Abstract
Aim: We investigated whether the ovarian reserve determined on the basis of anti-Müllerian hormone (AMH) and inhibin B predicted disease-free survival (DFS) in premenopausal patients with hormone receptor-positive breast cancer treated with neoadjuvant chemotherapy. Patients and Methods: Our analysis included 32 premenopausal women with clinical stage III hormone receptor-positive invasive ductal breast cancer treated by neoadjuvant chemotherapy. Blood samples were obtained after neoadjuvant chemotherapy completion. The median follow-up period was 57.7 months. Results: The median patient age was 41.5 years. The group with functional ovarian reserve was classified by higher AMH and higher inhibin B levels using cut-off values of 1,000 pg/ml and 30 pg/ml, respectively. The group with functional ovarian reserve had significantly worse DFS (p=0.043) than the group with ovarian failure. Conclusion: The functional ovarian reserve defined by higher serum AMH and inhibin B after neoadjuvant chemotherapy predicted poor DFS in premenopausal women with clinical stage III hormone receptor-positive breast cancer.
Menopausal status is considered a prognostic marker in patients with hormone receptor-positive breast cancer. Premenopausal patients tend to have a worse prognosis compared to postmenopausal patients, perhaps owing to differences in sex hormone levels. However, most of the data regarding the influence of ovarian function on breast cancer prognosis have been obtained indirectly (e.g. by assessing the survival benefits of ovarian ablation or ovarian suppression). For example, in premenopausal women with hormone receptor-positive breast cancer, oophorectomy resulted in breast cancer regression (1, 2). Ovarian suppression via administration of gonadotropin agonists or radiotherapy was as effective as chemotherapy (3-5). Despite the indirect evidence linking ovarian function to disease progression in sex hormone-sensitive breast cancer, studies evaluating the relationship between plasma sex hormone levels and therapeutic outcomes in premenopausal women with hormone receptor-positive breast cancer are few (6). This may reflect the lack of reliable and feasible methods for assessing ovarian function.
Examination of the functional ovarian reserve has been widely discussed in the context of infertility treatment. Classically, the ovarian reserve is defined as the number of non-growing follicles in the ovary at any specific time (7). As surgical oophorectomy alone can provide a direct count of the number of non-growing follicles within the ovary, biomarkers such as antral follicle count using ultrasonography, and levels of follicle-stimulating hormone (FSH), inhibin B, and anti-Müllerian hormone (AMH) are often used to estimate the ovarian reserve (7, 8). Although antral follicle count via ultrasonography is the most accurate method, it is costly, time-consuming, and subject to interobserver and intraobserver variation, all of which limit its use in routine practice in the clinic (9). On the other hand, serum FSH, inhibin B, and AMH levels can easily be measured in blood samples at any time.
Serum estradiol (E2) and FSH levels are widely used as indicators of menopausal status of patients with breast cancer in clinical trials (10-13), and of functional ovarian reserve during the course of breast cancer treatment in the clinic. The functional ovarian reserve estimated by these biomarkers enables the selection of appropriate endocrine therapy regimens such as those including tamoxifen, gonadotrophin-releasing hormone agonists, or aromatase inhibitors. However, the accuracy of E2 and FSH levels in estimating the functional ovarian reserve is questionable because of their wide intra-menstrual cycle fluctuations (14).
Recently, inhibin B and AMH levels were used to assess the functional ovarian reserve in infertile women. Inhibin B is a dimeric polypeptide secreted by the granulosa cells of the antral follicles under FSH stimulation (7, 8, 14). Inhibin B levels are thought to reflect the functional ovarian reserve (7). AMH is a dimeric glycoprotein secreted mainly by the glandulosa cells of the pre-antral and small antral follicles (7). Because it is produced during the pre-antral stage of the follicles, fluctuation of its levels during the menstrual cycle is minimal (7, 14-16). Basal AMH levels correlate with the number of retrieved oocytes after ovarian stimulation and better predict pregnancy through artificial reproduction than do FSH, inhibin B, or E2 levels (17). A recent report showed that AMH levels before chemotherapy predict whether ovarian function will resume after chemotherapy in young breast cancer survivors (8, 18). AMH and inhibin B might be better indicators of the functional ovarian reserve than FSH or E2. If inhibin B and AMH levels correlate with the functional ovarian reserve, even after chemotherapy, they might serve as prognostic markers in premenopausal women with hormone receptor-positive breast cancer.
We investigated the usefulness of functional ovarian reserve based on serum AMH, inhibin B, FSH and E2 levels, examined immediately after completion of neoadjuvant chemotherapy, as prognostic biomarkers in premenopausal women with clinical stage III hormone receptor-positive breast cancer.
Patients and Methods
By searching the database of the Breast Cancer Center at the Korea Cancer Center Hospital, we identified premenopausal women 20-50 years of age with clinical stage III primary hormone receptor-positive invasive ductal breast cancer treated with neoadjuvant chemotherapy and curative surgery between 2007 and 2010. This database has been previously described (19). Clinical data, surgical methods, treatment regimens, and tumor characteristics including estrogen receptor, progesterone receptor, and human epidermal growth factor receptor-2 (HER2) status were collected from the database. Hormone receptor positivity was defined as estrogen receptor or progesterone receptor expression in the primary tumor, as determined by immunohistochemistry. HER2 status was determined by immunohistochemistry or fluorescence in situ hybridization. If the patient had a menstrual period in the previous year or serum FSH level ≤30 mIU/ml at the time of diagnosis, she was considered premenopausal. Patients with a history of other primary malignancies were excluded.
Sixty-five patients from this database were eligible for this analysis. Of these patients, 32 had blood samples deposited in the Korea Institute of Radiological and Medical Sciences (KIRAMS) Biobank. All blood samples were obtained between the time of completion of neoadjuvant chemotherapy and the time of curative surgery and were kindly donated by each patient with written consent to KIRAMS Biobank. This study was approved by the Institutional Review Board of the Korea Cancer Center Hospital (IRB Number K-1401-002-014).
Levels of AMH, inhibin B, FSH, and E2 in serum were measured using enzyme-linked immunosorbent assay (ELISA). AMH assays were performed by USCN Life Science, Inc. (Buckingham, UK). Standards or samples were added to the wells along with an AMH-specific, biotin-conjugated polyclonal antibody and avidin conjugated to horseradish peroxidase (HRP). After incubation, a substrate solution was added to the wells, resulting in a colorimetric reaction. The optical intensity, which is proportional to the amount of AMH bound in the initial step, was measured at 405 nm (iMark microplate reader; BioRad, Hercules, CA, USA), and AMH concentrations were extrapolated from standard curves.
Inhibin B was quantified using a sandwich immune assay technique (ELISA kit; BlueGene Biotech Co., Ltd., Shanghai, P. R. China). Standards or samples were added to microtiter wells pre-coated with an inhibin B-specific monoclonal antibody. An HRP-conjugated polyclonal antibody specific for inhibin B was then added to the wells to “sandwich” the inhibin B bound to the immobilized monoclonal antibody. After incubation, the wells were thoroughly washed to remove all unbound components, and a substrate solution was added. The antibody–substrate reaction was terminated by adding a sulfuric acid solution, and optical intensity was measured spectrophotometrically at 450 nm. The inhibin B concentration in each sample was extrapolated from a standard curve plotting optical intensity as a function of the known inhibin B concentrations of the standards.
Levels of FSH and E2 were measured using electrochemiluminescence immunoassays (Roche Diagnostics, Mannheim, Germany) performed on an E170 module, according to the manufacturer's instruction. The FSH assay (Elecsys FSH; Roche Diagnostics) used two monoclonal antibodies recognizing different FSH epitopes. The limit of detection of the FSH assay was 0.100 mIU/ml. The E2 assay (Elecsys Estradiol II; Roche Diagnostics) was based on the competition of E2 and a labeled antigen for a binding site on an E2-specific biotinylated antibody. The limit of detection of the E2 assay was 5.00 pg/ml.
The events for disease-free survival were disease recurrence confirmed by biopsy or imaging or death by any cause. Disease-free survival was defined as the time between the date of disease diagnosis to the date of the first event or last follow-up. Kaplan–Meier analysis and the log-rank test were performed to assess the results of the survival analysis. The Cox regression model was used to identify significant independent factors related to disease-free survival. Spearman's rho test was used to assess the relationship between the levels of the biomarkers. Fisher's exact test was used to examine the relationships between biomarker levels and clinicopathological parameters. Significance was defined as p<0.05.
Results
The median age of the patients was 41.5 years (range=27-50 years). The clinical and pathological characteristics of the study population are shown in Table I. Thirty-one (96.9%) patients received a neoadjuvant anthracycline-docetaxel combination chemotherapy regimen. After surgery, disease in 11 (30.2%) patients was down-staged based on the results of a pathological examination. No patient had a pathological complete response. HER2 was overexpressed in the tumors of 14 (43.7%) patients; 10 of these patients received adjuvant trastuzumab chemotherapy after surgery.
Clinical and pathological characteristics of patients in this study.
The relationship between serum levels of anti-müllerian hormone, inhibin B, follicular-stimulating hormone and estradiol by Spearman's rho test in patients treated for hormone receptor-positive breast cancer.
The median levels of AMH, inhibin B, FSH, and E2 were 1150.40 pg/ml (range=752.74-2825.56 pg/ml), 18.50 pg/ml (range=0-140.69 pg/ml), 59.45 mIU/ml (range=4.21-122.30 mIU/ml), and 5.0 pg/ml (range=5.00-144.90 pg/ml), respectively. The levels of AMH, inhibin B, and E2 were significantly related to each other (p<0.05, Table II).
The cut-off values for AMH and inhibin B indicative of a functional ovarian reserve were determined at the level of the lowest p-value in the survival analysis. We used the reference value of our Institute as the premenopausal cut-off value for FSH and E2 (Table III). Based on the cut-off value for AMH, inhibin B and FSH, 68.8%, 34.4% and 15.6% of patients retained ovarian function after neoadjuvant chemotherapy, respectively. However, only one patient retained ovarian function based on the cut-off value for E2.
Cut-off value for functional ovarian reserve after neoadjuvant chemotherapy for hormone receptor-positive breast cancer.
The median follow-up period was 57.7 months (range=6-79 months). During the follow-up period, four patients had locoregional recurrence, and four had distant metastasis as the first event. We performed a survival analysis to evaluate the prognostic value of the biomarkers. Patients with a functional ovarian reserve identified according to the cut-off values for AMH or inhibin B had a tendency toward better disease-free survival compared with those identified as ovarian failure, although the difference was not significant (Figure 1). We, therefore, assessed the survival benefit of the functional ovarian reserve based on both AMH and inhibin B levels. Because E2 levels identified only one patient with a functional ovarian reserve, E2 was excluded from this analysis. Kaplan–Meier analysis showed that patients with a functional ovarian reserve based on lower AMH and lower inhibin B levels had significantly worse disease-free survival rates compared to patients with ovarian failure based on higher AMH or inhibin B level (p=0.043, Figure 2; Table IV). The hazard ratio for poorer survival for patients having a functional ovarian reserve versus ovarian failure based on both levels was 2.156 but was not significant (Table V).
Disease-free survival of patients with hormone receptor-positive breast cancer based on the levels of biomarkers for functional ovarian reserve: A: anti-Müllerian hormone, B: inhibin B, C: follicle-stimulating hormone and D: estradiol.
Disease-free survival of patients with hormone receptor-positive breast cancer according to the functional ovarian reserve determined by using both anti-Müllerian hormone and inhibin B levels.
Univariate analysis of prognostic factors for disease-free survival of patients with hormone receptor-positive breast cancer.
Multivariate of prognostic factors for disease-free survival of patients with hormone receptor-positive breast cancer.
Discussion
In this study, the functional ovarian reserve determined by higher AMH and inhibin B levels after neoadjuvant chemotherapy was related with the poor disease-free survival of premenopausal women with clinical stage III hormone receptor-positive breast cancer. To our knowledge, this is the first report to show that the functional ovarian reserve based on both AMH and inhibin B levels is a prognostic factor in patients with breast cancer.
Only a few studies have investigated the direct effects of ovarian function on the survival of premenopausal patients with breast cancer (1, 6). This may reflect the lack of appropriate biomarkers, and most studies used age, FSH or E2 levels. Although age is an important factor for determining the functional ovarian reserve, it can underestimate or overestimate the number of nongrowing follicles by 100-fold if used alone (7, 20). The fluctuation of FSH and E2 levels during the menstrual cycle compromises the accuracy of their use in premenopausal women. Compared to FSH and E2 levels, inhibin B and AMH levels are more reliable indicators of ovarian function (14).
In the chemotherapy era, the functional ovarian reserve has been studied in young women who became infertile because of cancer treatment. Gynecologists have used age, antral follicular count by ultrasonography, and FSH, E2, inhibin B, and AMH levels to predict reproductive potential in cancer survivors. AMH levels are thought to be useful in this regard, whereas the usefulness of inhibin B levels is less clear (8). As AMH is produced during the pre-antral stage of the follicles, its serum levels directly correlate with the functional ovarian reserve (7). In a study of 59 women with early breast cancer, Anderson et al. found that low pre-chemotherapy levels of AMH significantly correlated with amenorrhea 2 years after chemotherapy (p<0.0001), whereas pre-chemotherapy levels of FSH and inhibin B were unrelated to amenorrhea at this time point (21). Therefore, AMH is a useful predictor of long-term post-chemotherapy loss of ovarian function. Although we did not analyze AMH, inhibin B, FSH and E2 levels in relation to menstruation in patients after chemotherapy, their serum levels after neoadjuvant chemotherapy were significantly related to each other. This finding suggests that the serum levels of AMH and inhibin B might reflect the functional ovarian reserve, even after neoadjuvant chemotherapy.
Although AMH was undetectable or barely detectable after two or more menstrual cycles in the women in the study by Anderson et al. (21), the serum levels of AMH, as well as those of inhibin B and E2, after neoadjuvant chemotherapy were measurable in our study. This difference may reflect the inclusion of cyclophosphamide in chemotherapy regimens of all patients in the study by Anderson et al. (21) versus its inclusion in the regimen of only one patient in our study. It is well known that cyclophosphamide is more gonadotoxic than anthracycline or taxane (22). Lower gonadotoxicity may therefore account for the higher functional ovarian reserve in the patients in our study compared with those in the study by Anderson et al. (21).
Continuous assessment of ovarian function is critical throughout the treatment course in patients with hormone receptor-positive breast cancer. Established guidelines recommend administering aromatase inhibitors to menopausal women owing to their better survival benefits compared with those of tamoxifen, the drug of choice in premenopausal women (23-25). The recent SOFT-TEXT trial suggests using aromatase inhibitors to suppress ovarian function in premenopausal patients (12). However, estimating menopausal status after chemotherapy is difficult for physicians because chemotherapy-induced suppression of ovarian function can cause amenorrhea in premenopausal women. Chemotherapy-induced amenorrhea was reversed within 2 years after chemotherapy in 18-50% of patients in previous studies (22, 26), therefore it is not a useful marker of the functional ovarian reserve. FSH or E2 levels are also not useful because they are unrelated to the future resumption of menstruation (27, 28). In our study, the combination of AMH and inhibin B levels after neoadjuvant chemotherapy predicted the disease-free survival of premenopausal patients with hormone receptor-positive breast cancer. If this combination is a prognostic marker in premenopausal women with hormone receptor-positive breast cancer, it may help identify the best endocrine therapy strategies for patients. We hope that our findings will facilitate the design of personalized therapies.
In conclusion, the functional ovarian reserve defined by higher serum AMH and inhibin B levels after neoadjuvant chemotherapy predicted a poor survival outcome in premenopausal women with clinical stage III hormone receptor-positive breast cancer. This result may facilitate the design of tailored endocrine therapies for premenopausal patients. Additional prospective cohort studies are needed to validate this result.
Acknowledgements
This study was supported by a grant of the Korea Institute of Radiological and Medical Sciences (KIRAMS), funded by Ministry of Science, ICT and Future Planning, Republic of Korea (1711021931). The blood samples used in this study were obtained from the Korea Institute of Radiological and Medical Sciences Radiation Biobank (1711021877).
Footnotes
Conflicts of Interest
The Authors declare that they have no conflict of interest.
- Received December 17, 2015.
- Revision received January 21, 2016.
- Accepted January 22, 2016.
- Copyright© 2016 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
