancer-specific survival in multivariate analysis. CONCLUSIONS: Women with BRCA founder mutations are at increased risk for breast cancer-related events after breast conservation. However, mutation status is not an independent predictor of survival and should not influence decisions regarding adjuvant therapy. The increased contralateral breast cancer risk in women heterozygous for BRCA mutations mandates careful surveillance.

Five percent to 10% of breast cancers arise as the result of an inherited predisposition (1). Although the isolation of BRCA1 (2) and BRCA2 (3) genes has allowed the identification of individuals who are at increased risk for breast cancer, the optimal treatment of women who develop hereditary breast cancer remains undefined. One question that remains unanswered is whether breast conservation therapy is appropriate for such women. To investigate the importance of germline BRCA1 and/or BRCA2 (hereafter referred to as BRCA) status in determining the outcome after breast conservation therapy, we performed an anonymized, retrospective, cohort study of women treated at a single institution.

Patients and Methods

Patients

Review of databases maintained by the Departments of Radiation Oncology and Surgery identified 1248 patients who received breast-conserving treatment at the Memorial Sloan-Kettering Cancer Center for breast cancer diagnosed during the period from January 1, 1980, through December 31, 1990. Of the 415 women in the databases who identified themselves at the time of hospital registration as being Jewish, 393 were treated for early-stage invasive breast cancer. Archival pathology material and follow-up were available for 314 women who underwent treatment for 338 cancers.

Data Collection

The study was performed in an anonymized manner according to guidelines regarding the conduct of genetic research on stored tissue samples (4). After confirmation of the diagnosis of invasive breast cancer, archival pathologic material was accessioned from hospital tissue banks, and clinical records were reviewed. A limited amount of clinical information was extracted, including age at diagnosis, histologic description of the tumor, tumor size, extent of lymph node involvement, estrogen receptor status (if available), and basic treatment information (if available). To prevent a unique data pattern from identifying individual patients after anonymization, clinical information for each variable was recorded as a range rather than as a discrete value. Time to development of local or distant disease was recorded, as was the time to development of contralateral breast cancer, if any. The time to last follow-up and vital status were determined by reviewing clinical records and contacting the patient, her next of kin, or her primary physician by telephone. The tissue specimen and clinical information for each patient were associated through a unique study number. Samples were made anonymous by removal of all connection between the unique study number and patient identifiers before the linkage of genetic test results to clinical information. This anonymization obviated the need for specific informed consent for this study (4). The design and procedures of the study were approved by the Institutional Review Board of the Memorial Sloan-Kettering Cancer Center.

Genetic Analysis

Genomic DNA was isolated from archival tissue specimens by standard protocols (5). In nearly all cases, normal lymph node tissue from axillary dissection was used for analysis. All specimens were analyzed for the presence of three BRCA founder mutations common in individuals of Ashkenazi descent (BRCA1 185delAG, BRCA1 5382insC, and BRCA2 6174delT) (6-8). Polymerase chain reaction (PCR) products encompassing the site of each mutation were generated with the primers 5′-TCTGCTCTTCGCGTTGAAGAA-3′ and 5′-CACTCTTGTGCTGACTTACCA-3′ for BRCA1 185delAG (90-base-pair [ bp] product), 5′-CAGCATGATTTTGAAGTCAG-3′ and 5′-AGGGAGCTTTACCTTTCTGTC-3′ for BRCA1 5382insC (99-bp product), and 5′-GGGAAGCTTCATAAGTCAGTC-3′ and 5′-TTTGTAATGAAGCATCTGATACC-3′ for BRCA2 6174delT (97-bp product). Radiolabeled products were produced with a forward primer, end-labeled with [γ -33P]adenosine triphosphate, and then visualized by denaturing polyacrylamide gel electrophoresis followed by autoradiography as previously described in detail (9,10). The detection of characteristic variant bands indicated the presence of the mutation being studied. In two previous studies using this technique to screen archival ovarian cancer tissue for mutations (10,11), all 129 samples with variant bands were confirmed by sequencing to have the corresponding BRCA mutations, with no discordant results. In the current study, all mutations were confirmed by an independent PCR amplification from the corresponding DNA sample and repeat analysis. There were no discordant results upon repeat analysis.

Statistical Analysis

DNA samples from nine patients failed to yield PCR products, despite repeated attempts. Therefore, the final study group consisted of 305 women undergoing treatment for 329 invasive breast cancers. The median follow-up from first diagnosis for surviving patients is 124 months.

Freedom from ipsilateral breast tumor recurrence was defined as the time from initial diagnosis to diagnosis of cancer in the treated breast or last follow-up. Because of the difficulty in retrospectively differentiating true recurrence from an ipsilateral second primary breast cancer, all ipsilateral breast cancers were considered ipsilateral breast cancer recurrences. This end point was analyzed for all 329 invasive cancers. All other end points were analyzed in 305 patients. Distant disease-free survival was calculated as the time from first conservatively treated breast cancer to detection of any distant recurrence or last follow-up. The development of metachronous contralateral breast cancer was recorded as a separate end point. Women who had previously been diagnosed with breast cancer and had undergone a contralateral mastectomy before the breast cancer diagnosis for which they underwent breast conservation therapy were excluded from the analysis of metachronous contralateral breast cancer (n = 7). Breast cancer-specific survival was defined as the time from the diagnosis of the first conservatively treated cancer to death from breast cancer or last follow-up. Overall survival was the time from diagnosis of first conservatively treated cancer to death from any cause.

Patients were censored if they had not experienced the end point of interest at the time of last follow-up. Some patients died without experiencing any breast cancer-related end point (ipsilateral breast tumor recurrence, distant relapse, contralateral breast cancer, or death due to breast cancer). Hence, all survival probabilities presented in this report were calculated by use of a competing risk approach, and the comparison was done by use of the method of Gray (12). All P values were calculated with two-sided tests.

Prognostic variables considered were age at diagnosis (<50 years versus ⩾50 years), mutation status (founder mutation present versus no mutation detected), tumor (T) stage [ T1 versus T2; American Joint Committee on Cancer (13)], and lymph node (N) status [N0 versus N1; American Joint Committee on Cancer (13)]. Hormone receptor status was not evaluated because this information was not available for 166 specimens. Details of adjuvant chemotherapy were not available for 53 (16.1%) of 329 cancers. Information regarding the use of adjuvant tamoxifen was not available for 59 (17.9%) of 329 cancers. Because of the missing data points and potential biases in treatment assignment, adjuvant treatment was not evaluated as a separate prognostic variable.

Multivariate models were fit by using the Cox proportional hazards model. If only one variable was statistically significant in univariate analysis, a Cox model was fit by using this single variable to generate risk ratios and confidence intervals (CIs). Patients with missing information regarding tumor stage (17 patients) or lymph node status (14 patients) were excluded from the multivariate analysis. For the identification of factors associated with ipsilateral breast tumor recurrence, values for one or the other of these variables were missing for 31 of 329 breast carcinomas, leaving data from 298 breast carcinomas for multivariate analysis. For breast cancer-specific survival and overall survival, information was missing on 31 of 305 patients; thus, multivariate analysis for these end points was based on data from 274 patients.

Results

Mutation Analysis and Clinical Characteristics

BRCA mutations were identified in 28 (9.2%) of 305 women (19 with BRCA1 185delAG, two with BRCA1 5382insC, six with BRCA2 6174delT, and one with both BRCA1 185delAG and BRCA2 6174delT). These 28 women underwent conservative treatment for 35 invasive breast cancers. Therefore, of the 329 breast cancers in the series, 35 (10.6%) were associated with mutations (24 with BRCA1 185delAG, two with BRCA1 5382insC, eight with BRCA2 6174delT, and one with BRCA1 185delAG and BRCA2 6174delT). The clinical characteristics of the 329 cancers in women with or without founder mutations are shown in Table 1. Women with BRCA mutations were more likely to be diagnosed before the age of 50 years (62.9% versus 26.2%; P<.001) and to have axillary lymph node involvement (48.6% versus 32.3%; P = .04). There was no statistically significant difference between the groups with respect to tumor stage. All but two cancers associated with BRCA mutations were of the infiltrating ductal histology, and none were tubular, colloid, or papillary carcinomas. When hormone receptor status was known, breast cancers associated with BRCA mutations were more likely to be negative for estrogen receptor (14 of 18 versus 54 of 145; P = .001). Among those women for whom details of adjuvant therapy were available, systemic chemotherapy was used in 15 (51.7%) of 29 episodes of cancer associated with BRCA mutations and 78 (31.6%) of 247 episodes in women without mutations (P = .03). Among women for whom details of systemic therapy were known, adjuvant tamoxifen was given after local treatment of seven (24.1%) of 29 episodes of breast cancers associated with BRCA mutations and 92 (38.2%) of 241 women without mutations (P = .14).

Outcomes

The probability of ipsilateral breast cancer recurrence for patients with BRCA founder mutations was 14.9% at 5 years and 22.0% at 10 years, compared with 4.5% and 6.9%, respectively, for patients without these mutations (P = .25) (Table 2). Distant disease-free survival among women with BRCA mutations was 74.1% at 5 years and 66.2% at 10 years, compared with 90.5% at 5 years and 84.3% at 10 years among women without mutations (P = .05).

Women with BRCA mutations were more likely to die of breast cancer than women without these mutations. Breast cancer-specific survival at 5 years and 10 years was 85.3% and 71.9%, respectively, for women with mutations, compared with 95.9% and 87.2%, respectively, for women without mutations (P = .02; Fig. 1). Overall survival at 5 years and 10 years was 82.0% and 66.0%, respectively, for women with mutations and 93.0% and 80.6%, respectively, for women without mutations (P = .05).

Women with germline BRCA mutations were at increased risk for contralateral breast cancer. BRCA mutations were detected in nine (21.4%) of 42 women with bilateral disease. The risk of contralateral breast cancer at 5 years and 10 years after breast-conservation therapy was 14.8% and 27.0%, respectively, among women with mutations, compared with 3.7% and 9.5%, respectively, among women without mutations (P = .002).

Univariate Survival Analyses

Age at diagnosis, tumor stage, lymph node stage, and mutation status were analyzed individually for their prognostic association with each end point (Table 2). Age at diagnosis was the only variable associated with time to ipsilateral breast tumor recurrence. The relative risk (RR) of ipsilateral breast tumor recurrence associated with the presence of a germline BRCA mutation was 1.79 (95% CI = 0.64-5.03). Mutation status, tumor stage, and lymph node involvement were statistically significantly associated with distant disease-free survival, breast cancer-specific survival, and overall survival. Breast cancer-specific survival was also associated with age at diagnosis (P = .05). Mutation status was the only variable statistically significantly associated with contralateral breast cancer risk.

Multivariate Survival Analysis

Variables with univariate statistical significance were entered into the multivariate regression analysis by use of a Cox proportional hazards model. Because the only variable attaining univariate statistical significance for ipsilateral breast tumor recurrence was age at diagnosis, this was the only factor retained in the Cox model. Younger patients (<50 years old at diagnosis) had an RR for ipsilateral breast tumor recurrence of 2.51 (95% CI = 1.22-5.16; P = .01).

Mutation status was the only variable associated with contralateral breast cancer risk in the univariate analysis. In the Cox model, the presence of a mutation was associated with a hazard ratio for contralateral breast cancer of 3.50 (95% CI = 1.78-8.74; P = .001).

Mutation status, lymph node status, and tumor stage were the three variables examined in a multivariate Cox model for distant disease-free survival and breast cancer-specific survival. Age was also examined in the model for breast cancer-specific survival (Table 3, A). Although tumor stage and lymph node status retained prognostic significance, the hazard ratio associated with mutation status was not statistically significant for either end point. Age at diagnosis was not statistically significantly associated with breast cancer-specific survival in the Cox model. Therefore, the final model for both end points includes only tumor stage and lymph node status (Table 3, B).

Discussion

Surveillance and prevention options for women who are BRCA heterozygotes have been described in detail (14,15), but little is known about the management of established breast cancer in individuals with BRCA mutations. In particular, outcomes after breast conservation therapy in such women have not been described in detail. Young age at diagnosis of breast cancer has been shown to be a risk factor for ipsilateral breast tumor recurrence in several series (16-21). It is possible that an increased risk of local recurrence among women with germline BRCA1 or BRCA2 mutations could be contributory. However, in two retrospective cohort studies of women with invasive breast cancer treated by breast conservation (22, 23) and one case-control study (18), family history of breast cancer was not a statistically significant predictor of ipsilateral breast tumor relapse. BRCA genotyping was not performed in these studies, and an adverse outcome among women with mutations cannot be excluded. One report of outcomes after breast conservation in women who are BRCA heterozygotes has suggested an increased risk of ipsilateral breast tumor recurrence (24), although this has not been confirmed in other series (25-27). Because these analyses are based on prevalent case ascertainments, they may be subject to a survivorship bias.

The impact of germline BRCA status on survival is uncertain. Initial studies of patients with breast cancer associated with BRCA1 mutations identified through linkage analysis (28-30) suggested an improved survival. The positive effect of germline BRCA1 mutations was not confirmed in later studies of patients from familial cancer clinics. Such patients were compared with cancer registry (26) or population-based (31) control subjects. Studies of patients with early-onset breast cancer have yielded divergent results, indicating either similar (32) or inferior (33) survival for women who are BRCA heterozygotes. The published studies have a potential bias because they describe the outcome of selected women who consented to undergo germline genetic testing at some time after diagnosis of their breast cancer. An adverse effect of BRCA status on outcome may be obscured in such studies by the exclusion of women who were diagnosed in the same period as the subjects of the study but who did not undergo genetic testing because of early death. In support of this hypothesis, one series of unselected patients with breast cancer described a shorter survival among patients with BRCA1 mutations (34).

In this study, women with BRCA founder mutations had a greater risk of ipsilateral breast tumor recurrence than women without these mutations. The difference did not reach statistical significance, possibly because of the limited number of events observed. The risk of ipsilateral breast tumor recurrence in BRCA heterozygotes does not appear to be great enough to contraindicate breast conservation in this group. Larger studies will be required, however, to definitively resolve whether germline BRCA status independently influences ipsilateral breast tumor recurrence risk in young women.

In this study of unselected Ashkenazi women with breast cancer, women with BRCA founder mutations were at increased risk for distant recurrence, contralateral breast cancer, and breast cancer-related death when compared with women who did not carry the mutations. Only tumor stage and lymph node status retained statistical significance in the multivariate analysis of prognostic factors for distant disease-free and breast cancer-specific survival, although there was a statistically nonsignificant trend for mutation status to have an independent effect on breast cancer-specific survival. This result suggests that women with BRCA mutations in this series were at increased risk for death due to breast cancer because of presentation with more advanced disease. In this regard, axillary lymph node involvement was more common in women with BRCA mutations than in women without such mutations, as was extensive lymph node involvement (metastases in four or more lymph nodes). Alternatively, the observed worse outcome may result from adverse biologic features that have been described in BRCA-associated breast cancers. Such features include high histologic grade and proliferation rates (28,40), lack of hormone receptor expression (43-45). The differences in outcome observed in the current study could result from an excess of such features in breast cancer arising in women with BRCA mutations. There is no evidence that differences in adjuvant treatment contributed to the observed differences in outcome because women with BRCA mutations were more likely to receive chemotherapy and were not statistically significantly less likely to receive tamoxifen. Furthermore, women receiving adjuvant therapy did not have an improved survival compared with women not receiving such therapy, whether or not a mutation was present (data not shown).

Women with germline BRCA mutations were at a nearly fourfold increased risk for contralateral breast cancer. Nearly 30% of surviving women developed metachronous contralateral breast cancer 10 years after diagnosis. Comparison of contralateral cancer rates in heterozygotes treated with mastectomy alone will be required to discern whether radiation scatter contributes to the risk of contralateral breast cancer among BRCA heterozygotes undergoing breast conservation therapy.

A potential limitation of this study relates to the restriction of genetic testing to the three recurring mutations described among individuals of Ashkenazi descent. Women were selected for inclusion in this study on the basis of their self-declared religious preference. Because the great majority of American Jews are of Ashkenazi descent, the use of religious preference as a surrogate for ethnicity is unlikely to result in substantial misclassification. Unique BRCA mutations other than those analyzed in this study have been described in this population (46-48). We would not expect that the frequency of such mutations in the Jewish patients with breast cancer would be any greater than that in the population of patients with breast cancer as a whole. Because the frequency of BRCA1 mutations in all women in the United States who have breast cancer has been reported to be only 2.6% (49), the number of women with unique BRCA mutations included in the control group is likely to be small. The presence of such women among the control subjects would tend to lessen observed differences between the two groups.

Although the women in this series were members of a specific ethnic group, the results are likely to be valid for BRCA heterozygotes in the general population. To our knowledge, there are no data to suggest that women of Ashkenazi descent have a different clinical outcome than women of other ethnic groups after treatment for breast cancer. Moreover, to our knowledge, there are presently no data indicating that the outcome among BRCA heterozygotes is gene specific or allele specific. The breast cancer risks associated with BRCA1 and BRCA2 mutations appear to be similar (50,51), and the gene products interact in the same biologic pathway (52). Large collaborative trials will be required to achieve the sample size necessary to address potential differences in outcome conferred by specific BRCA alleles.

Our study indicates that women with breast cancer associated with BRCA mutations experience a different clinical outcome after breast conservation therapy than women without germline mutations. Women with mutations are more likely to experience distant relapse and to die of breast cancer. The risk of ipsilateral breast tumor recurrence in such women is also increased, but larger studies will be necessary to determine whether this effect is independent of age. Mutation status was not an independent predictor of outcome, suggesting that the shorter survival of women with cancers associated with BRCA mutations results from an association with known negative prognostic factors, such as axillary lymph node involvement. The presymptomatic identification of women with germline BRCA mutations and the implementation of intensified surveillance strategies may mitigate the prognostic impact of such mutations by facilitating cancer detection at the earliest most curable stage.

Table 1.

Clinical characteristics of conservatively treated cancers (n = 329)

Characteristic
 
Carcinomas in women with mutations (%) (n = 35)
 
Carcinomas in women with no mutation detected (%) (n = 294)
 
Age at diagnosis, y  
 <50 22 (62.9) 77 (26.2) 
 ⩾50 13 (37.1) 217 (73.8) 
Histology 
 Infiltrating ductal 33 (94.3) 243 (82.7) 
 Medullary 1 (2.9) 4 (1.4) 
 Other ductal (e.g., colloid or tubular) 0 (0.0) 23 (7.8) 
 Infiltrating lobular 1 (2.9) 24 (8.2) 
Tumor (T) stage* 
 T1 23 (65.7) 220 (74.8) 
 T2 8 (22.9) 61 (20.7) 
 Unknown 4 (11.4) 13 (4.4) 
Estrogen receptor status 
 Positive 4 (11.4) 91 (31.0) 
 Negative 14 (40.0) 54 (18.4) 
 Unknown 17 (48.6) 149 (50.7) 
Involved axillary lymph nodes, No. 
 0 16 (45.7) 182 (61.9) 
 1-3 10 (28.6) 74 (25.2) 
 ⩾4 7 (20.0) 21 (7.1) 
 Unknown 2 (5.7) 17 (5.8) 
Adjuvant therapy 
 Chemotherapy 15 (42.9) 78 (26.5) 
 Tamoxifen 7 (20.0) 92 (31.3) 
Characteristic
 
Carcinomas in women with mutations (%) (n = 35)
 
Carcinomas in women with no mutation detected (%) (n = 294)
 
Age at diagnosis, y  
 <50 22 (62.9) 77 (26.2) 
 ⩾50 13 (37.1) 217 (73.8) 
Histology 
 Infiltrating ductal 33 (94.3) 243 (82.7) 
 Medullary 1 (2.9) 4 (1.4) 
 Other ductal (e.g., colloid or tubular) 0 (0.0) 23 (7.8) 
 Infiltrating lobular 1 (2.9) 24 (8.2) 
Tumor (T) stage* 
 T1 23 (65.7) 220 (74.8) 
 T2 8 (22.9) 61 (20.7) 
 Unknown 4 (11.4) 13 (4.4) 
Estrogen receptor status 
 Positive 4 (11.4) 91 (31.0) 
 Negative 14 (40.0) 54 (18.4) 
 Unknown 17 (48.6) 149 (50.7) 
Involved axillary lymph nodes, No. 
 0 16 (45.7) 182 (61.9) 
 1-3 10 (28.6) 74 (25.2) 
 ⩾4 7 (20.0) 21 (7.1) 
 Unknown 2 (5.7) 17 (5.8) 
Adjuvant therapy 
 Chemotherapy 15 (42.9) 78 (26.5) 
 Tamoxifen 7 (20.0) 92 (31.3) 
*

Staging system is the American Joint Committee on Cancer system (13).

Details of adjuvant chemotherapy were not available for 53 cancers (six with mutations and 47 without); details of adjuvant tamoxifen were not available for 59 cancers (six with mutations and 53 without). Percentages use total number of episodes as denominator, whether or not treatment details are known.

Table 2.

Univariate analysis of clinical variables, restricted to patients for whom the variables are known

Variable
 
Ipsilateral breast tumor recurrence
 
Contralateral disease
 
Distant relapse
 
Death from breast cancer
 
No. of cases
 
No. of events
 
Two- sided P
 
No. of cases
 
No. of events
 
Two- sided P
 
No. of cases
 
No. of events
 
Two- sided P
 
No. of cases
 
No. of events
 
Two- sided P
 
Mutation 
 Yes 35 .25 27 .002 27 .05 28 .02 
 No 294 25  270 26  277 49  277 35 
Tumor (T) stage* 
 T1 243 23 .80 216 28 .13 221 30 <.001 222 20 <.001 
 T2 69  64  66 21  66 17 
Lymph node (N) stage* 
 N0 198 19 .86 175 20 .94 180 23 <.001 181 15 <.001 
 N1 112 11  110 12  110 33  110 26 
Age, y 
 <50 99 16 .01 94 14 .28 93 23 .13 94 19 .05 
 ⩾50 230 14  203 20  211 35  211 24 
Variable
 
Ipsilateral breast tumor recurrence
 
Contralateral disease
 
Distant relapse
 
Death from breast cancer
 
No. of cases
 
No. of events
 
Two- sided P
 
No. of cases
 
No. of events
 
Two- sided P
 
No. of cases
 
No. of events
 
Two- sided P
 
No. of cases
 
No. of events
 
Two- sided P
 
Mutation 
 Yes 35 .25 27 .002 27 .05 28 .02 
 No 294 25  270 26  277 49  277 35 
Tumor (T) stage* 
 T1 243 23 .80 216 28 .13 221 30 <.001 222 20 <.001 
 T2 69  64  66 21  66 17 
Lymph node (N) stage* 
 N0 198 19 .86 175 20 .94 180 23 <.001 181 15 <.001 
 N1 112 11  110 12  110 33  110 26 
Age, y 
 <50 99 16 .01 94 14 .28 93 23 .13 94 19 .05 
 ⩾50 230 14  203 20  211 35  211 24 
*

Staging system is the American Joint Committee on Cancer system (13).

Table 3.

Multivariate Cox analysis of factors influencing distant disease-free and breast cancer-specific survival


 
Distant disease-free survival
 
Breast cancer-specific survival
 
Estimate
 
P*
 
Relative risk (95% CI)
 
Estimate
 
P*
 
Relative risk (95% CI)
 
A) Initial Cox model including all factors statistically significant in univariate analysis 
Mutation status 0.372 .41 1.45 (0.60-3.49) 0.732 .14 2.08 (0.79-5.44) 
Tumor stage 0.796 .01 2.22 (1.24-3.97) 0.902 .01 2.47 (1.24-4.90) 
Lymph node stage 0.366 .03 1.44 (1.03-2.02) 0.697 .05 2.01 (1.00-4.04) 
Age at diagnosis — — — 0.011 .87 1.01 (0.88-1.16) 
B) Final Cox model of factors influencing distant disease-free and breast cancer-specific survival 
Tumor stage 0.792 .01 2.21 (1.23-3.96) 0.906 .01 2.47 (1.25-4.89) 
Lymph node stage 0.394 .02 1.48 (1.07-2.06) 0.785 .02 2.19 (1.11-4.33) 

 
Distant disease-free survival
 
Breast cancer-specific survival
 
Estimate
 
P*
 
Relative risk (95% CI)
 
Estimate
 
P*
 
Relative risk (95% CI)
 
A) Initial Cox model including all factors statistically significant in univariate analysis 
Mutation status 0.372 .41 1.45 (0.60-3.49) 0.732 .14 2.08 (0.79-5.44) 
Tumor stage 0.796 .01 2.22 (1.24-3.97) 0.902 .01 2.47 (1.24-4.90) 
Lymph node stage 0.366 .03 1.44 (1.03-2.02) 0.697 .05 2.01 (1.00-4.04) 
Age at diagnosis — — — 0.011 .87 1.01 (0.88-1.16) 
B) Final Cox model of factors influencing distant disease-free and breast cancer-specific survival 
Tumor stage 0.792 .01 2.21 (1.23-3.96) 0.906 .01 2.47 (1.25-4.89) 
Lymph node stage 0.394 .02 1.48 (1.07-2.06) 0.785 .02 2.19 (1.11-4.33) 
*

All P values are two-sided.

CI = confidence interval.

Staging system is the American Joint Committee on Cancer system (13).

Fig. 1.

Breast cancer-specific survival according to germline status (two-sided P = .02). At time 0, there were 28 patients with BRCA1 and/or BRCA2 (hereafter referred to as BRCA) mutations at risk and 277 patients without BRCA mutations at risk. At 2 years after diagnosis, there were 27 patients with BRCA mutations at risk and 273 patients without BRCA mutations at risk; breast cancer-specific survival was 96.4% (95% confidence interval [CI] = 77.2-99.5) for patients with BRCA mutations and 99.3% (95% CI = 97.16-99.8) for patients without BRCA mutations. At 5 years, there were 23 patients with BRCA mutations at risk and 251 patients without BRCA mutations at risk; breast cancer-specific survival was 85.3% (95% CI = 65.3-94.2) for patients with BRCA mutations and 95.9% (95% CI = 92.7-97.7) for patients without BRCA mutations. At 10 years, there were 16 patients with BRCA mutations at risk and 130 patients without BRCA mutations at risk; breast cancer-specific survival was 71.9% (95% CI = 49.6-85.7) for patients with BRCA mutations and 87.2% (95% CI = 82.2-90.9) for patients without BRCA mutations.

Fig. 1.

Breast cancer-specific survival according to germline status (two-sided P = .02). At time 0, there were 28 patients with BRCA1 and/or BRCA2 (hereafter referred to as BRCA) mutations at risk and 277 patients without BRCA mutations at risk. At 2 years after diagnosis, there were 27 patients with BRCA mutations at risk and 273 patients without BRCA mutations at risk; breast cancer-specific survival was 96.4% (95% confidence interval [CI] = 77.2-99.5) for patients with BRCA mutations and 99.3% (95% CI = 97.16-99.8) for patients without BRCA mutations. At 5 years, there were 23 patients with BRCA mutations at risk and 251 patients without BRCA mutations at risk; breast cancer-specific survival was 85.3% (95% CI = 65.3-94.2) for patients with BRCA mutations and 95.9% (95% CI = 92.7-97.7) for patients without BRCA mutations. At 10 years, there were 16 patients with BRCA mutations at risk and 130 patients without BRCA mutations at risk; breast cancer-specific survival was 71.9% (95% CI = 49.6-85.7) for patients with BRCA mutations and 87.2% (95% CI = 82.2-90.9) for patients without BRCA mutations.

Supported in part by the Society of the Memorial Sloan-Kettering Cancer Center. M. Robson is supported in part by an American Cancer Society Physicians' Research Training Award (PRTA-38).

We acknowledge the technical assistance of Ms. Nita Arroyo. We are also indebted to Colin Begg for assistance with study design and to Joan Marks.

(1)
Claus EB, Schildkraut JM, Thompson WD, Risch NJ. The genetic attributable risk of breast and ovarian cancer.
Cancer
 
1996
;
77
:
2318
-24.
(2)
Miki Y, Swensen J, Shattuck-Eidens D, Futreal PA, Harshman K, Tavtigian S, et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1.
Science
 
1994
;
266
:
66
-71.
(3)
Wooster R, Bignell G, Lancaster J, Swift S, Seal S, Mangion J, et al. Identification of the breast cancer susceptibility gene BRCA2.
Nature
 
1995
;
378
:
789
-92.
(4)
ASHG report. Statement on informed consent for genetic research. The American Society of Human Genetics.
Am J Hum Genet
 
1996
;
59
:
471
-4.
(5)
Strauss WB. Preparation of genomic DNA from mammalian tissue. In: Ausubel FM, Brent R, Kingston RE, editors. Current protocols in molecular biology. Vol 1. New York (NY): John Wiley & Sons; 1998. p. 2.2.1-2.2.3.
(6)
Oddoux C, Struewing JP, Clayton CM, Neuhausen S, Brody LC, Kaback M, et al. The carrier frequency of the BRCA2 6174delT mutation among Ashkenazi Jewish individuals is approximately 1%.
Nat Genet
 
1996
;
14
:
188
-90.
(7)
Roa BB, Boyd AA, Volcik K, Richards CS. Ashkenazi Jewish population frequencies for common mutations in BRCA1 and BRCA2.
Nat Genet
 
1996
;
14
:
185
-7.
(8)
Struewing JP, Abeliovich D, Peretz T, Avishai N, Kaback MM, Collins FS, et al. The carrier frequency of the BRCA1 185delAG mutation is approximately 1 percent in Ashkenazi Jewish individuals.
Nat Genet
 
1995
;
11
:
198
-200.
(9)
van der Heijden O, Chiu HC, Park TC, Takahashi H, LiVolsi VA, Risinger JI, et al. Allelotype analysis of uterine leiomyoma: localization of a potential tumor suppressor gene to a 4-cM region of chromosome 7q.
Mol Carcinog
 
1998
;
23
:
243
-7.
(10)
Rhei E, Bogomolniy F, Federici MG, Maresco DL, Offit K, Robson ME, et al. Molecular genetic characterization of BRCA1- and BRCA2-linked hereditary ovarian cancers.
Cancer Res
 
1998
;
58
:
3193
-6.
(11)
Boyd J, Sonoda Y, Federici MG, Bogomolniy F, Rhei E, Maresco DL, et al. Clinical and pathologic features of hereditary ovarian cancers associated with germline mutations in BRCA1 or BRCA2 [abstract].
Gynecol Oncol
 
1999
;
72
:
444
.
(12)
Gray RJ. A class of k-sample tests for comparing the cumulative incidence of a competing risk.
Ann Stat
 
1988
;
16
:
1141
-54.
(13)
Beahrs O, Henson D, Hutter R, editors. American Joint Committee on Cancer. Manual for staging of cancer. 4th ed. Philadelphia (PA): Lippincott; 1992.
(14)
Burke W, Daly M, Garber J, Botkin J, Kahn MJ, Lynch P, et al. Recommendations for follow-up care of individuals with an inherited predisposition to cancer. II. BRCA1 and BRCA2. Cancer Genetics Studies Consortium.
JAMA
 
1997
;
277
:
997
-1003.
(15)
Eisinger F, Alby N, Bremond A, Dauplat J, Espie M, Janiaud P, et al. Recommendations for medical management of hereditary breast and ovarian cancer: the French National Ad Hoc Committee.
Ann Oncol
 
1998
;
9
:
939
-50.
(16)
Fourquet A, Campana F, Zafrani B, Mosseri V, Vielh P, Durand JC, et al. Prognostic factors of breast recurrence in the conservative management of early breast cancer: a 25-year follow-up.
Int J Radiat Oncol Biol Phys
 
1989
;
17
:
719
-25.
(17)
Fowble BL, Schultz DJ, Overmoyer B, Solin LJ, Fox K, Jardines L, et al. The influence of young age on outcome in early stage breast cancer.
Int J Radiat Oncol Biol Phys
 
1994
;
30
:
23
-33.
(18)
Harrold EV, Turner BC, Matloff ET, Pathare P, Beinfield M, McKhann C, et al. Local recurrence in the conservatively treated breast cancer patient: a correlation with age and family history.
Cancer J Sci Am
 
1998
;
4
:
302
-7.
(19)
Kurtz JM, Spitalier JM, Amalric R, Brandone H, Ayme Y, Bressac C, et al. Mammary recurrences in women younger than forty.
Int J Radiat Oncol Biol Phys
 
1988
;
15
:
271
-6.
(20)
Kurtz JM, Jacquemier J, Amalric R, Brandone H, Ayme Y, Hans D, et al. Why are local recurrences after breast-conserving therapy more frequent in younger patients?
J Clin Oncol
 
1990
;
8
:
591
-8.
(21)
Recht A, Connolly JL, Schnitt SJ, Silver B, Rose MA, Love S, et al. The effect of young age on tumor recurrence in the treated breast after conservative surgery and radiotherapy.
Int J Radiat Oncol Biol Phys
 
1988
;
14
:
3
-10.
(22)
Chabner E, Nixon A, Gelman R, Hetelekidis S, Recht A, Bornstein B, et al. Family history and treatment outcome in young women after breast-conserving surgery and radiation therapy for early-stage breast cancer.
J Clin Oncol
 
1998
;
16
:
2045
-51.
(23)
Haas JA, Schultz DJ, Peterson ME, Solin LJ. An analysis of age and family history on outcome after breast-conservation treatment: the University of Pennsylvania experience.
Cancer J Sci Am
 
1998
;
4
:
308
-15.
(24)
Turner B, Harrold E, Matloff E, Smith T, Gumbs AA, Beinfield M, et al. BRCA1/BRCA2 germline mutations in locally recurrent breast cancer patients after lumpectomy and radiation therapy: implications for breast-conserving management in patients with BRCA1/BRCA2 mutations.
J Clin Oncol
 
1999
;
17
:
3017
-24.
(25)
Robson M, Levin D, Brown K, Haas B, Hampel H, Schulz C, et al. Results of breast conservation therapy (BCT) for invasive breast cancer in women with BRCA mutations [abstract].
Proc ASCO
 
1998
;
17
:
548a
.
(26)
Verhoog LC, Brekelmans CT, Seynaeve C, van den Bosch LM, Dahmen G, van Geel AN, et al. Survival and tumour characteristics of breast-cancer patients with germline mutations of BRCA1.
Lancet
 
1998
;
351
:
316
-21.
(27)
Pierce L, Strawderman M, Narod S, Olivotto I, Eisen A, Dawson L, et al. No deleterious effects of radiotherapy in women who are heterozygote for a BRCA-1 or BRCA-2 mutation following breast conservation therapy [abstract].
Proc ASCO
 
1999
;
18
:
86a
.
(28)
Marcus JN, Watson P, Page DL, Narod SA, Lenoir GM, Tonin P, et al. Hereditary breast cancer: pathobiology, prognosis, and BRCA1 and BRCA2 gene linkage.
Cancer
 
1996
;
77
:
697
-709.
(29)
Porter DE, Dixon M, Smyth E, Steel CM. Breast cancer survival in BRCA1 carriers.
Lancet
 
1993
;
341
:
184
-5.
(30)
Porter DE, Cohen BB, Wallace MR, Smyth E, Chetty U, Dixon JM, et al. Breast cancer incidence, penetrance and survival in probable carriers of BRCA1 gene mutation in families linked to BRCA1 on chromosome 17q12-21.
Br J Surg
 
1994
;
81
:
1512
-5.
(31)
Johannsson OT, Ranstam J, Borg A, Olsson H. Survival of BRCA1 breast and ovarian cancer patients: a population-based study from southern Sweden.
J Clin Oncol
 
1998
;
16
:
397
-404.
(32)
Robson M, Gilewski T, Haas B, Levin D, Borgen P, Rajan P, et al. BRCA-associated breast cancer in young women.
J Clin Oncol
 
1998
;
16
:
1642
-9.
(33)
Ansquer Y, Gautier C, Fourquet A, Asselain B, Stoppa-Lyonnet D. Survival in early-onset BRCA1 breast-cancer patients. Institut Curie Breast Cancer Group.
Lancet
 
1998
;
352
:
541
.
(34)
Foulkes WD, Wong N, Brunet JS, Begin LR, Zhang JC, Martinez JJ, et al. Germ-line BRCA1 mutation is an adverse prognostic factor in Ashkenazi Jewish women with breast cancer.
Clin Cancer Res
 
1997
;
3
:
2465
-9.
(35)
Breast Cancer Linkage Consortium. Pathology of familial breast cancer: differences between breast cancers in carriers of BRCA1 or BRCA2 mutations and sporadic cases.
Lancet
 
1997
;
349
:
1505
-10.
(36)
Eisinger F, Stoppa-Lyonnet D, Longy M, Kerangueven F, Noguchi T, Bailly C, et al. Germ line mutation at BRCA1 affects the histoprognostic grade in hereditary breast cancer.
Cancer Res
 
1996
;
56
:
471
-4.
(37)
Eisinger F, Nogues C, Birnbaum D, Jacquemier J, Sobol H. BRCA1 and medullary breast cancer.
JAMA
 
1998
;
280
:
1227
-8.
(38)
Jacquemier J, Eisinger F, Birnbaum D, Sobol H. Histoprognostic grade in BRCA1-associated breast cancer.
Lancet
 
1995
;
345
:
1503
.
(39)
Robson M, Rajan P, Rosen PP, Gilewski T, Hirschaut Y, Pressman P, et al. BRCA-associated breast cancer: absence of a characteristic immunophenotype.
Cancer Res
 
1998
;
58
:
1839
-42.
(40)
Johannsson OT, Idvall I, Anderson C, Borg A, Barkardottir RB, Egilsson V, et al. Tumour biological features of BRCA1-induced breast and ovarian cancer.
Eur J Cancer
 
1997
;
33
:
362
-71.
(41)
Loman N, Johannsson O, Bendahl PO, Borg A, Ferno M, Olsson H. Steroid receptors in hereditary breast carcinomas associated with BRCA1 or BRCA2 mutations or unknown susceptibility genes.
Cancer
 
1998
;
83
:
310
-9.
(42)
Osin P, Gusterson BA, Philp E, Waller J, Bartek J, Peto J, et al. Predicted anti-oestrogen resistance in BRCA-associated familial breast cancers.
Eur J Cancer
 
1998
;
34
:
1683
-6.
(43)
Crook T, Crossland S, Crompton MR, Osin P, Gusterson BA. p53 mutations in BRCA1-associated familial breast cancer.
Lancet
 
1997
;
350
:
638
-9.
(44)
Crook T, Brooks LA, Crossland S, Osin P, Barker KT, Waller J, et al. p53 mutation with frequent novel codons but not a mutator phenotype in BRCA1- and BRCA2-associated breast tumours.
Oncogene
 
1998
;
17
:
1681
-9.
(45)
Phillips KA, Nichol K, Ozcelik H, Knight J, Done SJ, Goodwin PJ, et al. Frequency of p53 mutations in breast carcinomas from Ashkenazi Jewish carriers of BRCA1 mutations.
J Natl Cancer Inst
 
1999
;
91
:
469
-73.
(46)
Frank TS, Manley SA, Olopade OI, Cummings S, Garber JE, Bernhardt B, et al. Sequence analysis of BRCA1 and BRCA2: correlation of mutations with family history and ovarian cancer risk.
J Clin Oncol
 
1998
;
16
:
2417
-25.
(47)
Robson ME, Offit K. New BRCA2 mutation in an Ashkenazi Jewish family with breast and ovarian cancer.
Lancet
 
1997
;
350
:
117
-8.
(48)
Schubert EL, Lee MK, Mefford HC, Argonza RH, Morrow JE, Hull J, et al. BRCA2 in American families with four or more cases of breast or ovarian cancer: recurrent and novel mutations, variable expression, penetrance, and the possibility of families whose cancer is not attributable to BRCA1 or BRCA2.
Am J Hum Genet
 
1997
;
60
:
1031
-40.
(49)
Newman B, Mu H, Butler LM, Millikan RC, Moorman PG, King MC. Frequency of breast cancer attributable to BRCA1 in a population-based series of American women.
JAMA
 
1998
;
279
:
915
-21.
(50)
Ford D, Easton DF, Stratton M, Narod S, Goldgar D, Devilee P, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium.
Am J Hum Genet
 
1998
;
62
:
676
-89.
(51)
Struewing JP, Hartge P, Wacholder S, Baker SM, Berlin M, McAdams M, et al. The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews.
N Engl J Med
 
1997
;
336
:
1401
-8.
(52)
Chen J, Silver DP, Walpita D, Cantor SB, Gazdar AF, Tomlinson G, et al. Stable interaction between the products of the BRCA1 and BRCA2 tumor suppressor genes in mitotic and meiotic cells.
Mol Cell
 
1998
;
2
:
317
-28.