Predictors of contralateral breast cancer in patients with unilateral breast cancer undergoing contralateral prophylactic mastectomy

Authors


Abstract

BACKGROUND:

Although contralateral prophylactic mastectomy (CPM) reduced the risk of contralateral breast cancer in unilateral breast cancer patients, it was difficult to predict which patients were most likely to benefit from the procedure. The objective of this study was to identify the clinicopathologic factors that predict contralateral breast cancer and thereby inform decisions regarding performing CPM in unilateral breast cancer patients.

METHODS:

A total of 542 unilateral breast cancer patients who underwent CPM at The University of Texas M. D. Anderson Cancer Center from January 2000 to April 2007 were included in the current study. A logistic regression analysis was used to identify clinicopathologic factors that predict contralateral breast cancer.

RESULTS:

Of the 542 patients included in this study, 25 (5%) had an occult malignancy in the contralateral breast. Eighty-two patients (15%) had moderate-risk to high-risk histologic findings identified at final pathologic evaluation of the contralateral breast. Multivariate analysis revealed that 3 independent factors predicted malignancy in the contralateral breast: an ipsilateral invasive lobular histology, an ipsilateral multicentric tumor, and a 5-year Gail risk ≥1.67%. Multivariate analysis also revealed that an age ≥50 years at the time of the initial cancer diagnosis and an additional ipsilateral moderate-risk to high-risk pathology were independent predictors of moderate-risk to high-risk histologic findings in the contralateral breast.

CONCLUSIONS:

The findings indicated that CPM may be a rational choice for breast cancer patients who have a 5-year Gail risk ≥1.67%, an additional ipsilateral moderate-risk to high-risk pathology, an ipsilateral multicentric tumor, or an ipsilateral tumor of invasive lobular histology. Cancer 2009. © 2009 American Cancer Society.

INTRODUCTION

Women diagnosed with breast cancer have a significantly increased risk of developing contralateral breast cancer over their lifetime.1-4 To reduce the risk of contralateral breast cancer, some breast cancer patients choose to undergo contralateral prophylactic mastectomy (CPM) in addition to mastectomy of the affected breast. The most common reasons for CPM may include physician advice, patient fear of another breast cancer diagnosis, desire for cosmetic symmetry, family history of breast cancer or other malignancy, and fibrocystic breast disease that makes surveillance more challenging.5

The observed impact of CPM on contralateral breast cancer reduction is in agreement with findings on familial breast cancer. McDonnell et al estimated a 90% reduction after a 10-year median follow-up using the Anderson model to predict contralateral breast cancer risk.6 Van Sprundel et al reported that CPM reduced the risk for contralateral breast cancer in BRCA1 or BRCA2 mutation carriers by 91%.7 Although CPM significantly reduces the risk of contralateral breast cancer,2, 6-8 the procedure is aggressive and irreversible and is also unnecessary for patients who are not likely to develop contralateral disease. Moreover, most patients will not experience any survival benefit from CPM because the risk of systemic metastases from their index cancer often exceeds the risk of developing a contralateral breast cancer. It is difficult to determine which patients are most likely to benefit from a CPM procedure. It stands to reason that the potential benefit of CPM is the greatest for those patients who have the highest risk of developing a contralateral breast cancer. Moderate-risk to high-risk findings of atypical ductal hyperplasia, atypical lobular hyperplasia, or lobular carcinoma in situ in the contralateral breast may represent surrogate markers of elevated risk of future contralateral breast cancer risk.1

The objective of the current study was to identify the clinical and pathologic factors that predict contralateral breast cancer or moderate-risk to high-risk pathology in the contralateral breast and thereby inform decisions regarding performing CPM in patients with a diagnosis of unilateral breast cancer.

MATERIALS AND METHODS

Patient Selection and Data Collection

We used the Surgical Oncology Breast Cancer Database to retrospectively evaluate unilateral breast cancer patients who underwent CPM at The University of Texas M. D. Anderson Cancer Center from January 2000 to April 2007. The M. D. Anderson Institutional Review Board approved this study.

Patients were included if they had unilateral primary breast cancer and had no clinical or radiographic evidence of a contralateral breast malignancy. Patients known to have had bilateral breast cancer before CPM were excluded. A total of 542 CPM patients were included in the study. Some of these patients were included in a previous report from our institution.9 Clinical factors examined included patient age, race, age at menarche, age at first live birth, number and findings of previous breast biopsies, family history of breast cancer, ipsilateral clinical tumor stage, reasons for CPM, histologic findings for both breasts, and follow-up status. We classified atypical ductal hyperplasia (ADH), atypical lobular hyperplasia (ALH), and lobular carcinoma in situ (LCIS) as moderate-risk to high-risk histologic findings. The presence of ipsilateral proliferative disease with atypia (either ductal or lobular) or LCIS in association with the diagnosed cancer also was evaluated. Additional ipsilateral moderate- to high-risk lesions were defined as concurrent high-risk proliferative lesions including ADH, ALH, or LCIS within the surgical specimen in the ipsilateral breast.

Mastectomies were performed on the ipsilateral side with or without lymph node staging as considered appropriate based on diagnostic biopsy findings. Synchronous or delayed CPM was usually performed with or without sentinel lymph node biopsy at the discretion of the operating surgeon. After surgery, the oncologist recommended adjuvant systemic therapy based on the final pathology information. Some patients received chemotherapy before surgery. Routine follow-up consisted of patient history and physical examination of the breast, chest wall, and regional lymph node basins at 3-month and 6-month intervals for 5 years followed by annual examinations for lifetime. Imaging studies were performed as needed based on symptoms and/or physical findings. Reconstructed breasts were not routinely imaged.

Gail Model

The Gail model has been validated in several studies to calculate the risk of developing an invasive breast cancer over the next 5 years and the lifetime probability of developing breast cancer for women who have not had a diagnosis of breast cancer.10-14 The 5-year risk of 1.67% has been used as the cutoff point for the definition of “high risk”. Goldflam et al and Kulkarni et al previously used the Gail model to calculate the risk of contralateral breast cancer for women with unilateral breast cancer.1, 15 In the current study, we evaluated breast cancer risk factors included in the Gail model (age, race, age at menarche, number and findings of previous breast biopsies, age at first live birth, and number of first-degree relatives with breast cancer).12 The Breast Cancer Risk Assessment Tool (the “Gail model” available at http://www.cancer.gov/bcrisktool/ accessed April 2008, or 800-4-CANCER) was used to calculate the 5-year Gail risk for each patient in our study.

Case-matched Control Group

We identified 1574 patients with unilateral primary breast cancer who did not undergo CPM and included them as a case-matched control group. The control group was matched for clinical TNM stage, age (±5 years), and race.

Statistical Analysis

For statistical analysis, patients who underwent CPM were separated into 3 groups: contralateral findings of none to low-risk pathology, contralateral findings of moderate-risk to high-risk pathology, and contralateral findings of malignancy on final pathology. Data were subjected to univariate analysis and then multivariate analysis. Stepwise multiple logistic regression analysis was used to identify variables that predicted malignant findings in the contralateral breast.16 The same analytic procedure was repeated to identify variables that predicted moderate-risk to high-risk histologic findings in the contralateral breast. All P values were 2-tailed, and values of P ≤ .05 were considered significant. Stata statistical software (StataSE 9; StataCorp LP, College Station, TX) was used for all statistical analyses.

RESULTS

Patient, Tumor, and Treatment Characteristics

A total of 542 patients underwent CPM and formed the primary study group. Patient, tumor, and treatment characteristics are summarized in Table 1. The median age of patients who underwent CPM was 49 years (range, 22 years-85 years). The clinical TNM stage of the primary tumor at the time of diagnosis was stage 0, stage I, stage II, and stage III in 18.2%, 36.9%, 33.4%, and 11.4% of patients, respectively. Of the 542 unilateral breast cancer patients who underwent CPM, 25 patients (4.6%) had an occult malignancy, 8 were diagnosed with invasive carcinomas (1.5%), and 17 were diagnosed with ductal carcinoma in situ (DCIS) (3.1%), in the contralateral breast on final pathology. There were 82 (15%) patients who had moderate-risk to high-risk histologic findings (16 with LCIS, 21 with ALH, and 37 with ADH) in the contralateral breast. A total of 74 patients (13.6%) had ipsilateral lobular histology.

Table 1. Patient, Tumor, and Treatment Characteristics of 542 Unilateral Breast Cancer Patients Who Underwent Contralateral Prophylactic Mastectomy
CharacteristicNo. of Patients (%)
Median age, y (range)49 (22-85)
Family history of breast cancer 
 Yes316 (58.3)
 Yes, first-degree relative171 (31.5)
 No222 (41.0)
 Unknown4 (0.7)
5-year Gail risk, % 
 Median (range)1.3 (0.2-2.2)
 No. of patients with risk ≥1.67144 (26.6)
Hormone replacement therapy 
 No345 (63.7)
 Yes179 (33.0)
 Unknown18 (3.3)
Ipsilateral multicentric tumor 
 No493 (91.0)
 Yes49 (9.0)
Ipsilateral breast cancer stage 
 099 (18.3)
 I200 (36.9)
 II181 (33.4)
 III62 (11.4)
Neoadjuvant chemotherapy 
 Yes197 (36.3)
 No345 (63.7)
Ipsilateral invasive lobular histology 
 No468 (86.3)
 Yes74 (13.7)
Histologic findings in the contralateral breast 
 Malignant25 (4.6)
 Moderate- to high-risk82 (15.1)
 None to low-risk435 (80.3)
Adjuvant chemotherapy 
 Yes209 (38.6)
 No329 (60.7)
 Unknown4 (0.7)
Adjuvant endocrine therapy 
 Yes266 (49.1)
 No272 (50.2)
 Unknown4 (0.7)
Radiotherapy 
 Yes146 (26.9)
 No392 (72.3)
 Unknown4 (0.7)

The information necessary to calculate risk using the Gail model was available for 492 (90.8%) patients. The median 5-year Gail risk was 1.3% (range, 0.2%-2.2%), and 144 patients (29%) had a 5-year risk ≥1.67%. Greater than half of these patients (58%) had a family history of breast cancer.

Reasons Patients Chose to Undergo CPM

The majority of patients (72%) chose to undergo CPM for 1 or 2 of the following reasons: a family history of breast cancer, difficulty in surveillance for contralateral breast cancer because of clinically and mammographically dense breast tissue or diffuse indeterminate microcalcifications in the contralateral breast, and psychologic fear of developing another breast cancer.

Contralateral Breast Cancer Incidence and Follow-up Status

Based on the known per-year incidence rates of contralateral breast cancer in breast cancer patients (approximately 0.5%-1%),2, 4 we predicted that 8 to 16 contralateral breast cancers would develop in those 542 patients at a mean follow-up of 34.7 months. At the time of this analysis, 91.8% of the patients were alive and without evidence of disease (Table 2). At a median follow-up of 32.7 months, 67 patients who underwent CPM had developed 36 local, 13 regional, and 36 systemic recurrences, but no patients had developed a contralateral breast cancer.

Table 2. Outcomes of 542 Patients With Unilateral Breast Cancer Patients Who Underwent CPM and 1626 Patients in a Case-Matched Control Group Who Did Not Undergo CPM
CharacteristicCPM Group N = 542 (%)Control Group N = 1574 (%)
  1. CPM indicates contralateral prophylactic mastectomy.

Median follow-up time (range), mo32.7 (0.2-90.2)49 (2.4-99.4)
Contralateral primary  
 Yes0 (0)37 (2.4)
 No542 (100)1537 (97.6)
Ipsilateral local  recurrence  
 Yes36 (6.6)21 (1.3)
 No506 (93.4)1553 (98.7)
Ipsilateral regional  recurrence  
 Yes13 (2.4)42 (2.7)
 No529 (97.6)1532 (97.3)
Distant recurrence  
 Yes36 (6.6)156 (9.9)
 No506 (93.4)1418 (90.1)
Status at last contact  
 Alive, free of disease497 (91.9)1337 (84.7)
 Alive with disease19 (3.5)46 (2.9)
 Deceased, from disease18 (3.3)125 (7.9)
 Deceased, other cause3 (0.6)34 (2.2)
 Deceased, cause unknown2 (0.4)26 (1.6)
 Unknown2 (0.4)6 (0.4)

In the case-matched control group, we calculated the incidence rates of contralateral breast cancer as well as local, regional, and systemic recurrences. There were 37 contralateral breast cancers (2.4%) found in the 1574 unilateral breast cancer patients at a mean follow-up of 50.2 months. The incidence rate of contralateral breast cancer in this case-matched control group was 0.56% per year. At a median follow-up of 49 months, patients who did not undergo CPM had developed 21 local (1.3%), 42 regional (2.7%), and 156 systemic recurrences (9.9%). At the time of the current study, 84.7% of patients who did not undergo CPM were alive and without evidence of disease (Table 2). Disease-free survival rates in the 2 groups were similar.

Univariate and Multivariate Analysis for Findings of Contralateral Breast Cancer

Clinical and pathologic predictors of malignant histologic findings in the contralateral breast are shown in Table 3. Univariate analysis revealed that patient age and a 5-year Gail risk ≥1.67%, an ipsilateral invasive lobular histology, an additional ipsilateral moderate-risk to high-risk pathology, and an ipsilateral multicentric tumor were shown to be significant predictors of breast cancer in the contralateral breast. However, patient race, estrogen receptor (ER) status, progesterone receptor (PR) status, previous hormone replacement therapy, and first-degree family history of breast cancer were not associated with increased contralateral breast cancer risk.

Table 3. Clinicopathologic Predictors of Contralateral Malignant Findings in Unilateral Breast Cancer Patients Who Underwent CPM
VariablePatients With No or Low-risk Findings n = 435 (%*)Patients With Malignant Findings n = 25 (%*)Univariate P ValueUnivariate Odds RatioMultivariate Odds Ratio (P)
  • CPM indicates contralateral prophylactic mastectomy; ER, estrogen receptor; PR, progesterone receptor.

  • *

    Percentages calculated in 3 groups: Patients with no or low-risk findings, patients with moderate-risk to high-risk findings, and patients with malignant findings.

Age at index tumor, y .01  
 Mean48.254.3   
 Median (range)47 (22-85)56 (26-76)   
Age, y     
 <50247 (86.7)8 (2.8)   
 ≥50188 (73.2)17 (6.6).022.8 
Ipsilateral tumor stage    
 077 (77.8)5 (5.0)   
 I153 (76.5)12 (6.0).731.2 
 II151 (83.4)6 (3.3).43.6 
 III54 (87.1)2 (3.2).51.6 
Ipsilateral invasive lobular histology    
 No383 (81.8)17 (3.6)   
 Yes52 (70.3)8 (10.8).0063.53.4 (.01)
Additional ipsilateral moderate to high-risk lesions   
 No392 (84.5)16 (3.5)   
 Yes43 (55.1)9 (11.5)<.00015.1 
Ipsilateral multicentric tumor    
 No398 (80.7)20 (4.1)   
 Yes37 (75.5)5 (10.2).062.73.1 (.04)
ER status ipsilateral tumor    
 Positive233 (77.2)16 (5.3)   
 Negative126 (88.1)5 (3.5).30  
 Unknown45 (80.4)4 (7.1)   
PR status ipsilateral tumor    
 Positive198(77.3)13 (5.1)   
 Negative156 (85.7)8 (4.4).59  
 Unknown51 (80.0)4 (6.3)   
Gail risk     
 <1.67%286 (82.2)10 (2.9)   
 ≥1.67%101 (70.1)13 (9.0)0.0033.73.5 (.005)
Hormone replacement    
 Yes135 (75.4)11 (6.2)   
 No285 (82.6)13 (3.8).17  
 Unknown15 (83.3)1 (5.6)   

Multivariate analysis revealed that 3 independent factors were predictive of a malignancy in the contralateral breast: an ipsilateral invasive lobular histology (odds ratio [OR] of 3.4; P = .01), an ipsilateral multicentric tumor (OR of 3.1; P = .004), and a 5-year Gail risk ≥1.67% (OR of 3.5; P = .005).

Multivariate Analyses for Contralateral Breast Cancer Between Patients Treated With or Without Neoadjuvant Chemotherapy

We compared multivariate models between patients who did not receive neoadjuvant chemotherapy and those who received neoadjuvant chemotherapy. In patients who did not receive neoadjuvant chemotherapy, an ipsilateral multicentric tumor (OR of 3.7; P = .03), and a 5-year Gail risk ≥1.67% (OR of 4.6; P < .0001) were associated with contralateral breast cancer. In patients who received neoadjuvant chemotherapy, only an ipsilateral invasive lobular histology (OR of 21.3; P = .0009) was found to be associated with contralateral breast cancer.

Univariate and Multivariate Analysis for Findings of Moderate-risk to High-risk Pathology in the Contralateral Breast

Clinical and pathologic predictors of moderate- to high-risk histologic findings in the contralateral breast are shown in Table 4. Univariate analysis revealed that patient age, negative ER status, negative PR status, and an additional ipsilateral moderate-risk to high-risk pathology were shown to be significant predictors of moderate- to high-risk histologic findings in the contralateral breast. However, patient race, an ipsilateral invasive lobular histology, an ipsilateral multicentric tumor, previous hormone replacement therapy, and first-degree family history of breast cancer were not found to be associated with moderate-risk to high-risk histologic findings in the contralateral breast.

Table 4. Clinicopathologic Predictors of Contralateral Moderate-risk to High-risk Findings in Patients With Unilateral Breast Cancer Patients Who Underwent CPM
VariablePatients With No or Low-risk Findings n = 435 (%*)Patients With Moderate- to High-risk Findings n = 82 (%*)Univariate P ValueUnivariate Odds RatioMultivariate Odds Ratio (P)
  • CPM indicates contralateral prophylactic mastectomy. ER, estrogen receptor; PR, progesterone receptor.

  • *

    Percentages were calculated in 3 groups: Patients with no or low-risk findings, patients with moderate-risk to high-risk findings, and patients with malignant findings.

Age, y     
 <50247 (86.7)30 (10.5)   
 ≥50188 (73.2)52 (20.2).0012.32.0 (.002)
Ipsilateral tumor stage    
 077 (77.8)17 (17.2)   
 I153 (76.5)35 (17.5).911.0 
 II151 (83.4)24 (13.3).34.7 
 III54 (87.1)6 (9.7).18.5 
Ipsilateral invasive lobular histology    
 No383 (81.8)68 (14.5)   
 Yes52 (70.3)14 (18.9).21  
Additional ipsilateral moderate- to high-risk lesions   
 No392 (84.5)56 (12.1)   
 Yes43 (55.1)26 (33.3)<.0014.24.2 (<.0001)
Ipsilateral multicentric tumor    
 No398 (80.7)75 (15.2)   
 Yes37 (75.5)7 (14.3).99  
ER status ipsilateral tumor    
 Positive233 (77.2)53 (17.5)   
 Negative126 (88.1)12 (8.4).010.4 
 Unknown45 (80.4)7 (12.5)   
PR status ipsilateral tumor    
 Positive198(77.3)45 (17.6)   
 Negative156 (85.7)18 (9.9).020.5 
 Unknown51 (80.0)9 (14.1)   
Gail risk     
 <1.67%286 (82.2)52 (14.9)   
 ≥1.67%101 (70.1)30 (20.8).061.6 
Hormone replacement    
 No285 (82.6)47 (13.6)   
 Yes135 (75.4)33 (18.4).12  
 Unknown15 (83.3)2 (11.1)   

Multivariate analysis revealed that an age ≥50 years at the initial cancer diagnosis (OR of 1.9; P = .01) and an additional ipsilateral moderate-risk to high-risk pathologic finding (OR of 3.7; P < .001) were independent predictors of moderate-risk to high-risk histologic findings in the contralateral breast.

Multivariate Analyses for Moderate-risk to High-risk Histologic Findings in the Contralateral Breast Between Patients Treated With or Without Neoadjuvant Chemotherapy

We compared multivariate models between patients who did not receive neoadjuvant chemotherapy and those who received neoadjuvant chemotherapy. In patients who did not receive neoadjuvant chemotherapy, age ≥50 years at the initial cancer diagnosis (OR of 1.9; P = .04) and an additional ipsilateral moderate-risk to high-risk pathology (OR of 3.3; P < .001) were associated with moderate-risk to high-risk histologic findings. In patients who received neoadjuvant chemotherapy, only an additional ipsilateral moderate-risk to high-risk pathology (OR of 7.0; P < .001) was found to be associated with moderate-risk to high-risk histologic findings.

DISCUSSION

We found that CPM was associated with a low risk of occult contralateral breast cancer (4.6%). Based on our analysis, CPM may be a rational choice for breast cancer patients who have a 5-year Gail risk ≥1.67%, an additional ipsilateral moderate-risk to high-risk pathology, an ipsilateral multicentric tumor, or an ipsilateral invasive lobular histology.

Occult contralateral breast cancer ranges from 4.6% to 68% depending on the method of diagnosis.1, 9, 17-19 Nielsen et al reported that 68% of unilateral breast cancer patients had evidence of contralateral breast cancer in 84 consecutive autopsies,18 whereas Wanebo et al, using randomly selected contralateral breast biopsies, found a rate of 18%.19 Other studies found that 5% of unilateral breast cancer patients had evidence of contralateral breast cancer.1, 9, 17 Our findings are consistent with the lower rate of contralateral breast cancer findings and to the best of our knowledge, is 1 of the largest recent cohorts of CPM patients evaluated for occult contralateral breast cancer.

In the current study, the rate of contralateral cancer among women without CPM was 2.4%, which is close to the 2.7% incidence reported in a nationwide study of 50,000 women diagnosed between 1979 and 1999.2 Peralta et al reported a higher rate of 19.8%.8 We noted 25 incidental contralateral cancers at the time of CPM compared with an expected 8 to 16 new contralateral cancers based on the known per-year incidence rates. This finding suggests that, in our cohort, CPM led to early detection and resection of already existing contralateral breast cancers more than prevention of expected contralateral cancers.

According to the consensus statement of the Society of Surgical Oncology, potential reasons breast cancer patients choose to undergo CPM include risk reduction, difficult surveillance, and reconstructive issues such as symmetry and/or balance.20 In the current study, the 2 most common reasons patients chose to undergo CPM were a family history of breast cancer and difficult surveillance.

The risk of developing contralateral cancer is not the same for all breast cancer patients, so it is unnecessary to routinely perform CPM in all patients.1, 2 A recent study showed the rate of CPM for patients with stages I through III unilateral breast cancer increased by 150% from 1998 to 2003 in the US.21 Defining the risk of contralateral cancer for breast cancer patients may help to reduce the rate of CPM. The current study findings suggest that specific histologic features of the ipsilateral tumor are associated with a higher incidence of contralateral cancer. The diagnosis of invasive lobular carcinoma increases the risk of contralateral breast cancer.4, 9, 22, 23 Erdreich et al reported that infiltrating lobular carcinoma (ILC) was diagnosed more frequently in elderly patients and that the risk of contralateral breast cancer was 2.6-fold higher in patients with invasive lobular cancer compared with other histologies.22 Boughey et al found a 5-fold higher risk of occult invasive cancer in patients with ILC over the average patient undergoing CPM.9 ILC has been shown to be more likely in older patients and to have a higher rate of contralateral breast cancer: 10.8% compared with 3.6% in other non-ILC cases in the current study and 21% compared with 11% in invasive ductal carcinoma (IDC) cases in the study by Arpino et al.24 The results of the current study provide additional evidence supporting that patients with invasive lobular carcinoma are at an increased risk of developing contralateral breast cancer.

Multicentric disease is another histopathologic feature that fits into the field-effect model for developing a contralateral breast cancer; however, to our knowledge, very few studies to date have specifically evaluated this risk factor. Lesser et al and Newman et al evaluated invasive breast cancers for multicentricity and found that multicentric tumors were a significant risk factor for bilateral cancer.25, 26 Robbins and Berg found that multicentric breast cancer increases the risk of harboring disease in the contralateral breast.4 In the current study, multicentric disease was found to be a predictor of contralateral breast cancer.

Although the Gail model was developed primarily to assess risk in patients without a breast cancer diagnosis, we found that a 5-year Gail risk ≥1.67% is a predictor of contralateral breast cancer in patients with unilateral breast cancer. This is in contrast to the findings of Goldflam et al.1 Our findings, similar to those of Kulkarni et al, suggest that the Gail model not only predicts the risk of breast cancer in women without a diagnosis of breast cancer, but may also further stratify the risk of contralateral breast cancer among breast cancer patients.15 Our findings are consistent with previous studies demonstrating that patients aged ≥50 years at the time of initial breast cancer diagnosis or with at least 1 affected first-degree relative have an increased risk of contralateral breast cancer.1, 3, 27 However, the Gail model takes into consideration other factors in addition to patient age and family history, and thus may be a better predictor of increased contralateral breast cancer risk.

Patient age may serve as an additional predictor for malignant histologic findings in the contralateral breast. It is likely that age and 5-year Gail risk are associated with each other. Increased age was also associated with an increased risk of moderate-risk to high-risk histologic findings in the contralateral breast on univariate analysis and multivariate analysis. This is not surprising because age is a known risk factor for breast carcinoma and has been found to be correlated with an increased risk of malignant or moderate-risk to high-risk histologic findings in prophylactic mastectomy samples.9, 28

In the current study, a first-degree family history of breast carcinoma alone was not found to be associated with an increased incidence of malignant or moderate-risk to high-risk contralateral breast findings. This is consistent with data from Lesser et al, who in a study of 880 breast cancer patients, found that family history of breast carcinoma was not related to the development of bilateral breast cancer.25

We identified 2 independent predictors of moderate-risk to high-risk pathology in the contralateral breast using multivariate analysis: age ≥50 years and an additional ipsilateral moderate-risk to high-risk pathology. It is known that moderate-risk to high-risk histologies such as ADH, ALH, and LCIS confer an increased risk for the development of breast cancer.29-34 The model for the evolution of IDC involves the progression from ductal hyperplasia to ADH to DCIS to IDC.35 Newsham reported that specific genomic alterations occurred in LCIS as well as in invasive cancer, suggesting there may be a genetic progression of LCIS to invasive cancer.36 Page et al found that invasive carcinoma occurred 3 times more frequently in contralateral breasts with ALH than without ALH, suggesting that ALH is an intermediate between a precursor and a marker of generalized risk of cancer in both breasts.33 Overall, the relative risk reported in the literature ranges from 3 to 5.3 for ALH, 2.03 to 4.3 for ADH, and 5.7 to 11 for LCIS.29-34 Taken together, these data suggest that moderate- to high-risk findings of ADH, ALH, and LCIS in the ipsilateral breast can be used as surrogate markers of elevated contralateral breast cancer risk.

We noted that different factors were observed in patients receiving neoadjuvant chemotherapy and those who did not receive neoadjuvant treatment. In patients receiving neoadjuvant chemotherapy, the association between ILC and contralateral breast cancer is likely due to the finding that patients with ILC usually had larger tumors and were treated with chemotherapy before surgery.

Patients with unilateral breast cancer have options less extreme than CPM for risk reduction. Systemic chemotherapy and endocrine therapy have been shown to reduce the risk of contralateral cancer in unilateral breast cancer patients. Using magnetic resonance imaging (MRI) in addition to mammography may improve rates of contralateral breast cancer detection. Lehman et al reported that MRI can detect contralateral cancer that is undetectable by mammography at the time of the initial breast cancer diagnosis.37 However, it is imperative to pathologically confirm suspicious MRI findings to prevent unnecessary CPMs. Finally, although several trials have demonstrated that tamoxifen significantly reduces the rate of contralateral cancers, this treatment is appropriate only for patients with hormone receptor-positive breast cancers.38, 39

Our study has limitations, including those inherent to any single-institutional, retrospective study. One limitation in this respect is the relatively short follow-up time. Whether the contralateral breast cancer incidence will remain low in our CPM cohort remains to be determined. Notwithstanding, the results of this study can be valuable to other institutions with respect to their own recommendations on the use of CPM in patients with unilateral breast cancer patients and may be useful in the design of trials addressing this issue.

In conclusion, we found that unilateral breast cancer patients who underwent CPM had a low risk of developing contralateral breast cancer. CPM may be a rational choice for breast cancer patients who have a 5-year Gail risk ≥1.67%, an additional ipsilateral moderate-risk to high-risk pathology, an ipsilateral multicentric tumor, or an ipsilateral invasive lobular histology. Evaluating 5-year Gail risk and histologic findings in the ipsilateral breast in unilateral breast cancer patients may help predict the likelihood of contralateral breast cancer and assist in stratifying risk and counseling patients.

Conflict of Interest Disclosures

The authors made no disclosures.

Ancillary