Introduction

Breast cancer is the most common cancer worldwide, affecting 2.3 million people annually and accounting for approximately a quarter of all female cancers [1, 2]. Approximately 70% of patients have cancers that are estrogen receptor (ER)-positive and human epidermal growth factor receptor 2 (HER2)-negative [3]. Notably, screening mammography has increased the detection of small breast cancers, with an approximately 5-fold increase in the incidence of subcentimeter breast cancers (from 12.8 to 66.0 per 100,000 women) [4]. In Japan, approximately 19,000 patients with T1a/b breast cancer annually undergo surgery [5]. Patients with T1a/b breast cancer, almost N0, have an excellent prognosis, with a 10-year breast cancer mortality rate of 3.2–4% [4, 6] and a 10-year recurrent-free survival of over 90% [7]. These clinical outcomes were comparable to those of distant metastasis-free intervals (DMFIs) in particularly low-risk cases using multi-gene assays, such as the 9-year DMFI 96.8% in cases with a recurrence score of ≤ 10 using a 21-gene assay and the 8-year DMFI 97.0% in ultralow-risk cases using the 70-gene signature [8, 9].

Adjuvant endocrine therapy (ET) is a global standard treatment for ER-positive breast cancer, including T1a/b disease [10,11,12,13]. The national comprehensive cancer network (NCCN) guideline classifies adjuvant ET for T1aN0 breast cancer as category 2B, meaning that “based upon lower-level evidence, the NCCN consensus prescribes that the intervention is appropriate.” The national surgical adjuvant breast and bowel project (NSABP)-B21 trial for T1a/bN0 breast cancer demonstrated that the distant metastasis rates were 3.3% in the radiotherapy + placebo group and 1.6% in the radiotherapy + tamoxifen group (P = .28) [14]. However, a previous cohort study of T1 breast cancer reported that the 7-year cumulative incidence of metastasis was 2%, and ET was not a predictor [15]. Therefore, the necessity of adjuvant ET for T1a/bN0 breast cancer remains uncertain.

Several patients discontinue adjuvant ET despite the recommendations of clinical guidelines. For instance, 31–73% of patients discontinued ET at five years of treatment, with a mean adherence rate of 66.2% [16, 17]. Notably, side effects of ET, such as hot flashes, arthralgia, fatigue, mood disturbance, and vaginal bleeding, cause long-term patient distress [18]. Additionally, ET is reported to persistently deteriorate the quality of life (QOL) [19]. Therefore, the selection of patients for adjuvant ET is necessary.

Furthermore, the patient involvement committee of the Japan clinical oncology group (JCOG) Breast Cancer Study Group discussed treatment preferences with patients, and one of the requests from patient advocators was to omit adjuvant ET if the clinical effect was minimal. In Japan, 86.3% of patients with ER-positive and HER2-negative stage I breast cancer underwent adjuvant ET in 2018 [5]. Therefore, we conducted a multicenter cohort study to investigate the long-term prognosis and effect of adjuvant ET in ER-positive and HER2-negative T1a/bN0M0 breast cancer.

Materials and methods

Study population

We retrospectively collected the medical data of patients with ER-positive and HER2-negative T1a/bN0M0 breast cancer who underwent radical surgery from January 2008–December 2012 in 42 JCOG Breast Cancer Study Group institutes. Tumor size and negative nodes were pathologically defined, and microinvasive disease (T1mi) was included in T1a. ER positivity was defined as ≥ 1% staining by immunohistochemistry [20]. We excluded patients who received neoadjuvant systemic therapy and those with confirmed pathogenic BRCA variants.

Statistical analysis

Baseline characteristics are described as numbers and percentages. Fisher’s exact test was used to compare the frequencies of the categorical variables. The primary endpoint was the cumulative incidence of distant metastasis. The secondary endpoints were distant disease-free survival (DDFS), disease-free survival (DFS), overall survival (OS), ipsilateral breast tumor recurrence (IBTR), and contralateral breast cancer. The effect of adjuvant ET was analyzed using Gray’s test model for cumulative incidence and log-rank test for survival. Predictive factors were assessed using the Fine-Gray model and Cox proportional hazards model. Hazard ratios (HRs) were adjusted by age, tumor size, nuclear grade, Ki-67 labeling index, lymphovascular invasion, and treatment. Statistical significance was defined as a two-sided P value < 0.05. All statistical analyses were performed using the SAS statistical software version 9.4 (SAS Institute).

The consensus of the investigators was that adjuvant ET would be the standard of care if the absolute difference in distant metastasis was 3% or greater, and no ET would be the standard of care if the absolute difference was 1% or less. If the difference was between 1 and 3%, a prospective trial would be planned or considered for shared decision-making.

Results

Among the 4914 patients with ER-positive and HER2-negative T1a/bN0M0 breast cancer who underwent surgery from January 2008–December 2012, 4758 were eligible for inclusion in this study (supplementary figure S1). Patient characteristics are presented in Table 1. Half were over 55 years old, 1202 (25.3%) had T1a tumors, and 3450 (72.5%) had undergone breast-conserving surgery. Adjuvant ET was administered to 3991 (83.9%) patients, and luteinizing hormone-releasing hormone (LH-RH) analog to 577 (34.7%) of 1661 premenopausal and 7 (13.0%) of 54 perimenopausal women. Absence of ET history was related to T1a tumor, low Ki-67 labeling index, mastectomy, and absence of radiation therapy history. The reasons for not performing ET were institutional policy and physician’s decision 395 (51.5%), patient preference 200 (26.1%), and co-morbidity 19 (2.5%). The median duration of ET was five years (ranging from 0 to 15). The rate of sufficient adherence, defined as the completion of 4.5 years of ET, was 83.9%. The median follow-up period was 9.2 years.

Table 1 Characteristics of Patients With T1a/b Breast Cancer
Full size table

Cumulative incidence of distant metastasis, DDFS, and OS

Overall, 84 patients developed distant recurrence. The 9-year cumulative incidence of distant metastasis was 1.5% (95% confidence interval [CI], 1.1–1.9%) in the ET group and 2.6% (95% CI, 1.5–4.1%) in the non-ET group (adjusted subdistribution hazard ratio [sHR], 0.54; 95% CI, 0.32–0.93, P = .027) (Fig. 1A). The 9-year DDFSs were 96.2% (95% CI, 95.5–96.8%) and 92.9% (95% CI, 90.5–94.6%) in the ET and non-ET groups, respectively (adjusted HR, 0.51; 95% CI, 0.36–0.71, P < .001) (Fig. 1B). The 9-year DFSs were 93.6% (95% CI, 92.6–94.4%) and 83.5% (95% CI, 80.3–86.2%) in the ET and non-ET groups, respectively (adjusted HR, 0.39; 95% CI, 0.31–0.50, P < .001) (Fig. 1C). The 9-year OSs were 97.0% (95% CI, 96.3–97.5%) and 94.4% (95% CI, 92.2–96.0%), respectively (adjusted HR, 0.57; 95% CI, 0.39–0.83, P = .004 (Fig. 1D).

Fig. 1
figure 1

Disease outcomes were compared according to adjuvant endocrine therapy in patients with T1a/bN0M0 breast cancer. Cumulative incidence of distant metastasis (A), distant disease-free survival (B), disease-free survival (C), and overall survival (D)

Full size image

Risk of distant metastasis and OS

In the multivariate analysis, risk factors for distant metastasis were high-grade, lymphatic invasion, mastectomy, and absence of ET history (Table 2). In contrast, the negative predictors of OS were older age, lymphatic invasion, and absence of ET history (supplementary table S1).

Table 2 Risk of Distant Metastasis in Patients With T1a/b Breast Cancer
Full size table

Distant metastasis according to risk factors

The 9-year cumulative incidences of distant metastasis were 1.5% (95% CI, 1.1–1.9%) and 4.1% (95% CI, 2.1–7.2%) in cases of nuclear grade 1–2 and 3 (Gray’s test P < .001), and 1.2% (95% CI, 0.9–1.6%) and 3.7% (95% CI, 2.2–5.9%) in lymphatic invasion negative and positive cases (Gray’s test P < .001), respectively (Fig. 2A, B). Distant metastatic events were stratified by nuclear grade and lymphatic invasion (Gray’s test P < .001, Fig. 2C). The cumulative incidence of distant metastasis and the 9-year rates with and without ET for each risk group are described in supplementary figure S2 and table S2. The effect of ET was expected, even in the low-risk group. No difference in the 9-year cumulative incidence of distant metastasis according to　tumor size was observed (T1a: 1.7% and 2.7%; T1b: 1.4% and 2.5%, in the ET and non-ET groups, respectively) (supplementary figure S3).

Fig. 2
figure 2

Cumulative incidence of distant metastasis according to risk factors. Nuclear grade (A), lymphatic invasion (B), and both (C). NG, nuclear grade; Ly, lymphatic invasion

Full size image

IBTR and contralateral breast cancer

IBTR was assessed only in patients who underwent breast-conserving surgery and included local recurrence and new primary tumors. The 9-year IBTRs were 1.1% (95% CI, 0.7–1.6%) in the ET group and 6.9% (95% CI, 4.6–9.7%) in the non-ET group (sHR, 0.17; 95% CI, 0.11–0.28, P < .001) (Fig. 3A). The risk factors for IBTR were younger age, high grade, vascular invasion, absence of ET history, and absence of radiation therapy history (supplementary table S3). The 9-year incidences of contralateral breast cancer were 1.4% (95% CI, 1.0–1.8%) and 5.2% (95% CI, 3.6–7.2%), respectively (sHR, 0.33; 95% CI, 0.22–0.49, P < .001) (Fig. 3B).

Fig. 3
figure 3

Cumulative incidence of ipsilateral (A) and contralateral (B) breast cancer according to adjuvant endocrine therapy in patients with T1a/bN0M0 breast cancer

Full size image

Discussion

Among the 4758 patients with ER-positive and HER2-negative T1a/bN0M0 breast cancer, distant metastasis occurred in 84 patients. Adjuvant ET significantly reduced the cumulative incidence of distant metastasis by an absolute difference of 1.1% at 9 years after surgery. Pathologically, the nuclear grade and lymphatic invasion were independent risk factors for distant recurrence.

The prognosis of T1a/bN0M0 breast cancer was previously investigated; however, most studies included all subtypes [6, 7, 15, 21]. These findings consistently indicate that hormone receptor-negative and HER2-positive status are poor prognostic factors. Furthermore, the NSABP-B21 trial did not require ER status for registration, and approximately 80% of the known cases were ER-positive (HER2 status was not reported) [14]. At a median follow-up of 8 years, adding tamoxifen to radiation therapy after lumpectomy reduced distant recurrence by 1.8% in absolute value, without statistical significance.　Moreover, the integrated analysis for ER-positive T1a/bN0M0 breast cancer from three randomized studies (NSABP-B06, B-14, and B-20) revealed that adjuvant tamoxifen improved recurrence-free survival than surgery alone (HR, 0.55; 95% CI, 0.35–0.88) [22]. Furthermore, in a previous cohort study, the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) reported that 5-year adjuvant tamoxifen and aromatase inhibitors reduced recurrence risk for ER-positive breast cancer, including T1 disease (tamoxifen vs. control: rate ratio [RR], 0.60 and aromatase inhibitors vs. tamoxifen: RR 0.76) [23,24,25]. In addition, a previous cohort study of ER-positive and HER2-negative T1N0M0 breast cancer from a Japanese institute reported that adjuvant ET did not improve DFS, DDFS, or OS for T1a/b disease, unlike T1c [26]. However, the median follow-up period in this study was 60 months, which was relatively short. Distant recurrence of ER-positive breast cancer persisted beyond 5 years [27]. Here, the cumulative incidence of distant metastasis was 0.7% and 1.1% at 5 years, and 1.5% and 2.6% at 9 years in the ET and non-ET groups, respectively. The recurrence prevention effect of ET is maintained after the treatment period, and the long-term outcomes should be highlighted for ER-positive breast cancer [24, 28].

Furthermore, because the absolute risk reduction of distant metastasis in all cohorts was minimal, stratification by risk factors might be helpful for treatment decisions. Our findings and previous reports indicated tumor grade and lymphatic invasion as poor prognostic factors [6, 7, 15, 29]. The 9-year cumulative incidence of distant metastasis increased by approximately 2% for each additional risk factor. Unfortunately, almost all patients with two risk factors, especially high nuclear grade, underwent ET, and it was impossible to estimate the effect of ET for each risk factor.

One of the disadvantages of ET is its side effects, which include the following: hot flashes, arthralgia, fatigue, mood disturbance, and vaginal bleeding [18]. These symptoms afflict patients and worsen their QOL in the long term during the ET period [19, 30, 31]. Furthermore, several integrative therapies and medications have been examined, including meditation, relaxation, yoga, massage, and music therapy for mood disturbance; acupuncture, yoga, and duloxetine for aromatase inhibitor-related arthralgia; and meditation, yoga, and exercise for QOL [32,33,34,35,36]. However, the effects of these supportive therapies have not been satisfactory, and approximately half of the patients discontinue ET at 5 years of treatment [16, 17, 37,38,39,40]. Recently, extended ET was offered even to patients with stage I breast cancer [41]. However, long-term ET raises concerns about further adverse events, including the incidence of endometrial cancer and pulmonary embolism with tamoxifen [42], osteoporosis, bone fracture, arthralgia, myalgia, and cardiovascular events with aromatase inhibitors [43, 44].

Interestingly, adjuvant ET also reduced IBTR (including local recurrence and second cancer) and contralateral breast cancer [14, 25, 45]. In addition, extended ET further reduced contralateral breast cancer [42, 46,47,48], equivalent to preventing the incidence of breast cancer [49]. Moreover, IBTR and contralateral breast cancer were more frequent than distant metastasis (6.9%, 5.2%, and 2.6% at 9 years after surgery). Therefore, clinicians should inform patients of the risk of developing a second breast cancer.

The patient involvement committee in JCOG was established in 2018. The committee requested the omission of excessive ET for breast cancer with an excellent prognosis because patients suffer from the side effects of long-term ET. This study determined the incidence of distant metastasis based on the risk of recurrence and the effectiveness of adjuvant ET but failed to conclude on a recommendation of adjuvant ET based on the predefined consensus. After several discussions in the committee, patients’ expected outcomes were diverse, with distant metastasis, survival, IBTR, and contralateral breast cancer all being important and prioritized by each individual. Therefore, the committee recommended that our findings should be used for shared decision-making for treatment selection rather than conducting a further prospective study.

Limitations and strengths

The strengths of this study include its large patient cohort with long-term follow-up and reliable data based on individual medical records. However, this study suffered from biases associated with retrospective studies, such as inherent selection bias. In addition, mastectomy was one of the potential risk factors for distant metastasis. Previous large, randomized trials have demonstrated equal survival for mastectomy and breast-conserving therapy in early breast cancer [50,51,52]. However, a population-based study reported mastectomy as a poor prognostic factor because of the lack of an additional value of radiation therapy [53]. This study did not identify the cause of this finding. Therefore, the analysis was designed to measure associations between patient variables, treatments, and outcomes.

Conclusions

The prognosis for patients with ER-positive and HER2-negative T1a/bN0M0 breast cancer was as favorable as that for genetically low-risk breast cancer and was stratified by clinical risks. Adjuvant ET improved the incidence of distant metastasis and overall survival with a small absolute risk difference. These findings support shared decision-making that considers the implementation of adjuvant ET by weighing the prevention of recurrence and second cancer, side effect, and QOL.