The relationship between serum E2 levels and recurrence in premenopausal,ER-positive breast cancer patients: A retrospective study

Abstract

BACKGROUND:

Among estrogens, estradiol (E2) has the strongest physiological activity as a stimulator in estrogen receptor (ER)-positive breast cancer. The aim of this study is to investigate E2 dynamics during endocrine therapy and to explore the optimal environment in which tamoxifen (TAM) exhibits better efficacy for ER-positive premenopausal early breast cancer patients.

METHODS:

This is a retrospective study enrolled 194 patients with premenopausal ER-positive early-stage breast cancer who aging ≤45 years at onset and receiving luteinizing hormone-releasing hormone-agonist (LHRH-a) and TAM–therapy. Approximately half of the patients also received pre- or post-operative chemotherapy as adjuvant systemic therapy. We studied the relationship between recurrence and serum hormonal dynamics during adjuvant therapy. We monitored the concentrations of E2 and, follicle–stimulating hormone (FSH) in the blood before, during, and after treatment. The median follow-up period was 80 (14–555) months.

RESULTS:

Forty–six (23.7%) patients developed recurrent breast cancer after surgery. The prognoses were favorable in the group receiving longer LHRH-a exposure if those patients did not receive chemotherapy as their adjuvant therapy. Paradoxically, patients with high serum E2 levels after LHRH-a showed a low recurrence ratio. This phenomenon might be explained by the similar mechanisms of estrogen therapy after estrogen depletion by aromatase inhibitor (AI) therapy for metastatic breast cancer.

CONCLUSION:

Among patients who received endocrine therapy without adjuvant chemotherapy, those with longer LHRH-a exposure had favorable prognoses. A potential association was observed between recurrence and E2 concentrations in women with premenopausal ER-positive early-stage breast cancer.

Introduction

Tamoxifen (TAM) is one of the most effective drugs for the treatment of estrogen-receptor (ER)-positive breast cancer. Clinical trials of TAM for chemoprevention of breast cancer in high-risk women have been performed in Europe, the United States, and Oceania [1–4].

TAM is a selective estrogen- receptor- modulator (SERM) that suppresses breast cancer by inhibiting the action of estrogen on breast epithelial cells, and acts as an agonist in uterine and ovarian cells. TAM directly stimulates estrogen release from the ovarium through negative feedback of follicle-stimulating hormone (FSH)-dependent signaling. In addition, serum estradiol (E2) levels increase during TAM therapy in premenopausal breast cancer patients [5].

E2 is known to have the strongest physiological activity among estrogens. Binding of E2 to ERs induces ER association with specific DNA sequences in the promoter of target genes, where it acts as a transcription factor. Whether transcription is up- or down- regulated depends on the organ.

E2 is thought to contribute to the progression of ER-positive breast cancer [6]. However, there is no evidence that increasing E2 levels by endocrine therapy cause breast cancer recurrence. The optimal length of treatment with luteinizing hormone-releasing hormone agonist (LHRH-a) together with TAM is unclear. The purpose of this study was to determine serum E2 dynamics during endocrine therapy and the optimal E2 environment for maximal TAM efficacy.

Patients and methods

Eligibility criteria and treatment

This is a retrospective study enrolling 194 patients with premenopausal, ER-positive (≧1%), early-stage breast cancer who were ≦45 years of age and had received both LHRH-a and TAM therapies between 1987 and 2012 in our hospital. We obtained their comprehensive informed consent before initiating their breast cancer treatment. We excluded patients with their consent sheets not confirmed. The Institutional Review Board in our hospital approved this study.

We monitored E2 and FSH levels in blood before, during, and after treatment and evaluated the relationship between local recurrence or distant metastasis and the duration of drug administration in selected patients (n = 49) who were available. We used their serum to measure hormonal levels of E2 (0.2 ml) and FSH (0.3 ml) using chemiluminescense enzyme immunoassay (CLEIA). We performed measurements immediately after blood collection from patients.

All included patients had confirmed ER-positive (≧1%) breast cancer in at least one breast. Treatment included daily oral TAM (20 mg) and subcutaneous injection of LHRH-a once a month or once every 3 months. Every patient received TAM for at least 5 years, with the duration dependent on their risk of recurrence. The range of LHRH-a treatment duration was 3–63 months (median; 20 months). Patients’ age and adverse events affected the duration of LHRH-a treatment.

Statistical analysis

Time-to-event endpoints (recurrence-free survival, RFS) were analyzed using the Kaplan–Meier method. RFS was measured from the date of the first surgery to the date on which breast cancer recurrence or death was documented.

We evaluated log-rank modeling to assess the significance of associations between patients’ characteristics possibility of recurrence. Factors considered to have potential prognostic value were age, tumor size (≤2.0 cm vs. >2 cm), number of nodal metastases (≤4 vs. >4), tumor grade (1 vs. 2 or 3), chemotherapy, LHRH-a exposure duration (≤24 months vs. >24 months), PgR status (positive or negative), and HER2 status (positive or negative).

Results

Patients

Of the patient, 63.9% (124/194) and 36.1% (70/194) received LHRH-a for ≦2 and >2 years, respectively (Table 1). About half of the patients (98/194) received chemotherapy as adjuvant therapy. In this study, we routinely checked HER2 status of the tumor since 2002. For the patients treated prior to 2002, we checked the HER2 status of surgical specimens only when they experienced relapse. For this reason, we included patients who had not received anti-HER2 therapies as adjuvant therapy in HER2-positive patients (23/194). Furthermore, HER2-negative group included ‘HER2 status- unknown’ cases.

Serum E2 and FSH

Serum E2 and FSH levels of selected patients at each treatment are shown in Fig. 1. To determine menopausal status, we evaluated the serum E2 and FSH levels before starting treatment. We also measured E2 levels during chemotherapy, LHRH-a + TAM therapy, and TAM monotherapy after LHRH-a treatment ended. Serum data from 1 year after the initiation of endocrine therapy was defined as “during endocrine therapy” and data from 1 month after the initiation of chemotherapy was defined as “during chemotherapy”. Baseline serum was sampled during the follicular, ovulatory, or luteal phase, depending on the cycle of each patient. All the data were within the normal range of serum E2 levels of premenopausal women (Fig. 1a). The range of serum E2 levels among patients treated with TAM only was 10.0–5248.0 pg/mL (mean, 41.0 pg/mL). This E2 level was higher than that of the same patients during chemotherapy (mean, 24.0 pg/mL; range, 10.0–56.0 pg/mL), higher than that of patients treated with TAM plus LHRH-a (mean, 26.0 pg/mL; range, 10.0–52.0 pg/mL), and higher than the normal level of E2 in general postmenopausal women (<31.0 pg/mL). In most cases, E2 levels decreased to a level equivalent to those of postmenopausal women during LHRH-a + TAM therapy or chemotherapy and increased during TAM monotherapy. The greatest increase in FSH was among women undergoing chemotherapy (Fig. 1b). Figure 1(c) shows the image of hormonal dynamics during each treatments.

Log-rank analysis

We performed Log-rank analysis to explore the relationship between prognosis and several clinicopathological factors including LHRH-a exposure time. Table 1 shows clinical factors and hazard ratios for recurrence. Forty-six patients experienced breast cancer recurrence after surgery. We intended to show the correlations on the relationship between RFS and tumor size, lymph node metastatic status, chemotherapy, LHRH-a treatment duration, and HER2 status; however, statistically significant relationship was unclear.

Table 2 shows the relationship between patients’ characteristics and LHRH-a treatment duration. No significant difference was observed in the distribution of clinico-pathological factors or the duration of LHRH-a treatment.

Kaplan Meier recurrence-free survival

Patients were divided into 2 groups according to the length of LHRH-a treatment (≦24 months, >24 months) (Fig. 2). The relapse rate tended to be less among patients with longer LHRH-a treatment (Fig. 2a). To avoid the confounding factor of chemotherapy, we analysed the prognosis in the patients who received endocrine therapy alone (Fig. 2b). Patients’ characteristics without chemotherapy is shown in Table 3. Among patients without adjuvant chemotherapy, the prognosis was significantly worse in patients who had shorter LHRH-a treatment [hazard ratio, 6.62; 95% confidence interval (CI), 1.08–13.37; P = 0.0042]. For the endocrine plus chemotherapy group, we observed no difference in RFS (hazard ratio, 1.24; 95% CI 0.55–2.83; P = 0.5996, data not shown). In almost all cases, serum E2 levels increased during TAM monotherapy after completion of LHRH-a + TAM therapy (data not shown). There was not statistically significant differences between the length of LHRH-a exposure and serum E2 levels during TAM monotherapy. Patients with higher E2 levels (>41 pg/mL) during TAM monotherapy had a lower recurrence rate (Fig. 3).

Discussion

Several studies have shown that OS and/or DFS are improved by inducing amenorrhea in premenopausal breast cancer patients [7–10]. In our study, those who sustained a longer menopausal status because of a longer LHRH-a treatment had a better prognosis among patients who received endocrine therapy without chemotherapy as adjuvant therapy.

Younger breast cancer patients often resume menstruation after completion of adjuvant therapy. We hypothesized that E2 levels during adjuvant therapy might be associated with recurrence, and therefore, we examined whether serum E2 level was associated with prognosis. In most cases, serum E2 decreased to the level found in postmenopausal women during LHRH-a + TAM therapy or chemotherapy and increased during TAM monotherapy. FSH increased to the highest levels in patients undergoing chemotherapy. This result is likely due to the LHRH-a-induced decrease in E2 via negative feedback of FSH secretion and the direct effects of chemotherapy on the ovum. In the absence of LHRH-a, TAM directly stimulates E2 secretion from the ovum but decreases FSH by means of a negative feedback mechanism.

Higher preoperative serum E2 level was reported to be a negative prognostic factor in postmenopausal breast cancer patients without E2 depletion [11]. In contrast, estrogen exposure after estrogen depletion was observed to cause apoptosis of ER-positive breast cancer cells in several studies [12–16]. We believe that this phenomenon could occur during adjuvant therapy even in early-stage cancer patients, as our patients demonstrated high serum E2 level after completion of LHRH-a. This E2 elevation might suppress recurrence by inducing apoptosis via ER signaling in residual breast cancer cells. Our experience shows that some cases of ER-positive breast cancer are affected by fluctuations in serum E2 levels.

This study is limited by the small number of serum samples taken from each point. More samples from each patient and a longer follow-up are required to prove our hypothesis.

Among patients who received endocrine therapy without chemotherapy as adjuvant therapy, those who maintained a longer menopausal status because of a longer LHRH-a treatment had a better prognosis. These data suggest an association between breast cancer recurrence and serum E2 levels in premenopausal ER-positive, early-stage breast cancer patients.

Conflict of Interest

Takuwa Haruko and other co-authors have no conflicts of interest to declare.