SEARCH

SEARCH BY CITATION

Keywords:

  • breast carcinoma;
  • epidemiology;
  • contralateral;
  • hereditary;
  • family history

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

BACKGROUND

The objectives of the current study were to estimate the risk of developing contralateral breast carcinoma (CBC) among women with hereditary/familial non-BRCA1/BRCA2 breast carcinoma and to determine the factors that may predict their risk of CBC.

METHODS

The study sample consisted of all families (n = 217 families) that were referred between 1994–2001 to the Clinic of Cancer Genetics at the University Hospital of Umeå for suspected hereditary breast carcinoma. The study included all women in the 217 families who had carcinoma of the breast as their first primary invasive malignancy diagnosed between 1970–2001 in northern Sweden. Exclusion criteria were an estimated lifetime risk < 20%, BRCA1/BRCA2 mutation, noninvasive carcinoma (ductal or lobular carcinoma in situ), and bilateral breast carcinoma. In the final analysis, 204 women were included from 120 families.

RESULTS

The cumulative probability of developing CBC among women who had hereditary/familial non-BRCA1/BRCA2 breast carcinoma after 20 years was 27.3% (95% confidence interval, 15.0–37.8) compared with the expected risk (4.9%) among women in northern Sweden who had primary breast carcinoma. A significantly increased risk of CBC was associated with age younger than 50 years at the time of diagnosis of the first primary breast carcinoma (P = 0.006). Adjuvant hormone therapy reduced the risk of CBC (P = 0.036).

CONCLUSIONS

Women with hereditary/familial non-BRCA1/BRCA2 breast carcinoma had a high risk of developing CBC. This risk was attenuated further among women who were younger at the time of onset, who had a cumulative probability of developing CBC of nearly 40% after 15 years, which is similar to the estimated risk among BRCA1/BRCA2 mutation carriers. The results of this study emphasized the importance of genetic counseling for these women. Cancer 2006. © 2006 American Cancer Society.

Understanding the etiology of contralateral breast carcinoma (CBC) will help to identify patients who are at increased risk for developing a second primary breast tumor. Between 5–10% of women with primary breast carcinoma develop CBC.1 Accounting for nearly 45% of all second primary malignancies, CBC is the most common second primary malignancy reported among women who are diagnosed with breast carcinoma.2

Generally, women with primary breast carcinoma have a two to six times greater risk of developing CBC compared with the risk of developing primary breast carcinoma among women in the general population.3 Confirmed risk factors for CBC are young age at the time of diagnosis and a family history of breast carcinoma. Vaittinen and Hemminki used the Swedish Family Cancer Data Base to analyze age-incidence correlations for CBC and observed a high incidence (800/105 person yrs) for women ages 25–49 years. Their results indicate that young age is an important risk factor for CBC,4 which is in agreement with several other studies.5–14 However, others have reported no association with early age at onset.15, 16

Between 5–10% of all breast carcinomas are caused by germline mutations in dominant inherited genes, such as BRCA1 and BRCA2.17 Female BRCA1 and BRCA2 mutation carriers who are diagnosed with breast carcinoma have a strongly elevated lifetime risk of developing a tumor in the contralateral breast. BRCA1 carriers have an estimated cumulative risk for CBC of 48% by age 50 years and 64% by age 70 years.5 For BRCA2 carriers, the risk is 37% by age 50 years and 52% by age 70 years.18

However, only a minority (10–40%) of all hereditary breast carcinoma is caused by mutations in the BRCA1 and BRCA2-genes; to our knowledge, the risk of developing CBC among women with non-BRCA1/BRCA2 hereditary breast carcinoma has not been evaluated thoroughly. Therefore, in the current study, we investigated the risk of CBC among women with hereditary/familial non-BRCA1/BRCA2 breast carcinoma in northern Sweden.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Study Population

The study population consisted of all families (n = 217 families) that were referred between 1994–2001 to the Clinic of Cancer Genetics at the University Hospital of Umea because of suspected hereditary breast carcinoma (at least 2 close relatives with breast carcinoma). All genetic counseling in northern Sweden is performed in our Clinic of Cancer Genetics at Umea University Hospital. No referrals are made primarily to other regions in Sweden. Among 60 families, all women with breast carcinoma cases were either diagnosed before 1970, or they were living outside the Northern Healthcare Region, or their medical records were unavailable. In the remaining 157 families, 259 women with breast carcinoma had medical records available for review at the University Hospital in Umea, and those women were included in the current study.

During genetic counseling at the clinic, pedigrees were drawn for all families, and their estimated risk of breast carcinoma was calculated according to the Claus model,19 which estimates a woman's age-specific breast carcinoma risk with regard to the number of first-degree and second-degree relatives with breast carcinoma and their age at diagnosis. From the Claus model, 27 families (36 women) were defined as low-risk (< 20% life-time risk in unaffected women) and were excluded from the study. Of the remaining 130 families, 180 women were defined with hereditary breast carcinoma (≥ 3 close relatives with breast carcinoma), and 43 women were defined with familial breast carcinoma (2 close relatives with breast carcinoma).

In 61 families that had a medium-to-high probability of developing BRCA1/BRCA2 breast carcinoma according to BRCAPRO (available at URL: www3.utsouthwestern.edu/cancergene/ [accessed February 3, 2006]), 1 woman was screened for mutation in BRCA1/BRCA2 by DHPCC/PTT at a reference laboratory at the University Hospital in Lund, Sweden.20 The BRCAPRO model is used to estimate the risk that a women will carry a mutation in BRCA1 or BRCA2 considering the number of first-degree relatives with breast and/or ovarian carcinoma. In addition, screening for large deletions in BRCA1/BRCA2 also was performed in selected families that had a high probability of mutations. Five BRCA1 mutations and no BRCA2 mutations were identified, and an additional eight BRCA1 carriers were excluded. Because the current study included only women with invasive breast carcinoma, seven women with ductal carcinoma in situ (DCIS) or lobular carcinoma in situ (LCIS) were excluded. CBC was defined as a new, invasive breast carcinoma that was diagnosed ≥ 6 months after diagnosis of the primary breast carcinoma. Therefore, four women with bilateral carcinoma also were excluded. In summary, exclusion criteria were an estimated lifetime risk < 20%, BRCA1/BRCA2 mutation, noninvasive carcinoma (DCIS/LCIS), and bilateral breast carcinoma. In the final analysis, we included 204 women from 120 families (Fig. 1).

thumbnail image

Figure 1. This chart illustrates the selection procedure for the 204 women with primary breast carcinoma who were diagnosed in northern Sweden between 1970–2001.

Download figure to PowerPoint

Written informed consent was obtained from each participant. The study was approved by the Ethics Committee at the participating academic institution (Umeå University).

Data Collection and Analysis

We used medical records to identify the age of women at the time of diagnosis, methods of surgery, adjuvant treatments, tumor size, and other tumor characteristics (Table 1). All diagnoses of primary breast carcinoma were confirmed by using the Swedish Northern Cancer Register (available at URL: http://www.oc.umu.se [accessed February 3, 2006]). A contralateral tumor was defined as a histologically confirmed, secondary breast carcinoma that was diagnosed ≥ 6 months after diagnosis of the primary breast carcinoma.

Table 1. Demographic, Clinical, and Treatment Characteristics of 204 Women with Hereditary/Familial Non-BRCA1/BRCA2 Breast Carcinoma
CharacteristicNo. of patients (%)
  • ER: estrogen receptor; PR: progesterone receptor.

  • a

    Women may have received radiation therapy in several areas.

Age at diagnosis of primary breast carcinoma 
 Median52 yrs
 Range31–86 yrs
 Birth–49 yrs93
 > 50 yrs111
Tumor classification 
 T177 (38)
 T2104 (51)
 T310 (5)
 T413 (6)
Multifocal tumor 
 Yes23 (11)
 No181 (89)
Pathologic lymph node status 
 N1 (positive)84 (41)
 N0 (negative100 (49)
 Unknown20 (10)
ER status of primary tumor 
 Positive59 (29)
 Negative23 (11)
 Unknown122 (60)
PR status of primary tumor 
 Positive48 (234)
 Negative26 (13)
 Unknown130 (64)
Surgery 
 No surgery5 (3)
 Mastectomy125 (60)
 Breast conservation74 (36)
Radiation therapy 
 No46 (23)
 Yesa231 (83)
 Thorax76 (37)
 Breast72 (35)
 Axillary83 (41)
Systemic adjuvant therapy 
 No adjuvant therapy124 (61)
 Chemotherapy29 (14)
 Hormone therapy63 (31)
 Any adjuvant treatment80 (39)

The incidence of CBC among women with hereditary breast carcinoma was analyzed by the calculating standardized incidence ratio (SIR) (i.e., the ratio of the observed to expected numbers of CBC diagnoses). Follow-up for malignancies among women began 6 months after the date of primary breast carcinoma diagnosis and ended either at death, the date of emigration, or on December 31, 2002, whichever occurred first. The number of observed diagnoses and the person-years at risk were stratified according to calendar year and 5-year age groups. The expected number of diagnoses was calculated by multiplying the observed person-years by age, and calendar-specific incidence rates were derived from the Regional Cancer Registry of Northern Sweden. The cumulative probability of developing CBC was estimated using the Kaplan–Meier method. Univariate evaluation of risk factors for developing CBC was performed using the log-rank test. In multivariate analysis of risk factors, a Cox regression model was used. All reported P values were based on two-sided tests.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Table 1 presents the demographic, clinical, and treatment characteristics for the 204 women with primary non-BRCA1/BRCA2 hereditary/familial breast carcinoma in the current study. The mean follow-up was 7.4 years. In 25 of 204 women, CBC was diagnosed ≥ 6 months after the diagnosis of primary breast carcinoma, and 24 of 25 tumors were registered in the Cancer Registry. One of the tumors had been identified as part of distant metastasis; therefore, it was not reported to the Cancer Registry.

A significantly increased SIR of 5.02 (95% confidence interval [95% CI], 3.22–7.35) was observed for CBC among women with hereditary breast carcinoma: There were 25 observed diagnoses compared with 4.98 expected diagnoses (Table 2). The excess risk was pronounced most among women with early-onset (age < 50 yrs) primary breast carcinoma.

Table 2. Incidence of CBC among Swedish Women with Primary Hereditary/Familial Non-BRCA1/BRCA2 Breast Carcinoma as a Function of Age at Diagnosis
Age at diagnosis of primary breast carcinomaObservedExpectedSIR95% CI
  1. CBC: contralateral breast carcinoma; SIR: standardized incidence ratio; 95% CI: 95% confidence interval.

< 50 yrs192.876.633.99–10.35
≥ 50 yrs62.162.781.02–6.06
Total255.024.983.22–7.35

The estimated cumulative probability of developing CBC among Swedish women with hereditary/familial non-BRCA1/BRCA2 breast carcinoma was 5.5% at 5 years, 13.4% at 10 years, 24.2% at 15 years, and 27.3% at 20 years after the primary diagnosis, compared to 1.9%, 3.0%, 3.9%, and 4.9%, respectively, among all women with primary breast carcinoma in northern Sweden (Table 3; Fig. 2). When analyses were stratified according to age at diagnosis of primary breast carcinoma (age < 50 yrs vs. age ≥ 50 yrs), the highest risk for CBC was among women with early-onset primary breast carcinoma, with a cumulative probability of 41.0% at 20 years after primary diagnosis compared with +10.1% among women with late-onset disease (age > 50 yrs).

Table 3. Cumulative Probability (%) and 95% CIs of Developing CBC in Swedish Women with Hereditary/Familial Non-BRCA1/BRCA2 Breast Carcinoma by Age at Diagnosis of Primary Tumor
Yrs since diagnosis of primary breast carcinomaAge at primary diagnosis: cumulative probability (95% CI)
< 50 yrs≥ 50 yrsAllExpected
  1. CBC: contralateral breast carcinoma; 95% CI: 95% confidence interval.

5 yrs7.8 (1.5–13.8)3.5 (0.0–7.4)5.5 (1.9–9.0)1.9
10 yrs19.9 (8.4–29.9)6.9 (0.6–12.7)13.4 (6.9–19.5)3.0
15 yrs35.9 (18.0–49.9)10.1 (1.1–18.4)24.2 (13.5–33.6)3.9
20 yrs41.0 (20.5–56.3)10.1 (1.1–18.4)27.3 (15.0–37.8)4.9
thumbnail image

Figure 2. This chart illustrates a comparison of the cumulative probability (percentage) with 95% confidence intervals of developing contralateral breast carcinoma among women with primary hereditary/familial non-BRCA1/BRCA2 breast carcinoma (n = 204 patients) versus an unselected patient population with primary breast carcinoma in northern Sweden.

Download figure to PowerPoint

In the univariate analysis, age at diagnosis (age < 50 yrs vs. age ≥ 50 yrs; P = 0.006) and adjuvant hormone therapy (no adjuvant therapy vs. any adjuvant hormone therapy; P = 0.036) were found to be significant predictors of CBC development (Table 4). Significance was not reached for adjuvant chemotherapy (P > 0.05). Other modalities of treatment (e.g., the type of surgery and radiotherapy) were not found to be associated significantly with the risk of CBC.

Table 4. Results of Univariate Survival Analysis of Clinical, Histologic, and Treatment Factors Assessed for Risk of CBC that Displayed Significance at the 95% CI using the Kaplan–Meier Log-Rank Approximation Test
VariableNo. of women (n = 204)Women with CBC (n = 25)PercentageP
  • CBC: contralateral breast carcinoma; 95% CI: 95% confidence interval; NS: nonsignificant.

  • a

    There were 14 women with unknown values.

  • b

    Staging according to UICC/TNM classification.

  • c

    Grades are defined as follows: 1, well differentiated; 2, moderately differentiated; 3, poorly differentiated.

  • d

    There were eight women with unknown values.

  • e

    There was one woman with unknown values.

Age at onset    
 < 50 yrs931920.4 
 > 50 yrs11165.40.006
Risk group    
 Medium risk (20–40%)41614.6 
 High risk (> 40%)1631911.7NS
Tumor sizea    
 < 2 cm1091311.9 
 2 cm to < 5 cm70912.9NS
 > 5cm80  
 Overgrowth30  
Pathologic lymph node status    
 N1 (positive)8467.1 
 N0 (negative)1001515NS
Stage at initialb diagnosis    
 Stages I/IIA1511811.9 
 Stages IIB, IIIA, IIIB, and IV53713.2NS
Histologic grade of primary tumorcd    
 Grade 1140  
 Grade 2781215.4 
 Grade 31041110.6NS
Histology of first tumore    
 Ductal1712011.7 
 Other33515.2NS
Multifocal    
 Yes2314.3 
 No1812413.3NS
Surgical treatment    
 No surgery50  
 Mastectomy1251512 
 Breast-conserving surgery741013.5NS
Radiotherapy    
 Yes1582012.7 
 No46510.9NS
Adjuvant hormone therapy    
 Yes6334.8 
 No1412215.60.036
Chemotherapy    
 Yes2927 
 No1752313.1NS

Characteristics of the primary tumor, such as size, grade, differentiation, and type, did not differ significantly among the groups (Table 4). In a multiple Cox regression analysis (including age, adjuvant hormone treatment, risk group, histology, tumor grade, tumor stage, and multifocality as explanatory variables), age (P = 0.003) and adjuvant hormone therapy (P = 0.03) remained significant as independent risk factors.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

In the current study, the results indicated that women with hereditary/familial non-BRCA1/BRCA2 breast carcinoma have a very high risk of developing CBC. This risk is even more apparent among women who are younger at the onset of primary breast carcinoma (age < 50 yrs), with a cumulative risk of nearly 40% after 15 years. Although the association between family history and an increased risk of CBC is well documented,1, 3, 4, 6, 7, 13, 15, 17–19, 21–26 to our knowledge, this is the first study published to date regarding women with hereditary/familial non-BRCA1/BRCA2 breast carcinoma. In general, those other studies reported a two to four times increased risk of CBC among women who had a positive family history for breast carcinoma. In the majority of previous studies, women who had a strong family history were not analyzed separately because of a lack of power or because a detailed family history was not available.

When analyzing the age at primary breast carcinoma diagnoses (age < 50 yrs vs. age ≥ 50 yrs), the probability of CBC was found to be high among women who were diagnosed before age 50 years (Table 3) (Fig. 2), reaching 20% after 10 years and nearly 40% after 15 years. Women age > 50 years, however, had a probability of developing CBS of 7% after 10 years, indicating that age at diagnosis is an independent risk factor apart from genetic susceptibility. The increased risk of young age at primary diagnosis largely remained unchanged (P = 0.003) in our multivariate analysis.

The results of the current study are in agreement with the results from a cohort study of 83 families with an identified germline mutation in the BRCA1 gene. In that study, the occurrence of a second CBC was studied in relation to the age of onset of the first primary breast carcinoma. Of the evaluated 164 BRCA1 patients, 40% of BRCA1 patients who were age < 50 years at the time of their primary breast carcinoma diagnosis (n = 124 women) had developed CBC after 10 years of follow-up.5 Another recently published cohort study of BRCA1/BRCA2 mutation carriers demonstrated that, for women who were diagnosed with their first primary breast carcinoma before age 49 years who did not undergo oophorectomy or receive treatment with tamoxifen, the 10-year risk of CBC was 43.4% for BRCA1 carriers and 34.6% for BRCA2 carriers.21

The other important finding was that adjuvant hormone therapy appeared to lower the risk of CBC significantly (P = 0.036). In a meta-analysis of 55 studies of tamoxifen treatment in patients with early breast carcinoma, a reduction in the incidence of CBC (47%) was noted among women who received adjuvant tamoxifen in 5 years irrespective of their estrogen-receptor status.27 Results from a case–control study of BRCA1 and BRCA2 mutation carriers indicated that adjuvant tamoxifen reduced the risk of CBC (BRCA1: odds ratio [OR], 0.38; 95% CI, 0.19–0.74; and BRCA2: OR, 0.63; 95% CI, 0.2–1.5).28 Furthermore, in the study by Buzdar et al., which was a randomized, double-blind trial comparing anastrozole with tamoxifen as adjuvant therapy in postmenopausal women with early breast carcinoma, the investigators found that the incidence of CBC as a first event was reduced more in the anastrozole group than in the tamoxifen group (OR, 0.42; 95% CI, 0.22–0.79).29 We did not observe any effect of adjuvant chemotherapy; however, our sample size was small, and therefore the power to detect small effects was limited. However, a study overview of polychemotherapy noted a 20% reduction in the annual odds of developing CBC.30 This means there is overwhelming evidence that adjuvant, systemic treatment of breast carcinoma is one way to reduce the risk of CBC among women with early-onset breast carcinoma. The results from the current study also indicate that women with hereditary/familial non-BRCA1/BRCA2 breast carcinoma can benefit from adjuvant hormone therapy with respect to CBC.

There are some limitations to the current study. Because this study was based on clinical material, not all women were tested for BRCA1/BRCA2 mutations. Among the 60 families that were not tested, the risk according to BRCAPRO was < 10% that breast carcinomas in the family were caused by BRCA1/BRCA2 mutations. Therefore, it is unlikely that undetected BRCA1/BRCA2 mutations explain the current results. Because this study was not population-based, selection bias also may have affected the results. If women with bilateral carcinoma were more likely to be referred to our clinic, this would then result in an overestimate of the risk of CBC. However, we believe that this had only a minor effect on the results because we observed the same increased risk of CBC even after the families received genetic counseling (data not shown).

The risk of CBC among premenopausal women with hereditary/familial non-BRCA1/BRCA2 breast carcinoma is high, with a cumulative probability of 20% after 10 years. When genetic counseling is provided for this group of women, it is important to consider and provide information regarding the risk of CBC. With respect to the side effects of hormone treatment in younger women, prophylactic contralateral mastectomy may be an option. For postmenopausal women, who have a marginally increased risk of developing CBC, conservative management with hormone treatment is a sensible strategy.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES