The study population consisted of fifty-nine premenopausal breast cancer patients whose stage of breast cancer was restricted to stage 0 (e.g. ductal carcinoma in situ). They were referred to the Department of Obstetrics and Gynecology at Seoul National University Hospital and all of the participants underwent gynecologic surveillance during tamoxifen treatment (20 mg/day). The menstrual history and gynecologic information, including the number of pregnancy and live birth, was reviewed from the medical records. Amenorrhea was calculated in months and it was defined as the absence of menstruation for a time equal to a total of three previous cycles since the previous menstrual period. Those who had a previous history of chemotherapy or serum follicle stimulating hormone (FSH) concentrations equal to or higher than 30 mIU/mL were excluded since they were at menopausal transition. Those who received aromatase inhibitor, herceptin or gonadotropin releasing hormone (GnRH) agonist treatment were also excluded. All clinical information was collected and retrospectively reviewed. The study protocol was approved by the Institutional Review Board of Seoul National University Hospital.

Serum hormones were measured between cycle days 2 and 5 in patients with menstruation or at any time in amenorrheic participants. Serum luteinizing hormone (LH) and FSH were determined with immunoradiometric assay kits (Siemens Healthcare Diagnostics Inc., Deerfield, IL, USA). Serum estradiol was measured using a radioimmunoassay kit (Siemens Healthcare Diagnostics Inc.), and minimal detection limits were 8 pg/mL. Immunoradiometric assay kits (DiaSorin, Saluggia, Italy; ShinJin Medics Inc., Goyang, Korea) were used to determine the levels of thyroid stimulating hormone (TSH) and prolactin.

Serum carcinoembryonic antigen (CEA) and cancer antigen 15-3 (CA 15-3) were measured using non-competitive immunoradiometric assay kits (ShinJin Medics Inc.) and solid-phase sandwich immunoradiometric assay kits (Cisbio Bioassays, Coldolet, France) with minimal detection limits of 0.18 ng/mL and 0.2 U/mL, respectively.

During the gynecologic examination, transvaginal ultrasound was performed to detect uterine and ovarian abnormalities using an ultrasound system with a transvaginal probe of 5.0 MHz (Medison, Seoul, Korea). The endometrial thickness (EM thickness) was measured as the width of the anterior and posterior layer of the endometrium in the sagittal plane of the uterus.

Immunohistochemical (IHC) staining was conducted in regard to estrogen receptor (ER) (1:100, 1D5; Dako, Glostrup, Denmark), progesterone receptor (PR) (1:200, PgR636; Dako), B-cell lymphoma 2 (Bcl-2) (1:200, 124; Dako), Ki-67 (1:300, MIB-1; Dako), epidermal growth factor receptor (EGFR) (1:100; Zymed, San Francisco, CA, USA), and human epidermal growth factor receptor 2 (HER2) (1:200, CB11; Leica Microsystems, Berlin, Germany) on the cut sections of specimens. Positive staining was defined as nuclear staining in ≥1% of the tumor cells for ER and PR. For Bcl-2, cytoplasmic staining in ≥10% of the tumor cells was defined as positive. The Ki-67 proliferation index of each case was evaluated from the percentage of Ki-67-positive cells among at least 200 tumor cells. Only the membranous staining of tumor cells was scored for the determination of HER2 expression, and the intensity was scored as follows: 0 for negative staining, 1+ for weak staining, 2+ for moderate staining, and 3+ for strong staining in at least 30% of the tumor cells.

All data were expressed as mean±standard deviation or number of participants. All statistical analyses were performed using the Statistical Package for Social Science 21.0 (SPSS Inc., Chicago, IL, USA). The data were analyzed with the Student t-test, χ² test and Fisher’s exact test. The p value of less than 0.05 was considered to be significant for all analyses.

A total of 59 premenopausal tamoxifen users were identified for analysis and the overall prevalence of amenorrhea was 61.0% (36/59) (Table 1). The duration of amenorrhea was 10.2 months and mean age of amenorrheic women was comparable to that of women with regular menstruation (42.3±5.3 vs. 41.6±4.0 years). No significant differences were observed in regard to the body habitus, obstetric history, previous menstrual regularity and mode of treatments between the two groups. Menstrual intervals were 26.8±2.6 vs. 27.8±3.0 days (regular menses vs. amenorrhea); the duration of tamoxifen use 24.5±4.7 vs. 23.7±4.4 months. The onset time of amenorrhea from the initiation of tamoxifen was 22.4±3.6 months. The study groups were followed for 39–55 months after breast cancer surgery and tamoxifen was started from within one month after surgery during this follow-up period.

IHC marker positivity of surgical specimen was compared between the two groups. The compared variables included ER, PR, Bcl-2, Ki-67, EGFR and HER2. None of the compared markers showed any significant difference according to the presence of menstruation (Table 2). No correlation was observed between IHC results and hormone profiles, however, amenorrheic group showed a higher prevalence of HER2 0 and HER2 1+ when compared to regular menses group (p=0.006).

Amenorrheic women had a significantly higher level of serum FSH than that of women with regular menses (8.1±5.7 vs. 5.1±2.2 mIU/mL, p=0.01), which may suggest decreased ovarian reserve (Table 3). Regularly menstruating women demonstrated significantly increased EM thickness compared to that of amenorrheic group (7.4±4.1 vs. 5.4±2.3 mm, p=0.04), suggesting increased serum estrogen concentrations, however the difference of estrogen levels did not reach a statistical significance. Serum TSH level was significantly lower in amenorrheic group (p=0.04), although the mean values of both groups were within a normal range (Table 3). The distribution of serum TSH and prolactin level were not different according to the presence of menstruation (Table 4).