This cohort study of patients with breast cancer in South Korea examines differences in patterns of locoregional recurrence according to the molecular subtype of the tumor.
Key Points
Question
What are the patterns of locoregional recurrence in patients with breast cancer stratified by molecular subtype?
Findings
In this cohort study of 16 462 Korean patients, locoregional recurrence occurred with different patterns according to breast cancer subtypes. Patients aged 40 years or younger had greater differences in ipsilateral breast tumor recurrence, regional recurrence, and contralateral breast cancer patterns among subtypes than did patients older than 40 years.
Meaning
The findings suggest that tailoring breast cancer surveillance should be recommended regarding differences in locoregional recurrence patterns according to tumor subtypes, particularly for younger patients.
Abstract
Importance
While numerous studies have consistently reported that the molecular subtypes of breast cancer (BC) are associated with different patterns of distant metastasis, few studies have investigated the association of tumor subtypes with locoregional recurrence.
Objective
To investigate the patterns of ipsilateral breast tumor recurrence (IBTR), regional recurrence (RR), and contralateral BC (CBC) according to tumor subtypes.
Design, Setting, and Participants
This retrospective cohort study used the clinical records of patients who underwent BC surgery at a single institution in South Korea between January 2000 and December 2018. Data were analyzed from May 1, 2019, to February 20, 2023.
Exposures
Ipsilateral breast tumor recurrence, RR, and CBC events.
Main Outcomes and Measures
The primary outcome was differences in annual incidence patterns of IBTR, RR, and CBC according to tumor subtypes. Hormone receptor (HR) status was assessed by immunohistochemical staining assay, and ERBB2 status was evaluated according to American Society of Clinical Oncology and College of American Pathologists guidelines.
Results
A total of 16 462 female patients were included in the analysis (median age at time of operation, 49.0 years [IQR, 43.0-57.0 years]). The 10-year IBTR-, RR-, and CBC-free survival rates were 95.9%, 96.1%, and 96.5%, respectively. On univariate analysis, HR−/ERBB2+ tumors had the worst IBTR-free survival (vs HR+/ERBB2− subtype: adjusted hazard ratio, 2.95; 95% CI, 2.15-4.06), while the HR−/ERBB2– subtype had the worst RR- and CBC-free survival among all subtypes (vs HR+/ERBB2− subtype, RR: adjusted hazard ratio, 2.95; 95% CI, 2.37-3.67; CBC: adjusted hazard ratio, 2.12; 95% CI, 1.64-2.75). Subtype remained significantly associated with recurrence events in Cox proportional hazards regression analysis. Regarding the annual recurrence pattern, the IBTR patterns of HR−/ERBB2+ and HR−/ERBB2− subtypes showed double peaks, while HR+/ERBB2− tumors showed a steadily increasing pattern without distinguishable peaks. Additionally, the HR+/ERBB2− subtype seemed to have a steady RR pattern, but other subtypes showed the highest RR incidence at 1 year following surgery, which then gradually decreased. The annual recurrence incidence of CBC gradually increased among all subtypes, and patients with the HR−/ERBB2− subtype had a higher incidence than patients with other subtypes over 10 years. Younger patients (age ≤40 years) had greater differences in IBTR, RR, and CBC patterns between subtypes than did older patients.
Conclusions and Relevance
In this study, locoregional recurrence occurred with different patterns according to BC subtypes, with younger patients having greater differences in patterns among subtypes than older patients. The findings suggest that tailoring surveillance should be recommended regarding differences in locoregional recurrence patterns according to tumor subtypes, particularly for younger patients.
Introduction
Breast cancer (BC) is the leading cause of newly diagnosed malignant tumors and cancer-related death in women worldwide.1 Locoregional recurrence after initial treatment is associated with higher BC mortality.2,3 Accordingly, BC surveillance, including mammography, focuses on early detection of recurrence before clinical symptoms develop.4,5 Moreover, early treatment of locoregional recurrence is also well known to be associated with improved prognosis.6,7,8 Major guidelines recommend annual breast mammography for BC surveillance.9,10,11
Identifying tumor subtypes is essential for decision-making about treatments and prognosis; thus, most institutions routinely conduct immunohistochemical (IHC) tests.12 Interestingly, BC subtypes are correlated with recurrence in that the cumulative recurrence rates of the luminal subtype are more favorable than those of other subtypes.3,13,14 Additionally, the annual hazard ratio of recurrence, including distant metastasis, varies over time after initial treatment according to subtype.15,16 Patients with the ERBB2-enriched subtype and basal-like BC had higher recurrence rates than did those with the luminal subtype at 12 months after surgery, but rates declined dramatically after that.17 In contrast, the recurrence incidence of the luminal subtype showed a steadily increasing pattern exceeding that of other subtypes after 3 years of treatment.17
Nevertheless, current major guidelines do not include tailored postoperative surveillance for tumor subtypes.9,10,11 Furthermore, few studies have reported the annual pattern of locoregional recurrence according to tumor subtypes. We retrospectively investigated the patterns of ipsilateral breast tumor recurrence (IBTR), regional recurrence (RR), and contralateral breast cancer (CBC) with a large number of patients with BC and long-term follow-up at a single institution.
Methods
Patients
This cohort study was approved by the institutional review board of Seoul National University Hospital in South Korea. All procedures were performed following the Declaration of Helsinki,18 and the requirement for informed consent was waived because this was a retrospective study that had no potential harm to the included patients. The study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.19 We retrospectively reviewed the clinicopathological records associated with BC survival of all patients who underwent surgery for BC between January 2000 and December 2018.20 We used the database of our institution, which collects comprehensive data from a single institution.21 Patients with phyllodes tumors, synchronous or metachronous malignant tumors, and secondary BC were excluded. In addition, we excluded patients with bilateral breast cancer to minimize potential bias, as it is difficult to distinguish in-breast recurrence as either IBTR or CBC and axillary lymph node metastasis as either RR or distant metastasis. Lastly, patients with indistinguishable subtypes were excluded (eFigure 1 in Supplement 1). Age was categorized with a cutoff of 40 years due to the small number of patients younger than 35 years, which is a generally accepted cutoff for categorizing women as younger.22,23
Pathologic Assessment
Positive hormone receptor (HR) was defined as 1% or more of stained cells with estrogen or progesterone receptor on IHC staining assay. The ERBB2 status was evaluated according to the American Society of Clinical Oncology and College of American Pathologists guidelines.24 In the case of equivocal results on IHC assay, the specimen was reevaluated with fluorescence in situ hybridization (FISH). Patients with 2+ on IHC assay for ERBB2 but who did not undergo FISH or silver in situ hybridization (SISH) testing were excluded. A Ki-67–labeling index of less than 10% was regarded as low according to a previous study conducted in our institution.25
Definition of Recurrence
We classified the recurrence events into IBTR, RR, and CBC. Ipsilateral breast tumor recurrence was confined to the first recurrence in the ipsilateral breast. Accordingly, only patients who underwent breast-conserving surgery were analyzed for IBTR. Regional recurrence events included any recurrence at the ipsilateral chest wall, mastectomy scar, skin of the breast, or regional lymph nodes including ipsilateral axillary, supraclavicular, infraclavicular, or internal mammary lymph nodes. Recurrence-free survival was defined as the interval between the date of surgery and the date of pathologic or radiologic confirmation of recurrence. Considering the competing risk, distant metastasis without prior IBTR, RR, or CBC was treated as censored at the time of occurrence. The maximum observation period was set at up to 10 years following surgery since no further surveillance was performed for most patients who were disease-free for 10 years after surgery in our institution.
Statistical Analysis
The continuous variables were compared using the Kruskal-Wallis test and post hoc Mann-Whitney U test. The Pearson χ2 test was used for categorical variables. The Cox proportional hazards regression model was used for multivariate analysis and to estimate the adjusted hazard ratios. Variables with a 2-sided P value <.05 in the univariate analysis were included in multivariate analysis, and those that violated the proportional hazards assumption or showed multicollinearity were excluded. The proportional hazards assumption was assessed by the Schoenfeld residual test, and a variance inflation factor greater than 4.0 was considered to indicate the presence of multicollinearity. Missing data were addressed by applying a complete case analysis approach. The goodness of fit of the model was assessed using the Hosmer-Lemeshow test, with 2-sided P > .05 indicating a good fit. All analyses were performed using SPSS, version 26.0 (IBM), and the Kaplan-Meier curves were drawn using GraphPad Prism, version 8.0 (GraphPad Software). Patterns of annual recurrence incidence were smoothed with the kernel smoothing method using the ksmooth function in R, version 3.6.3 (R Project for Statistical Computing).26 Statistical significance was set at 2-sided P < .05. Data were analyzed from May 1, 2019, to February 20, 2023.
Results
Patients
We identified 16 462 female patients who met the inclusion criteria. The median age at the time of operation was 49.0 years (IQR, 43.0-57.0 years) (Table 1). Patients underwent mammography and breast sonography with a median interval of 12.0 months (IQR, 10.7-13.3 months) and 10.1 months (IQR, 7.5-12.1 months), respectively. Approximately 73.8% and 50.4% of patients underwent mammography and breast sonography more than once a year, respectively. During the median follow-up period of 73.7 months (IQR, 46.3-116.2 months), RR occurred in 466 (2.8%) and CBC in 325 (2.5%) of all patients, and IBTR occurred in 286 of 10 313 patients (2.8%) who underwent breast-conserving surgery.
Table 1. Demographic and Clinicopathological Characteristics of Patients.
| Characteristic | Patients, by breast cancer subtypea | ||||
|---|---|---|---|---|---|
| All (N = 16 462) | HR+/ERBB2− (n = 10 075) | HR+/ERBB2+ (n = 1846) | HR−/ERBB2+ (n = 1908) | HR−/ERBB2− (n = 2633) | |
| Age at operation, median (IQR), y | 49.0 (43.0-57.0) | 48.0 (43.0-57.0) | 48.0 (42.0-56.0) | 52.0 (46.0-58.0) | 50.0 (42.0-57.0) |
| BMI, median (IQR) | 23.1 (21.2-25.4) | 23.1 (21.1-25.5) | 22.9 (21.1-25.2) | 23.4 (21.5-25.4) | 23.3 (21.5-25.6) |
| Surgeon | |||||
| A | 7780 (47.3) | 4500 (44.7) | 890 (48.2) | 952 (49.9) | 1439 (54.6) |
| B | 6443 (39.1) | 4121 (40.9) | 724 (39.2) | 682 (35.7) | 916 (37.8) |
| C | 1303 (7.9) | 853 (8.5) | 144 (7.8) | 144 (7.5) | 162 (6.2) |
| Other | 936 (5.7) | 601 (6.0) | 88 (4.8) | 130 (6.8) | 117 (4.4) |
| Year of operation | |||||
| 2000-2009 | 5478 (33.3) | 2989 (29.7) | 607 (32.9) | 741 (38.8) | 1141 (43.3) |
| 2010-2018 | 10 984 (66.7) | 7086 (70.3) | 1239 (67.1) | 1167 (61.2) | 1492 (56.7) |
| Breast operation | |||||
| Breast conserving | 10 313 (62.6) | 6618 (65.7) | 996 (54.0) | 918 (48.1) | 1781 (67.6) |
| Mastectomy | 6149 (37.4) | 3457 (34.3) | 850 (46.0) | 990 (51.9) | 852 (32.4) |
| Axillary operation | |||||
| Sentinel lymph node biopsy | 8419 (51.1) | 5441 (54.0) | 873 (47.3) | 868 (45.5) | 1237 (47.0) |
| Axillary lymph node dissection | 6543 (39.7) | 3604 (35.8) | 825 (44.7) | 896 (47.0) | 1218 (46.3) |
| Not done | 413 (2.5) | 326 (3.2) | 38 (2.1) | 33 (1.7) | 16 (0.6) |
| Unknown | 1087 (6.6) | 704 (7.0) | 110 (6.0) | 111 (5.8) | 162 (6.2) |
| T stageb | |||||
| Tis | 1549 (9.4) | 936 (9.3) | 239 (12.9) | 289 (15.1) | 85 (3.2) |
| T1 | 8597 (52.2) | 5354 (53.1) | 947 (51.3) | 968 (50.7) | 1328 (50.4) |
| T2 | 5531 (33.6) | 3329 (33.0) | 566 (30.7) | 558 (29.2) | 1078 (40.9) |
| T3-4 | 733 (4.5) | 432 (4.3) | 86 (4.7) | 85 (4.5) | 130 (4.9) |
| Unknown | 52 (0.3) | 24 (0.2) | 8 (0.4) | 8 (0.4) | 12 (0.5) |
| N stageb | |||||
| N0 | 10 809 (65.7) | 6459 (64.1) | 1143 (61.9) | 1330 (69.7) | 1877 (71.3) |
| N1 | 3579 (21.7) | 2329 (23.1) | 413 (22.4) | 350 (18.3) | 487 (18.5) |
| N2 | 1031 (6.3) | 622 (6.2) | 149 (8.1) | 113 (5.9) | 147 (5.6) |
| N3 | 507 (3.1) | 268 (2.7) | 83 (4.5) | 66 (3.5) | 90 (3.4) |
| Unknown | 536 (3.3) | 397 (3.9) | 58 (3.1) | 49 (2.6) | 32 (1.2) |
| Lymphovascular invasion | |||||
| Present | 4397 (26.7) | 2687 (26.7) | 580 (31.4) | 477 (25.0) | 653 (24.8) |
| Absent | 10 052 (61.1) | 6373 (63.3) | 985 (53.4) | 1059 (55.5) | 1635 (62.1) |
| Unknown | 2013 (12.2) | 1015 (10.1) | 281 (15.2) | 372 (19.5) | 345 (13.1) |
| Ki-67 index, % | |||||
| ≥10 | 4212 (25.6) | 1222 (12.1) | 639 (34.6) | 819 (42.9) | 1532 (58.2) |
| <10 | 12 022 (73.0) | 8719 (86.5) | 1181 (64.0) | 1058 (55.5) | 1064 (40.4) |
| Unknown | 228 (1.4) | 137 (1.3) | 26 (1.4) | 31 (1.6) | 37 (1.4) |
| Histologic grade, No./total No. (%)c | |||||
| I-II | 8252/14 861 (55.5) | 6582/9115 (72.2) | 733/1599 (45.8) | 435/1611 (27.0) | 502/2536 (19.8) |
| III | 5648/14 861 (38.0) | 1981/9115 (21.7) | 780/1599 (48.8) | 1038/1611 (64.4) | 1849/2536 (72.9) |
| Unknown | 961/14 861 (6.5) | 552/9115 (6.1) | 86/1599 (5.4) | 138/1611 (8.6) | 185/2536 (7.3) |
| Resection margin, No./total No. (%)d | |||||
| Clear | 8412/10 313 (81.6) | 5279/6618 (79.8) | 812/996 (81.5) | 757/918 (82.5) | 1564/1781 (87.8) |
| Close or involved | 1834/10 313 (17.8) | 1301/6618 (19.7) | 175/996 (17.6) | 149/918 (16.2) | 209/1781 (11.7) |
| Unknown | 67/10 313 (0.6) | 38/6618 (0.6) | 9/996 (0.9) | 12/918 (1.3) | 8/1781 (0.4) |
| Neoadjuvant chemotherapy | |||||
| Administered | 2674 (16.2) | 1119 (11.1) | 494 (26.8) | 439 (23.0) | 622 (23.6) |
| Not administered | 13 788 (83.8) | 8956 (88.9) | 1352 (73.2) | 1469 (77.0) | 2011 (76.4) |
| Adjuvant chemotherapy | |||||
| Administered | 6955 (42.2) | 3615 (35.9) | 851 (46.1) | 916 (48.0) | 1573 (59.7) |
| Not administered | 8244 (50.1) | 5613 (55.7) | 866 (46.9) | 876 (45.9) | 889 (33.8) |
| Unknown | 1263 (7.7) | 847 (8.4) | 129 (7.0) | 116 (6.1) | 171 (6.5) |
| Adjuvant radiotherapy | |||||
| Administered | 11 093 (67.4) | 6894 (68.4) | 1225 (66.4) | 1058 (55.5) | 1916 (72.8) |
| Not administered | 4634 (28.1) | 2689 (26.7) | 560 (30.3) | 771 (40.4) | 614 (23.3) |
| Unknown | 735 (4.5) | 492 (4.9) | 61 (3.3) | 79 (4.1) | 103 (3.9) |
| Adjuvant hormonal treatment | |||||
| Administered | 11 368 (69.1) | 9620 (95.5) | 1677 (90.8) | 29 (1.5) | 42 (1.6) |
| Not administered | 5008 (30.4) | 388 (3.9) | 151 (8.2) | 1879 (98.5) | 2590 (98.4) |
| Unknown | 86 (0.5) | 67 (0.7) | 18 (1.0) | 0 | 1 (0.0) |
| ERBB2-targeted treatment | |||||
| Administered | 1894 (11.5) | 0 | 1001 (54.2) | 893 (46.8) | 0 |
| Not administered | 14 568 (88.5) | 10 075 (100) | 845 (45.8) | 1015 (53.2) | 2633 (100) |
| Interval of mammography, median (IQR), mo | 12.0 (10.7-13.3) | 12.0 (10.6-13.3) | 12.1 (10.9-13.5) | 12.1 (10.8-13.5) | 12.0 (10.8-13.5) |
| Interval of clinic visits, median (IQR), mo | |||||
| ≤5 y After surgery | 6.0 (4.2-8.5) | 5.8 (4.1-8.2) | 6.2 (4.4-8.6) | 6.5 (4.5-9.1) | 6.3 (4.3-9.1) |
| >5 y After surgery | 35.6 (24.0-61.2) | 36.1 (24.0-61.6) | 34.3 (24.0-60.0) | 36.8 (24.0-61.8) | 32.4 (24.0-60.0) |
| IBTR, No./total No. (%)d | 286/10 313 (2.8) | 137/6618 (2.1) | 33/996 (3.3) | 53/918 (5.8) | 63/1781 (3.5) |
| RR | 466 (2.8) | 187 (1.9) | 54 (2.9) | 82 (4.3) | 143 (5.4) |
| CBC | 325 (2.0) | 167 (1.7) | 33 (1.8) | 35 (1.8) | 90 (3.4) |
| Follow-up period, median (IQR), mo | 73.7 (46.3-116.2) | 74.4 (48.4-114.8) | 69.8 (43.3-110.1) | 72.0 (43.2-116.0) | 76.0 (41.1-120.0) |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); CBC, contralateral breast cancer; IBTR, ipsilateral breast tumor recurrence; RR, regional recurrence; Tis, tumor in situ.
Data are presented as the number (percentage) of patients unless otherwise indicated.
Stratified according to the AJCC Cancer Staging Manual, Eighth Edition TNM stage. Patients who received neoadjuvant chemotherapy were evaluated clinically.
Among patients with invasive cancer.
Among patients who underwent breast-conserving surgery.
When examining tumor subtypes, the HR+/ERBB2−, HR+/ERBB2+, HR−/ERBB2+, and HR−/ERBB2− constituted 61.2%, 11.2%, 11.6%, and 16.0% of the sample, respectively. Patients with the HR+/ERBB2− subtype had less aggressive pathologic features, such as lower T/N stage, histologic grade, and Ki-67 level, and less lymphovascular invasion (eTable 1 in Supplement 1). They also received significantly less adjuvant and neoadjuvant chemotherapy compared with patients who had other subtypes. Patients with the HR−/ERBB2− subtype required more adjuvant chemotherapy (vs HR+/ERBB2−: odds ratio [OR], 2.75 [95% CI, 2.51-3.01]; vs HR+/ERBB2+: OR, 1.80 [95% CI, 1.59-2.04]; vs HR−/ERBB2+: OR: 1.69 [95% CI, 1.49-1.92]) and radiotherapy (vs HR+/ERBB2−: OR, 1.22 [95% CI, 1.10-1.35]; vs HR+/ERBB2+: OR, 1.43 [95% CI, 1.25-1.63]; vs HR−/ERBB2+: OR, 2.27 [95% CI, 2.00-2.59]). Adjuvant hormonal treatment was administered to 95.5% and 90.8% of the groups with HR+/ERBB2− and HR+/ERBB2+ subtypes, respectively. While none of the patients with the ERBB2− subtype received ERBB2-targeted treatment, it was administered to 54.2% and 46.8% of the groups with HR+/ERBB2+ and HR−/ERBB2+ subtypes, respectively.
Survival Outcomes According to Tumor Subtypes
The 10-year IBTR-, RR-, and CBC-free survival rates were 95.9%, 96.1%, and 96.5%, respectively. The IBTR-free survival rate was significantly better for patients with the HR+/ERBB2− subtype than for patients with other subtypes, while those with the HR−/ERBB2+ subtype had the worst survival (vs HR+/ERBB2−: adjusted hazard ratio, 2.95; 95% CI, 2.15-4.06) (Figure 1A). In contrast, patients with the HR−/ERBB2− subtype had the worst RR-free survival among all subtypes (vs HR+/ERBB2−: adjusted hazard ratio, 2.95; 95% CI, 2.37-3.67) (Figure 1B). Similarly, CBC-free survival among patients with the HR−/ERBB2− subtype was significantly poorer than among patients with other subtypes (vs HR+/ERBB2−: adjusted hazard ratio, 2.12; 95% CI, 1.64-2.75), but there were no significant differences among the other subtypes (Figure 1C).
Figure 1. Kaplan-Meier Curves for Recurrence-Free Survival.
Hazard ratios and 95% CIs were adjusted for surgeon and year of surgery. Reference is the HR+/ERBB2− subtype. HR indicates hormone receptor.
After adjusting for clinicopathological variables affecting recurrence, older age at operation was an independent factor associated with lower IBTR and CBC (Table 2). In addition, both IBTR and RR were significantly associated with lymphovascular invasion and administration of radiotherapy. There was a significant association between the HR−/ERBB2+ subtype and worse IBTR-free survival, and patients with the HR−/ERBB2− subtype had the worst RR- and CBC-free survival among tumor subtypes.
Table 2. Univariate and Multivariate Analyses for Ipsilateral Breast Tumor Recurrence–, Regional Recurrence–, and Contralateral Breast Cancer–Free Survivala.
| Characteristic | Univariate analysis | Multivariate analysisb | ||
|---|---|---|---|---|
| Hazard ratio (95% CI) | P value | Hazard ratio (95% CI) | P value | |
| Ipsilateral breast tumor recurrence | ||||
| Initial age, yc | 0.96 (0.94-0.97) | <.001 | 0.97 (0.95-0.98) | <.001 |
| BMIc | 0.94 (0.90-0.98) | .002 | 0.97 (0.93-1.02) | .21 |
| T staged | ||||
| Tis | 1 [Reference] | .009 | 1 [Reference] | .02 |
| T1 | 0.64 (0.45-0.90) | 0.54 (0.29-0.98) | ||
| T2-4 | 0.87 (0.60-1.24) | 0.77 (0.41-1.46) | ||
| N staged | ||||
| N0 | 1 [Reference] | .55 | NA | NA |
| N1-3 | 1.09 (0.83-1.43) | NA | ||
| Lymphovascular invasion | ||||
| Absent | 1 [Reference] | <.001 | 1 [Reference] | .001 |
| Present | 1.90 (1.44-2.50) | 1.70 (1.25-2.30) | ||
| Ki-67 index, % | ||||
| <10 | 1 [Reference] | <.001 | 1 [Reference] | .17 |
| ≥10 | 1.86 (1.44-2.40) | 1.26 (0.91-1.74) | ||
| Resection margin | ||||
| Clear | 1 [Reference] | <.001 | 1 [Reference] | <.001 |
| Close or involved | 2.02 (1.54-2.65) | 1.77 (1.29-2.44) | ||
| Tumor subtype | ||||
| HR+/ERBB2− | 1 [Reference] | <.001 | 1 [Reference] | <.001 |
| HR+/ERBB2+ | 1.68 (1.15-2.46) | 1.49 (0.93-2.40) | ||
| HR−/ERBB2+ | 2.95 (2.15-4.06) | 3.08 (2.01-4.63) | ||
| HR−/ERBB2– | 1.75 (1.29-2.36) | 1.84 (1.24-2.71) | ||
| Neoadjuvant chemotherapy | ||||
| Administered | 1 [Reference] | .02 | 1 [Reference] | .10 |
| Not administered | 0.70 (0.51-0.95) | 0.74 (0.51-1.07) | ||
| Adjuvant chemotherapy | ||||
| Administered | 1 [Reference] | .78 | NA | NA |
| Not administered | 0.97 (0.76-1.23) | NA | ||
| Adjuvant radiotherapy | ||||
| Administered | 1 [Reference] | <.001 | 1 [Reference] | <.001 |
| Not administered | 2.70 (1.95-3.73) | 2.52 (1.68-3.77) | ||
| Adjuvant hormonal treatmente | ||||
| Administered | 1 [Reference] | <.001 | NA | NA |
| Not administered | 2.16 (1.71-2.73) | NA | ||
| ERBB2-targeted treatment | ||||
| Administered | 1 [Reference] | .18 | NA | NA |
| Not administered | 0.78 (0.54-1.13) | NA | ||
| Regional recurrence | ||||
| Initial age, y | 0.98 (0.97-0.99) | <.001 | 0.99 (0.97-1.00) | .004 |
| BMI | 1.01 (0.99-1.04) | .34 | NA | NA |
| T staged | ||||
| Tis | 1 [Reference] | <.001 | 1 [Reference] | <.001 |
| T1 | 1.91 (1.12-3.24) | 2.37 (0.58-9.76) | ||
| T2-4 | 4.64 (2.76-7.81) | 4.92 (1.19-20.35) | ||
| N staged | ||||
| N0 | 1 [Reference] | <.001 | 1 [Reference] | <.001 |
| N1-3 | 2.67 (2.21-3.21) | 1.64 (1.27-2.10) | ||
| Lymphovascular invasion | ||||
| Absent | 1 [Reference] | <.001 | 1 [Reference] | <.001 |
| Present | 3.40 (2.77-4.17) | 2.41 (1.88-3.08) | ||
| Ki-67 index, % | ||||
| <10 | 1 [Reference] | <.001 | 1 [Reference] | .004 |
| ≥10 | 2.49 (2.07-3.01) | 1.40 (1.11-1.77) | ||
| Resection margin | ||||
| Clear | 1 [Reference] | .17 | NA | NA |
| Close or involved | 0.79 (0.57-1.10) | NA | ||
| Tumor subtype | ||||
| HR+/ERBB2− | 1 [Reference] | <.001 | 1 [Reference] | <.001 |
| HR+/ERBB2+ | 1.64 (1.21-2.22) | 1.18 (0.82-1.69) | ||
| HR−/ERBB2+ | 2.37 (1.83-3.08) | 1.98 (1.43-2.73) | ||
| HR−/ERBB2− | 2.95 (2.37-3.67) | 3.03 (2.31-3.98) | ||
| Neoadjuvant chemotherapy | ||||
| Administered | 1 [Reference] | <.001 | 1 [Reference] | <.001 |
| Not administered | 0.31 (0.26-0.38) | 0.31 (0.24-0.40) | ||
| Adjuvant chemotherapy | ||||
| Administered | 1 [Reference] | <.001 | 1 [Reference] | .37 |
| Not administered | 0.49 (0.40-0.60) | 1.13 (0.87-1.48) | ||
| Adjuvant radiotherapy | ||||
| Administered | 1 [Reference] | <.001 | 1 [Reference] | <.001 |
| Not administered | 1.88 (1.55-2.26) | 2.93 (2.33-3.68) | ||
| Adjuvant hormonal treatmente | ||||
| Administered | 1 [Reference] | <.001 | NA | NA |
| Not administered | 2.55 (2.12-3.06) | NA | ||
| ERBB2-targeted treatment | ||||
| Administered | 1 [Reference] | .08 | NA | NA |
| Not administered | 0.78 (0.59-1.03) | NA | ||
| Contralateral breast cancer | ||||
| Initial age, y | 0.98 (0.97-1.00) | .005 | 0.99 (0.97-1.00) | .02 |
| BMI | 0.97 (0.93-1.00) | .050 | 0.97 (0.94-1.01) | .13 |
| T staged | ||||
| Tis | 1 [Reference] | .87 | NA | NA |
| T1 | 0.91 (0.63-1.33) | NA | ||
| T2-4 | 0.95 (0.64-1.39) | NA | ||
| N staged | ||||
| N0 | 1 [Reference] | .37 | NA | NA |
| N1-3 | 0.90 (0.70-1.14) | NA | ||
| Lymphovascular invasion | ||||
| Absent | 1 [Reference] | .69 | NA | NA |
| Present | 1.06 (0.81-1.37) | NA | ||
| Ki-67 index, % | ||||
| <10 | 1 [Reference] | .06 | NA | NA |
| ≥10 | 1.26 (0.99-1.62) | NA | ||
| Resection margin | ||||
| Clear | 1 [Reference] | .66 | NA | NA |
| Close or involved | 1.08 (0.77-1.53) | NA | ||
| Tumor subtype | ||||
| HR+/ERBB2− | 1 [Reference] | <.001 | 1 [Reference] | <.001 |
| HR+/ERBB2+ | 1.14 (0.79-1.66) | 1.10 (0.75-1.62) | ||
| HR−/ERBB2+ | 1.14 (0.79-1.64) | 1.25 (0.87-1.81) | ||
| HR−/ERBB2− | 2.12 (1.64-2.75) | 2.13 (1.63-2.79) | ||
| Neoadjuvant chemotherapy | ||||
| Administered | 1 [Reference] | .04 | 1 [Reference] | .22 |
| Not administered | 0.73 (0.54-0.99) | 0.82 (0.60-1.12) | ||
| Adjuvant chemotherapy | ||||
| Administered | 1 [Reference] | .80 | NA | NA |
| Not administered | 1.03 (0.82-1.30) | NA | ||
| Adjuvant radiotherapy | ||||
| Administered | 1 [Reference] | .56 | NA | NA |
| Not administered | 1.07 (0.85-1.37) | NA | ||
| Adjuvant hormonal treatmente | ||||
| Administered | 1 [Reference] | <.001 | NA | NA |
| Not administered | 1.83 (1.47-2.29) | NA | ||
| ERBB2-targeted treatment | ||||
| Administered | 1 [Reference] | .26 | NA | NA |
| Not administered | 1.26 (0.85-1.87) | NA | ||
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); NA, not applicable; Tis, tumor in situ.
Hosmer-Lemeshow goodness-of-fit test results: ipsilateral breast tumor recurrence: χ28, 7.55; P = .48; regional recurrence: χ28, 5.79; P = .67; contralateral breast cancer: χ28, 12.04; P = .49. All variables were adjusted for the surgeon and the year of surgery.
Variables with a P value <.05 in the univariate analysis were included in the multivariate analysis.
Analyzed as a continuous variable. The hazard ratio represents the value when the age or BMI increased by 1.
Stratified according to the AJCC Cancer Staging Manual, Eighth Edition TNM stage. Patients after neoadjuvant chemotherapy were evaluated with clinical stage.
Adjuvant hormonal treatment showed multicollinearity for ipsilateral breast tumor recurrence, so it was excluded from multivariate analysis.
Annual Recurrence Pattern According to Tumor Subtypes
The rate of IBTR showed a double-peak pattern, with the first peak at year 2 and the second peak between years 8 and 9 after surgery (Figure 2A). The HR−/ERBB2+ subtype was associated with a higher annual IBTR incidence than other subtypes and had a double-peak pattern. The HR−/ERBB2− subtype showed a similar annual pattern but was associated with lower recurrence incidence compared with the HR−/ERBB2+ subtype. On the contrary, HR+/ERBB2– tumors showed a steadily increasing pattern with indistinguishable peaks.
Figure 2. Patterns of Annual Recurrence Incidence by Tumor Subtype.
Ipsilateral breast tumor recurrence was analyzed for patients who underwent breast-conserving surgery. HR indicates hormone receptor.
In the case of RR, the HR−/ERBB2− subtype showed the highest incidence, with a peak at 1 year after surgery, followed by the HR−/ERBB2+, HR+/ERBB2+, and HR+/ERBB2− subtypes (Figure 2B). The difference in the incidence of recurrence among tumor subtypes was evident until 2 years after surgery. After that, the incidence of all subtypes except the HR+/ERBB2− subtype gradually decreased, showing a similar annual recurrence pattern after 5 years postsurgery. The HR+/ERBB2− subtype showed a stable and increasing pattern of annual RR incidence. Conversely, the annual recurrence rate of CBC gradually increased for all subtypes, and the HR−/ERBB2− subtype showed the highest rate over 10 years (Figure 2C).
Annual Recurrence Pattern According to Age and Tumor Subtypes
In total, 13 556 patients (82.3%) were older than 40 years, and we compared the recurrence patterns between younger (age ≤40 years) and older (age >40 years) patients. The difference in the early recurrence patterns between HR+ and HR− subtypes was larger in the younger age group. In the younger age group, the double peaks of annual IBTR incidence were more prominent for younger patients than for older patients, especially for the HR−/ERBB2+ subtype (Figure 3A and D). Additionally, the incidence of RR decreased during the first 5 years, and all subtypes except the HR−/ERBB2− subtype showed a rebounded increasing pattern thereafter (Figure 3B and E). The annual CBC incidence among younger patients was higher with more diverse patterns than that among older patients (Figure 3C and F). On the contrary, older patients showed a steady annual IBTR and RR incidence after 5 years postsurgery, and the pattern was nearly identical among subtypes.
Figure 3. Patterns of Annual Recurrence Incidence Among Patients Aged 40 Years or Younger or Older Than 40 Years.
Ipsilateral breast tumor recurrence was analyzed for patients who underwent breast-conserving surgery. HR indicates hormone receptor.
Annual Recurrence Pattern According to Adjuvant Treatment
Patients who received ERBB2-targeted treatment had better locoregional recurrence-free survival (hazard ratio, 1.42; 95% CI, 1.08-1.87) but comparable CBC-free survival (hazard ratio, 1.40; 95% CI, 0.86-2.28) compared with those who did not receive this treatment (eFigure 2 in Supplement 1). The annual recurrence pattern indicated that patients treated with ERBB2-targeted treatment had a lower risk of locoregional recurrence until 8 years after surgery (eFigure 3 in Supplement 1). The peaks of the 2 curves were consistent at 2 to 3 years after surgery.
Similarly, compared with patients who were administered hormonal treatment, both locoregional recurrence-free survival (hazard ratio, 2.27; 95% CI, 1.32-3.90) and CBC-free survival (hazard ratio, 2.75; 95% CI, 1.95-3.87) were poorer for those who omitted hormonal treatment (eFigure 2 in Supplement 1). The annual incidence of locoregional recurrence was consistently higher in the hormonal treatment–omission group for 10 years, peaking between 1 and 4 years after surgery (eFigure 3 in Supplement 1).
Recurrence Outcomes Without Treating Distant Metastasis as a Competing Risk
Treating distant metastasis as a competing event could lead to underestimation of events and overestimation in survival analysis. Therefore, we performed post hoc analyses without considering distant metastasis as a competing event and including all events that occurred after competing events. Multivariate analysis showed that IBTR-free survival was still significantly associated with the HR−/ERBB2+ subtype, whereas CBC-free survival was worst for the HR−/ERBB2− subtype (eTable 2 in Supplement 1). The HR−/ERBB2+ subtype showed poorer RR-free survival than did the other subtypes. The annual incidence pattern showed a more pronounced increase in incidence rates in the late period, but the overall trend of the pattern was not significantly different (eFigure 4 in Supplement 1).
Discussion
Locoregional recurrence after initial treatment of BC is associated with higher mortality.2,3 In the current study, we found differences in annual patterns of recurrence incidence according to tumor subtypes in a large cohort with long-term follow-up. Knowledge of the recurrence pattern may help identify the periods of prevalent BC recurrence and optimize the surveillance strategy. For example, physicians may consider surveillance every 6 months for patients with HR− subtypes until 5 years after surgery, while surveillance for patients with HR+ tumors may continue to follow the current guideline of 1-year intervals. For younger individuals, more frequent surveillance may be recommended.
Several studies have reported different locoregional recurrence survival rates among patients with different BC subtypes.27,28,29 Ignatov et al27 analyzed 12 053 patients and showed that patients with ERBB2-enriched and triple-negative BC had significantly poorer locoregional recurrence-free survival than did those with the luminal subtype. Lee et al28 reported significantly higher local, CBC, and nodal recurrence for ERBB2-enriched and triple-negative BC subtypes, and another study29 analyzing 514 patients without lymph node involvement observed similar results. Our study has strengths in that we included a larger cohort than did previous studies and provided the results of recurrence patterns according to a 10-year time frame.
Regarding the recurrence pattern, Kimura et al30 analyzed 2209 patients with 10 years of follow-up and reported that the recurrence pattern peaked at the second year and then gradually decreased. Additionally, patients with HR− subtypes showed double peaks at the second and between the sixth and seventh years after surgery, while patients with HR+ subtypes had a constant recurrence pattern. Demicheli et al31 also found double-peak recurrence patterns for all subtypes. They assumed that the unstable status of micrometastatic foci affected the first peak, while the second peak reflected the effect of adjuvant treatments that delay or suppress recurrence. In the current study, a second peak was present at 8 or 9 years after surgery, while the second peak in previous studies was at 5 or 7 years.30,31 This difference may be due to the considerable development in treatments compared with studies published more than 10 years ago.30,31 Notably, a large-scale study in the Netherlands used registry data to analyze BC recurrence patterns and found that subtype was an important factor associated with recurrence.32 Despite similar results to ours, that study used registry data, included nearly half the number of patients in the cohort in our study, and analyzed patients diagnosed more than a decade ago. Moreover, as both IBTR and RR events were relatively small compared with distant metastasis, a diverse change in patterns of locoregional recurrence was difficult to distinguish.
Patients with the HR−/ERBB2+ subtype had higher IBTR incidence than those with other subtypes in the current study, especially during the first postoperative year. We assumed that the delayed peak at 1 year would have been associated with 1 year of ERBB2-targeted treatment.33 Recurrence patterns after stratifying patients with ERBB2+ subtypes according to the administration of ERBB2-targeted treatment indicated that the significantly increased risk of early recurrence in patients with ERBB2+ subtypes may be associated with the biological properties of tumor subtypes rather than the suppression effect of recurrence during the adjuvant treatments. The crossover between the group that received ERBB2-targeted treatment and the group that did not at 8 years after surgery may be due to the decreased number of patients in the group that omitted ERBB2-targeted treatment, as those patients are at a higher risk of distant metastasis or death.17 Similarly, hormonal therapy was assumed to have been associated with the steady and late recurrence pattern for the HR+ subtype. Although the number of patients was too small to provide strong statistical evidence, these results suggest that administering appropriate adjuvant treatment is critical in reducing recurrences.
Our results showed an increasing and diverse pattern beyond 5 years following surgery for younger patients (age ≤40 years). As younger patients, especially those with HR+ subtypes, are known to develop more distant metastasis than older patients,34 the late recurrence incidence among younger patients was expected to be higher than in the current study. Additionally, our findings showed that IBTR incidence was more prominent in younger patients than in older patients with the HR−/ERBB2+ subtype. The results are consistent with a previous study35 that included 25 284 patients and found that among young patients, those with the HR−/ERBB2+ subtype had significantly more IBTR events than did those with other subtypes. These findings support the need for intensive surveillance for younger patients with the HR− subtype.
To assess the need for different surveillance according to subtypes, we conducted a post hoc analysis to investigate the association between intensive surveillance and early detection of recurrence. To reduce the selection bias raised by the high chance of frequent surveillance in the high-risk group, we analyzed only the patients who experienced recurrence. Survival curves showed that conducting mammography and breast sonography at least twice a year detected IBTR, RR, and CBC events significantly earlier than conducting these imaging examinations less frequently. The result remained consistent for each subtype (eFigures 5 and 6 in Supplement 1). Despite the small number of patients in the subgroups, this finding supports the need for intensive surveillance during high-risk periods after BC treatment.
Limitations
This study has several limitations. First, as IBTR can be classified as either true recurrence or secondary BC, the number of true IBTR events in our study may have been overestimated.36,37 However, since several IBTR events were diagnosed without histologic confirmation and the purpose of this study was to identify the pattern of recurrence for appropriate treatment, we did not differentiate IBTR as a true recurrence or a new tumor. Second, the relatively small number of subgroups may limit the statistical power of the subgroup analysis and the generalizability of our findings. Third, the HR−/ERBB2− subtype distribution at our institution was higher than previously reported in Asian patients.38 This may be due to the role of our institution as a leading national hospital in Korea that treats patients with aggressive BC, and this distribution was similar to that reported in a previous study conducted at a branch institution of Seoul National University Hospital.28 Fourth, excluding patients with 2+ on IHC for ERBB2 but not tested with FISH or SISH may have led to selection bias. Fifth, we included a period of over 20 years, and the results could not reflect the change in modalities of treatments over the years. To address this issue, we conducted a post hoc subgroup analysis for the patients who underwent surgery after 2010. The second peak of the recurrence pattern for several subtypes was more prominent in the late period compared with that of all subtypes, but the overall features were similar to that of overall subtypes (eFigure 7 in Supplement 1). Finally, we conducted a sensitivity analysis excluding distant metastasis as a competing event but could not further adjust for competing risk due to events that may have occurred if patients had not died.
Conclusions
In this cohort study, locoregional recurrence occurred in different patterns according to BC subtypes, with younger patients having greater differences in patterns among subtypes than older patients. Patients with HR− subtypes had high recurrence rates in the early period of treatment, while those with HR+ subtypes had a steady pattern beyond 5 years after surgery. These findings suggest that tailoring surveillance should be recommended regarding differences in locoregional recurrence patterns according to subtypes, particularly for young patients. Validating our study at a multi-institutional level with a large number of patients, taking into account the recent advancement in BC treatments, would be necessary to generalize our findings.
eTable 1. Pairwise Comparison for Demographic and Clinicopathological Characteristics Between Tumor Subtypes
eTable 2. Univariate and Multivariate Analyses for Recurrences Without Treating Distant Metastasis as a Competing Risk
eFigure 1. Flowchart of Study Inclusion
eFigure 2. Kaplan-Meier Curves for Recurrence-Free Survival According to Adjuvant ERBB2-Targeted Treatment and Hormonal Treatment
eFigure 3. Annual Recurrence Pattern According to Adjuvant ERBB2-Targeted Treatment and Hormonal Treatment
eFigure 4. Annual Recurrence Pattern Without Treating Distant Metastasis as a Competing Risk
eFigure 5. Kaplan-Meier Curves for Recurrence-Free Survival According to Frequency of Breast Sonography or Mammography
eFigure 6. Kaplan-Meier Curves for Recurrence-Free Survival According to Breast Cancer Subtypes and Frequency of Breast Sonography or Mammography
eFigure 7. Annual Recurrence Pattern for Breast Cancer Patients Who Underwent Surgery Since 2010
Data Sharing Statement
References
- 1.Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209-249. doi: 10.3322/caac.21660 [DOI] [PubMed] [Google Scholar]
- 2.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin. 2020;70(1):7-30. doi: 10.3322/caac.21590 [DOI] [PubMed] [Google Scholar]
- 3.Clarke M, Collins R, Darby S, et al. ; Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) . Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet. 2005;366(9503):2087-2106. doi: 10.1016/S0140-6736(05)67887-7 [DOI] [PubMed] [Google Scholar]
- 4.Monticciolo DL, Newell MS, Hendrick RE, et al. Breast cancer screening for average-risk women: recommendations from the ACR Commission on Breast Imaging. J Am Coll Radiol. 2017;14(9):1137-1143. doi: 10.1016/j.jacr.2017.06.001 [DOI] [PubMed] [Google Scholar]
- 5.De Rose F, Meduri B, De Santis MC, et al. Rethinking breast cancer follow-up based on individual risk and recurrence management. Cancer Treat Rev. 2022;109:102434. doi: 10.1016/j.ctrv.2022.102434 [DOI] [PubMed] [Google Scholar]
- 6.Houssami N, Ciatto S. Mammographic surveillance in women with a personal history of breast cancer: how accurate? how effective? Breast. 2010;19(6):439-445. doi: 10.1016/j.breast.2010.05.010 [DOI] [PubMed] [Google Scholar]
- 7.Lu WL, Jansen L, Post WJ, Bonnema J, Van de Velde JC, De Bock GH. Impact on survival of early detection of isolated breast recurrences after the primary treatment for breast cancer: a meta-analysis. Breast Cancer Res Treat. 2009;114(3):403-412. doi: 10.1007/s10549-008-0023-4 [DOI] [PubMed] [Google Scholar]
- 8.Paszat L, Sutradhar R, Grunfeld E, et al. Outcomes of surveillance mammography after treatment of primary breast cancer: a population-based case series. Breast Cancer Res Treat. 2009;114(1):169-178. doi: 10.1007/s10549-008-9986-4 [DOI] [PubMed] [Google Scholar]
- 9.Cardoso F, Kyriakides S, Ohno S, et al. ; ESMO Guidelines Committee . Early breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2019;30(8):1194-1220. doi: 10.1093/annonc/mdz173 [DOI] [PubMed] [Google Scholar]
- 10.National Comprehensive Cancer Network . NCCN guidelines for breast cancer, version 8.2021. Published August 30, 2021. Accessed September 21, 2021. http://www.nccn.org/login?
- 11.Runowicz CD, Leach CR, Henry NL, et al. American Cancer Society/American Society of Clinical Oncology breast cancer survivorship care guideline. J Clin Oncol. 2016;34(6):611-635. doi: 10.1200/JCO.2015.64.3809 [DOI] [PubMed] [Google Scholar]
- 12.Duraiyan J, Govindarajan R, Kaliyappan K, Palanisamy M. Applications of immunohistochemistry. J Pharm Bioallied Sci. 2012;4(6)(suppl 2):S307-S309. doi: 10.4103/0975-7406.100281 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ponzone R, Baum M. Loco-regional therapy and breast cancer survival: searching for a link. Breast. 2013;22(4):510-514. doi: 10.1016/j.breast.2012.10.004 [DOI] [PubMed] [Google Scholar]
- 14.Anderson SJ, Wapnir I, Dignam JJ, et al. Prognosis after ipsilateral breast tumor recurrence and locoregional recurrences in patients treated by breast-conserving therapy in five National Surgical Adjuvant Breast and Bowel Project protocols of node-negative breast cancer. J Clin Oncol. 2009;27(15):2466-2473. doi: 10.1200/JCO.2008.19.8424 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Metzger-Filho O, Sun Z, Viale G, et al. Patterns of recurrence and outcome according to breast cancer subtypes in lymph node-negative disease: results from International Breast Cancer Study Group trials VIII and IX. J Clin Oncol. 2013;31(25):3083-3090. doi: 10.1200/JCO.2012.46.1574 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Yin W, Di G, Zhou L, et al. Time-varying pattern of recurrence risk for Chinese breast cancer patients. Breast Cancer Res Treat. 2009;114(3):527-535. doi: 10.1007/s10549-008-0022-5 [DOI] [PubMed] [Google Scholar]
- 17.Park S, Koo JS, Kim MS, et al. Characteristics and outcomes according to molecular subtypes of breast cancer as classified by a panel of four biomarkers using immunohistochemistry. Breast. 2012;21(1):50-57. doi: 10.1016/j.breast.2011.07.008 [DOI] [PubMed] [Google Scholar]
- 18.World Medical Association . World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191-2194. doi: 10.1001/jama.2013.281053 [DOI] [PubMed] [Google Scholar]
- 19.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative . The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453-1457. doi: 10.1016/S0140-6736(07)61602-X [DOI] [PubMed] [Google Scholar]
- 20.Yekedüz E, Dizdar Ö, Kertmen N, Aksoy S. Comparison of clinical and pathological factors affecting early and late recurrences in patients with operable breast cancer. J Clin Med. 2022;11(9):2332. doi: 10.3390/jcm11092332 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Moon HG, Han W, Noh DY. Underweight and breast cancer recurrence and death: a report from the Korean Breast Cancer Society. J Clin Oncol. 2009;27(35):5899-5905. doi: 10.1200/JCO.2009.22.4436 [DOI] [PubMed] [Google Scholar]
- 22.Han W, Kim SW, Park IA, et al. Young age: an independent risk factor for disease-free survival in women with operable breast cancer. BMC Cancer. 2004;4:82. doi: 10.1186/1471-2407-4-82 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kim HJ, Kim S, Freedman RA, Partridge AH. The impact of young age at diagnosis (age <40 years) on prognosis varies by breast cancer subtype: a US SEER database analysis. Breast. 2022;61:77-83. doi: 10.1016/j.breast.2021.12.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wolff AC, Hammond MEH, Allison KH, et al. Human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists Clinical Practice Guideline focused update. J Clin Oncol. 2018;36(20):2105-2122. doi: 10.1200/JCO.2018.77.8738 [DOI] [PubMed] [Google Scholar]
- 25.Jung SY, Han W, Lee JW, et al. Ki-67 expression gives additional prognostic information on St. Gallen 2007 and Adjuvant! online risk categories in early breast cancer. Ann Surg Oncol. 2009;16(5):1112-1121. doi: 10.1245/s10434-009-0334-7 [DOI] [PubMed] [Google Scholar]
- 26.Chacón JE, Duong T. Multivariate Kernel Smoothing and Its Applications. Chapman & Hall/CRC; 2018. doi: 10.1201/9780429485572 [DOI] [Google Scholar]
- 27.Ignatov A, Eggemann H, Burger E, Ignatov T. Patterns of breast cancer relapse in accordance to biological subtype. J Cancer Res Clin Oncol. 2018;144(7):1347-1355. doi: 10.1007/s00432-018-2644-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lee Y, Kang E, Lee AS, et al. Outcomes and recurrence patterns according to breast cancer subtypes in Korean women. Breast Cancer Res Treat. 2015;151(1):183-190. doi: 10.1007/s10549-015-3390-7 [DOI] [PubMed] [Google Scholar]
- 29.Laurberg T, Alsner J, Tramm T, et al. Impact of age, intrinsic subtype and local treatment on long-term local-regional recurrence and breast cancer mortality among low-risk breast cancer patients. Acta Oncol. 2017;56(1):59-67. doi: 10.1080/0284186X.2016.1246803 [DOI] [PubMed] [Google Scholar]
- 30.Kimura M, Yanagita Y, Fujisawa T, Koida T. Study of time-course changes in annual recurrence rates for breast cancer: data analysis of 2209 patients for 10 years post-surgery. Breast Cancer Res Treat. 2007;106(3):407-411. doi: 10.1007/s10549-007-9510-2 [DOI] [PubMed] [Google Scholar]
- 31.Demicheli R, Abbattista A, Miceli R, Valagussa P, Bonadonna G. Time distribution of the recurrence risk for breast cancer patients undergoing mastectomy: further support about the concept of tumor dormancy. Breast Cancer Res Treat. 1996;41(2):177-185. doi: 10.1007/BF01807163 [DOI] [PubMed] [Google Scholar]
- 32.van Maaren MC, de Munck L, Strobbe LJA, et al. Ten-year recurrence rates for breast cancer subtypes in the Netherlands: a large population-based study. Int J Cancer. 2019;144(2):263-272. doi: 10.1002/ijc.31914 [DOI] [PubMed] [Google Scholar]
- 33.Slamon D, Eiermann W, Robert N, et al. ; Breast Cancer International Research Group . Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med. 2011;365(14):1273-1283. doi: 10.1056/NEJMoa0910383 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Anders CK, Hsu DS, Broadwater G, et al. Young age at diagnosis correlates with worse prognosis and defines a subset of breast cancers with shared patterns of gene expression. J Clin Oncol. 2008;26(20):3324-3330. doi: 10.1200/JCO.2007.14.2471 [DOI] [PubMed] [Google Scholar]
- 35.Li Y, Lu S, Zhang Y, Wang S, Liu H. Loco-regional recurrence trend and prognosis in young women with breast cancer according to molecular subtypes: analysis of 1099 cases. World J Surg Oncol. 2021;19(1):113. doi: 10.1186/s12957-021-02214-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Panet-Raymond V, Truong PT, McDonald RE, et al. True recurrence versus new primary: an analysis of ipsilateral breast tumor recurrences after breast-conserving therapy. Int J Radiat Oncol Biol Phys. 2011;81(2):409-417. doi: 10.1016/j.ijrobp.2010.05.063 [DOI] [PubMed] [Google Scholar]
- 37.Smith TE, Lee D, Turner BC, Carter D, Haffty BG. True recurrence vs new primary ipsilateral breast tumor relapse: an analysis of clinical and pathologic differences and their implications in natural history, prognoses, and therapeutic management. Int J Radiat Oncol Biol Phys. 2000;48(5):1281-1289. doi: 10.1016/S0360-3016(00)01378-X [DOI] [PubMed] [Google Scholar]
- 38.DeSantis CE, Ma J, Gaudet MM, et al. Breast cancer statistics, 2019. CA Cancer J Clin. 2019;69(6):438-451. doi: 10.3322/caac.21583 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Pairwise Comparison for Demographic and Clinicopathological Characteristics Between Tumor Subtypes
eTable 2. Univariate and Multivariate Analyses for Recurrences Without Treating Distant Metastasis as a Competing Risk
eFigure 1. Flowchart of Study Inclusion
eFigure 2. Kaplan-Meier Curves for Recurrence-Free Survival According to Adjuvant ERBB2-Targeted Treatment and Hormonal Treatment
eFigure 3. Annual Recurrence Pattern According to Adjuvant ERBB2-Targeted Treatment and Hormonal Treatment
eFigure 4. Annual Recurrence Pattern Without Treating Distant Metastasis as a Competing Risk
eFigure 5. Kaplan-Meier Curves for Recurrence-Free Survival According to Frequency of Breast Sonography or Mammography
eFigure 6. Kaplan-Meier Curves for Recurrence-Free Survival According to Breast Cancer Subtypes and Frequency of Breast Sonography or Mammography
eFigure 7. Annual Recurrence Pattern for Breast Cancer Patients Who Underwent Surgery Since 2010
Data Sharing Statement
