Age-Related Incidence and Peak Occurrence of Contralateral Breast Cancer

This cohort study compares the cumulative incidence and peak occurrence period of contralateral breast cancer between patients aged 35 years or younger and those older than 35 years at the time of surgery for primary breast cancer.


Introduction
Improved survival for patients with breast cancer coupled with advances in treatment have resulted in higher numbers of patients developing contralateral breast cancer (CBC). 1 The cumulative incidence rate for development of CBC at 5 to 10 years and more than 10 years of follow-up ranges from 1.15% to 3.8% and 3.8% to 8.3%, respectively. 2Factors associated with developing CBC include

Key Points
Question Do the cumulative incidence and peak occurrence period of contralateral breast cancer (CBC) differ according to age at primary breast cancer surgery?
Findings In this cohort study of 16 251   patients with stage 0 to III breast cancer, the younger group (Յ35 years) had significantly higher incidence of CBC than the older group (>35 years).The peak occurrence period for CBC was earlier in the younger group vs the older group with hormone receptornegative, ERBB2-positive subtype.

Meaning
The findings provide information that may be valuable when evaluating risk for CBC.

+ Supplemental content
Author affiliations and article information are listed at the end of this article.
Open Access.This is an open access article distributed under the terms of the CC-BY License.(HR) status, lobular type, and not undergoing adjuvant endocrine therapy or chemotherapy. 3,4agnosis of PBC at a younger age is a significant risk factor.Kurian et al 5 reported that women diagnosed with PBC before age 30 years had a 36-fold higher risk of developing CBC than those at any other age.7][8] Suggested reasons for the high incidence of CBC development in younger patients with breast cancer are genetic factors such as family history or BRCA variation, a high prevalence of triple-negative breast cancer, and the aggressive character of PBC. 7,9,10anwhile, the rate of contralateral prophylactic mastectomy (CPM) is increasing, especially in young patients. 11As a higher proportion of younger patients was observed in the age distribution of Korean patients with breast cancer compared with those in Western countries, 12,13

Study Population
In this retrospective, single-center cohort study, we used the Asan database, a prospectively

Pathology
Pathologic data were evaluated at the Department of Pathology at the Asan Medical Center.We used pathologic TNM staging for patients who had up-front surgery and clinical TNM staging for those who underwent neoadjuvant chemotherapy.TNM stage was assigned according to the American Joint Committee on Cancer Staging Manual (7th edition) classification.Tumor subtypes were categorized according to HR and ERBB2 status as HR+/ERBB2−, HR+/ERBB2+, HR−/ERBB2+, and HR−/ERBB2−.
Hormone receptor was considered positive if either estrogen or progesterone receptor was positive.
Immunohistochemistry was used to determine the estrogen HR and ERBB2 status.Estrogen and progesterone receptor statuses were considered positive if more than 10% of the cells were positive.
For ERBB2 overexpression, patients with immunohistochemistry grades 0 or 1 were considered negative and those with grade 3 or higher were considered positive.Patients with grade 2 were further evaluated using fluorescence in situ hybridization.

Statistical Analysis
All statistical analyses were performed using SAS, version 9.4 (SAS Institute Inc) and R, version 3.

Patient Characteristics
We collected data from 20 405 female patients, and after exclusion, 16 251 were included in the analysis; all patients were Korean, and the mean (SD) age was 48.61 (10.06) years.The flowchart for the study population is shown in Figure

Factors Associated With CBC
In both univariate analysis and multivariate analysis adjusted for family history, histology grade, subtype, T stage, and hormone therapy status of PBC, PBC surgery at younger age (Յ35 years) was significantly associated with development of CBC.In univariate analysis, the hazard ratio was 2.49 (95% CI, 1.93-3.21),and in multivariate analysis, the hazard ratio was 2.10 (95% CI, 1.62-2.74).Other factors associated with increased risk of CBC were family history; higher nuclear grade; HR-/ERBB2subtype; carcinoma in situ (patients with Tis-stage cancer had higher risk of developing CBC than those with invasive cancer with lesions less than 5 cm in diameter); in case of invasive cancer, higher T stage; and no treatment with hormone therapy (eTable 2 in Supplement 1).
We conducted a separate analysis of patients who underwent a BRCA test by adding BRCA variation status as an adjusting variable.Younger age at surgery was an independent factor associated with developing CBC regardless of BRCA variation status (hazard ratio, 1.74; 95% CI, 1.09-2.77)(eTable 3 in Supplement 1).

Analysis for Cumulative Incidence of CBC
We compared the cumulative incidence of CBC in each age group.We found that 347 of 14 933 patients (2.32%) in the older group and 71 of 1318 (5.39%) in the younger group developed CBC.According to the Kaplan-Meier curve, the younger group showed significantly higher incidence of CBC than the older group (10-year cumulative CBC incidence, 7.1% vs 2.9%; P < .001)(Figure 2A).
In subgroup analysis according to the subtype of PBC, the cumulative incidence of CBC was significantly higher in the younger group for every subtype (Figure 2B-E).The HR-/ERBB2-subtype was associated with the highest CBC incidence among the subtypes (10-year cumulative CBC incidence, 12.4% vs 4.4% in the younger and older group, respectively; P < .001).Meanwhile, the 10-year cumulative CBC incidence for the HR+/ERBB2-subtype was 4.8% vs 2.7% (P = .002)and for the HR-/ERBB2+ subtype was 6.1% vs 2.9% (P = .02)in the younger and older groups, respectively.

Analysis for Hazard Rate of CBC
The younger group showed a higher hazard rate for CBC throughout the follow-up period, and both age groups presented a similar pattern of hazard rate; risk for CBC increased until the peak at around 10 years (Figure 3A).In subgroup analyses, the peak occurrence period of CBC varied according to the subtype of PBC.For the HR+/ERBB2-subtype, the hazard rate continuously increased over time in both age groups.For the HR-/ERBB2-subtype, CBC risk peaked at around 10 years.For both subtypes, younger patients showed higher hazard rates.A total of 170 of 1318 patients in the younger group (12.90%) were categorized as having the HR-/ERBB2+ subtype; 8 of these patients (4.71%), with median follow-up of 128 months (IQR, 87-163 months), developed CBC.For the HR-/ERBB2+ subtype, the hazard rate in the younger group peaked earlier, at 1.7 years, compared with 4.8 years since first surgery in the older group (Figure 3B-E).

Discussion
In this large-scale retrospective study, the younger group, which constituted 8.11% of the total study population, had a significantly higher cumulative incidence of CBC compared with the older group, especially in those with the HR-/ERBB2-subtype.The CBC risk pattern showed different timing according to the subtype of PBC, and in the younger group with the HR-/ERBB2+ subtype, the peak occurrence time for CBC was earlier (before 3 years since surgery for PBC) than among other patients.
The cumulative incidence of CBC in the younger group was significantly higher compared with that in the older group, with a 10-year cumulative incidence of 7.1% vs 2.9%.Our findings align with those of a systematic review 13 that reported a median 10-year cumulative incidence of 3.1% for CBC.
Similarly, a nationwide Danish study 14 reported 10-year cumulative incidences of 5.5%, 4.7%, and 4.2% for age groups younger than 45 years, 45 to 54 years, and older than 55 years, respectively, which are consistent with the CBC incidence range observed in our study.
According to large population studies 15,16 conducted in North America and Europe, the annual incidence of CBC ranges relatively consistently from 0.2% to 0.7%, even with long follow-up, leading to a continuous increase in the cumulative incidence of CBC.We evaluated whether risk patterns at certain time frames after surgery for PBC varied according to the subtype or age at surgery.In a previous study by some of us, 9 young patients (age <35 years) with the ERBB2+ subtype of PBC had the highest risk for CBC at around 4.6 years after PBC surgery, while the risk among young patients with the HR+/ERBB2-subtype peaked at 7.1 years; other patients showed a constant increase over the follow-up period.As that study was propensity matched for young patients (4 controls per case), we analyzed general population data in the current study.This study showed a similar result in the younger group with HR-/ERBB2+ subtype, with the peak hazard rate reached earlier in the younger group compared with the older group (Figure 3D).This may be interpreted as having a shorter interval for developing CBC.This was also shown in a Kaplan-Meier graph for cumulative incidence; unlike the HR+/ERBB2-or HR-/ERBB2-subtype, which showed a constant increase in cumulative incidence, the younger group with HR-/ERBB2+ subtype showed a relatively rapid increase at first and then a plateau afterward (Figure 2).To our knowledge, there is no study analyzing the hazard ).Incidence of CBC in this subgroup (4.7%) was less than the 10-year cumulative incidence for the whole younger group (7.1%).We can conclude that although patients with HR-/ERBB2+ breast cancer had a short interval for developing CBC, the incidence was lower than for other subtypes.Our findings suggest that a more thorough examination, such as breast magnetic resonance imaging for CBC, can be considered in young patients with the HR-/ERBB2+ subtype at an early period after the surgery for PBC, as the short interval for developing CBC is a poor prognostic factor. 17However, the number of patients who developed CBC in this subgroup was small; therefore, multicenter studies with larger populations are needed to further validate this result.
In our study, diagnosis of PBC at a young age was significantly associated with higher risk for CBC.Many other studies have shown similar results. 3,4To examine whether young patients with breast cancer are at risk for developing CBC regardless of family history or BRCA variation, we performed a separate multivariate Cox proportional hazards regression analysis with 1506 patients who underwent a BRCA test.Age of 35 years or younger at surgery was an independent factor associated with developing CBC regardless of BRCA variation status and family history, with a hazard ratio of 1.74 (95% CI, 1.09-2.77).However, patients who underwent BRCA testing were not randomized.In Korea, only patients with both ovarian and breast cancer, bilateral breast cancer, family history of breast or ovarian cancer, or a diagnosis of breast cancer before age 40 years were covered by insurance for BRCA testing until 2020.In other words, many patients with test results in this study had a higher chance of BRCA variation; thus, our results may not accurately represent the whole study population.
In our study, patients with Tis-stage cancer had higher risk of developing CBC than those with invasive cancer with lesions less than 5 cm in diameter (eTable 2 in Supplement 1).Similar results were found in a nationwide population study in the Netherlands, 18 which showed that the group with ductal carcinoma in situ had slightly higher risk for CBC compared with the group with invasive carcinoma.Also, in a population study 15 using data from the Surveillance, Epidemiology, and End Results program, diagnosis of ductal carcinoma in situ for PBC was associated with greater risk of CBC compared with invasive cancer for the first 6 years.The proposed reason for these results was more extensive use of systemic therapy, which led to a lower prevalence of CBC in patients with invasive cancer.
Our study included a relatively large population to show incidence and hazard rates for developing CBC, especially in young patients who are at higher risk.This could be valuable information for young patients seeking consultation for CBC and CPM.

Strengths and Limitations
A strength is that the Asan Medical Center is 1 of the largest hospitals in Korea, and approximately 10% of Korean patients with breast cancer undergo surgery at the hospital.This allowed for a relatively large cross-section of the relevant population to be included in our study.Moreover, as a single-center study, there was a relatively uniform approach to pathology and treatment, which may have minimized the variability in these factors and allowed for a more focused analysis.The long follow-up period in our study further strengthens the reliability of the data.
The main limitation of this study was the relatively small number of patients with CBC, especially of those with the ERBB2+ subtypes in the younger group.Additionally, our study included possible selection bias owing to its retrospective and single-center nature.

Conclusions
In this cohort study, we demonstrated that patients aged 35 years or younger with breast cancer, especially those with triple-negative breast cancer, had higher risk for developing CBC compared with older patients.Furthermore, for the HR-/ERBB2+ subtype, younger patients with breast cancer had a distinct timing for developing CBC compared with older patients.These findings might provide valuable information for physicians and could assist in the decision-making process for patients considering CPM.However, further research with larger sample sizes and multicenter studies are warranted to confirm and validate our results.

Figure 3 .
Figure 3. Hazard Rate for Developing Contralateral Breast Cancer (CBC) According to Age in All Patients and in Subgroup Analysis by Subtype 9t is important to identify and educate patients who develop breast cancer at a young age about the risk of developing CBC before deciding on a treatment option.A previous study by some of us9investigating age-related risk factors for developing CBC using a propensity-matched cohort (age <35 vs Ն35 years) concluded that patients younger than 35 years had 2.5 times the risk of CBC development 9ompared with patients aged 35 years or older.In the younger cohort, patients with ERBB2+ subtype PBC and family history of breast cancer had increased risk of developing CBC.Additionally, in the younger cohort, the hazard rate (incidence in a certain time frame) of CBC differed according to the PBC subtype.Patients with ERBB2 overexpression tended to develop CBC in a shorter time interval compared with patients who had other subtypes; patients with the ERBB2+ subtype had the highest incidence of developing CBC at 4.6 years after surgery for PBC, and the incidence among patients with the HR+/ERBB2-subtype peaked at 7.1 years.9Inthis study, we aimed to evaluate the risk of developing CBC in younger patients by comparing incidence of CBC segregated by age at surgery for PBC.In addition, we examined the varying interval for developing CBC in younger patients (age Յ35 years) vs older patients (age >35 years).
1.Among these patients, 1318 (8.11%) were classified in the (9.27%) underwent a BRCA test, among whom 859 (57.04%) were younger patients.Although there was no statistically significant difference in BRCA variation status, a numerically higher percentage of patients in the younger group had BRCA variation than in the older group (137 of 859 [15.95%] vs 82 of 647 [12.67%];P = .07).The clinicopathological characteristics of CBC in patients who developed CBC after PBC are summarized in eTable 1 in Supplement 1.The data for CBC were missing for 2 patients in the younger group and 11 in the older group; therefore, 69 patients in the younger group (5.24%) and 336 patients in the older group (2.25%) were analyzed.There were no significant differences in the histologic grade or the nuclear grade between PBC and CBC in both age groups.There was no significant difference in composition of breast cancer subtype between PBC and CBC in the younger group, and the proportion of patients with HR-/ERBB2-subtype was high (PBC, 25 of 69 [36.23%];CBC, 24 of 69 [34.78%]).In the older group, there was a significant difference in breast cancer subtype, with a decrease in patients with HR+/ERBB2-(PBC, 99 of 336 [29.46%];CBC, 68 of 336 [20.24%]) and an increase in patients with ERBB2+ (PBC, 57 of 336 [16.96%];CBC, 84 of 336 [25.00%]) (P = .002).Patients of both age groups who developed CBC after PBC had a higher T and N stage of PBC compared with the T and N stage of CBC.

Table .
Baseline Characteristics of Patients at Primary Breast Cancer Surgery a Data are presented as the number/total number (percentage) of patients unless otherwise indicated.bFirst-and/or second-degree relative with breast cancer.c A total of 1506 patients underwent a BRCA test.

JAMA Network Open | Oncology Age
-Related Incidence and Peak Occurrence of Contralateral Breast Cancer Figure 2. Cumulative Incidence of Developing Contralateral Breast Cancer (CBC) According to Age in All Patients and in Subgroup Analysis by Subtype for CBC according to the subtype of PBC, especially according to ERBB2 status.In our study, 170 of 1318 patients in the younger group (12.90%) were categorized as having the HR-/ERBB2+ subtype.In this subgroup, 8 patients (4.71%), with median follow-up of 128 months (IQR, 87-163 months), developed CBC, which was a longer observation period compared with the whole study population (median follow-up, 107 months [IQR, 79-145 months] rate