Introduction The clinical impact on fertility in carriers of BRCA1 and BRCA2 mutations remains unclear. The aim of this study was to assess ovarian reserve as measured by anti-mullerian hormone levels in BRCA1 or BRCA2 mutation carriers, as well as to investigate the impact of anti-mullerian hormone levels on reproductive outcomes.
Methods The study involved a cohort of women who tested positive for BRCA1 and BRCA2 screening or were tested for a BRCA1 or BRCA2 family mutation. Blood samples were collected for anti-mullerian hormone analysis and the reproductive outcomes were analyzed after a mean follow-up of 9 years. Participants were classified into BRCA mutation-positive versus BRCA mutation-negative. Controls were healthy relatives who tested negative for the family mutation. All patients were contacted by telephone to collect data on reproductive outcomes. Linear regression was used to predict anti-mullerian hormone levels by BRCA status adjusted for a polynomial form of age.
Results Results of anti-mullerian hormone analysis and reproductive outcomes were available for 135 women (BRCA mutation-negative, n=66; BRCA1 mutation-positive, n=32; BRCA2 mutation-positive, n=37). Anti-mullerian hormone curves according to BRCA status and adjusted by age showed that BRCA2 mutation-positive patients have lower levels of anti-mullerian hormone as compared with BRCA-negative and BRCA1 mutation-positive. Among the women who tried to conceive, infertility was observed in 18.7% of BRCA mutation-negative women, in 22.2% of BRCA1 mutation-positive women, and in 30.8% of BRCA2 mutation-positive women (p=0.499). In the multivariable analysis, there were no factors independently associated with infertility.
Discussion BRCA2 mutation-positive carriers showed more diminished anti-mullerian hormone levels than BRCA1 mutation-positive and BRCA mutation-negative women. However, these differences do not appear to have a negative impact on reproductive outcome. This is important to consider at the time of reproductive counseling in women with BRCA1 or BRCA2 mutations.
- anti-Müllerian hormone
- ovarian reserve
Statistics from Altmetric.com
BRCA2 mutation carriers had decreased anti-müllerian hormone levels.
Decreased anti-müllerian hormone levels do not appear to have an impact on reproductive outcome.
This finding is important at the time of reproductive counseling in women with BRCA mutations.
Factors that influence the reproductive prognosis of women are increasingly being more extensively investigated. It is currently known that age,1 endometriosis,2 thrombophilia,3 and unhealthy lifestyle behaviors such as cigarette smoking,4 among other factors, can adversely impact fertility in women. However, the clinical impact on fertility in carriers of BRCA1 and BRCA2 mutations remains unclear. Low response to ovarian stimulation during fertility treatments in carriers of germline mutations in BRCA genes has been shown,5 which may reflect a diminished oocyte reserve in these women. BRCA is known to be involved in homologous DNA recombination and plays an essential role in double-strand DNA break repair.6 Increasing evidence shows that BRCA mutations can result in defective DNA repair, leading to oocyte compromise with apoptotic cell death or the clinical consequence of breast and ovarian carcinogenesis.6
On the other hand, estimates of BRCA1 and BRCA2 mutations in general populations range from 0.2–1%.7 8 Latin American ancestries had a significantly higher prevalence of deleterious BRCA1 and BRCA2 mutations compared with women of Western European ancestry.9 Therefore, assessing the impact of BRCA mutations on fertility is of vital importance for a significant part of the population and is integral to reproductive counseling in all high-risk women.
Over the past 5 years, numerous studies on fertility and ovarian reserve in women with BRCA mutations have been published with inconsistent results; some have reported reduced circulating anti-müllerian hormone levels in women with either BRCA1 10 or BRCA2 mutations,11 while others found similar anti-mullerian hormone levels in carriers and non-carriers.12 Given that controversy still exists about the impact of BRCA1 and BRCA2 mutations on female fertility, the aim of this study was to assess serum anti-mullerian hormone levels in BRCA1 or BRCA2 mutation carriers, and to compare them with healthy non-carriers. A second objective of the study was to determine the clinical impact of serum anti-mullerian hormone levels on long-term reproductive outcomes in BRCA mutation carriers. It was hypothesized that decreased anti-mullerian hormone levels were associated with worse reproductive outcomes in women with either BRCA1 or BRCA2 mutations.
The study was carried out in collaboration with Hospital Universitari de Bellvitge, Institut Català d’Oncologia (ICO), and Fundació Institut d’Investigació Biomèdica de Bellvitge (IDIBELL) in Barcelona, Spain. The study protocol for the analysis of circulating anti-mullerian hormone levels was approved by the Ethical Committee for Clinical Research of Hospital Universitari de Bellvitge (ID PR243/12). Written informed consent was obtained from all women undergoing testing of BRCA mutation status.
The inclusion criteria were women aged between 18 and 45 years who were proven carriers of BRCA1 or BRCA2 gene mutations. Controls were healthy relatives that tested negative for the BRCA1 or BRCA2 family mutation. Also, viable frozen serum samples had to be available for anti-mullerian hormone analysis, and subjects had to respond to a telephone survey to obtain data on reproductive outcomes. Exclusion criteria were unilateral or bilateral adnexectomy, history of pelvic radiation therapy, cancer, chemotherapy, and refusal to participate in the study.
The initial study sample included all attending patients in whom BRCA mutation status was analyzed. All inclusion criteria and no exclusion criteria were applied, and thereafter patients were classified as BRCA mutation-positive or BRCA mutation-negative. Anti-mullerian hormone measurements were performed in frozen stored samples collected between January 2005 and November 2011, using a fully automated anti-mullerian hormone electrochemiluminescence immunoassay (ECLIA) on the Roche Modular E-170 platform. According to anti-mullerian hormone levels, the ovarian response was classified as “low” (≤0.58 ng/mL), “normal” (>0.58 and ≤1.48 ng/mL), and “high” (>1.48 ng/mL) as reported by Jayaprakasan et al.13 Between March and June 2018, BRCA mutation-positive and BRCA mutation-negative patients were contacted by telephone to collect data on reproductive outcomes and epidemiological variables. Verbal informed consent for the telephone survey was obtained.
Information obtained from each participant included age at the time of blood extraction, age at the current telephone survey, age of menarche, menstrual cycle regularity (yes/no), use and type of contraceptive methods, parity, history of spontaneous miscarriages, maternal age at first birth, and maternal age at last birth. Women were also asked if they had a history of infertility (“Did you need more than a year of frequent relations to become pregnant?”), use of fertility treatment (“Have you ever received fertility treatment such as artificial insemination or in vitro fertilization-embryo transfer to help you to get pregnant?”), and the cause of fertility problems (“Did your gynecologist or infertility center explain to you the cause of your infertility, such as endometriosis, fibroids, polycystic ovary syndrome, male factor … or unknown?”).
Categorical data are expressed as frequencies and percentages, and quantitative data as mean±SD or median and range. Demographic variables and the clinical characteristics of participants were described for the subsets of BRCA mutation-negative women, BRCA1 mutation-positive women, and BRCA2 mutation-positive women. Categorical variables were compared with the χ2 test, the Fisher’s exact test or the Mann-Whitney U test, and continuous variables with one-way analysis of variance (ANOVA) or the Kruskal-Wallis tests according to conditions of application. Linear regression was used to predict anti-mullarian hormone levels by BRCA status adjusted for a polynomial form of age, according to the following formula:
Antimullerian-Hormone= β + βage + βage² + βage³ + ε
Logistic regression analysis was used to assess the association of demographic and clinical variables with infertility. Odds ratio (OR) and 95% confidence interval (95% CI) were calculated. Statistical significance was set at p<0.05 (two-sided). Analyses were performed using R (version 3.5.0) R: A Language and Environment for Statistical Computing (Vienna, Austria).
The study's initial cohort included 760 women with known BRCA mutation status. A total of 586 (77.1%) patients were found ineligible for the following reasons: age <18 years or >45 years in 344 patients, cancer history in 97 patients, history of adnexectomy in one patient, and unavailable frozen samples for anti-mullerian hormone analysis in 144 patients. Therefore, measurement of serum anti-mullerian hormone levels was performed in 174 women, of whom 39 (22.4%) were ineligible for the results of the reproductive outcome survey because of unsuccessful telephone contact in 38 patients and refusal to participate in one patient. The flow chart of the study population is shown in Figure 1 of the online supplementary material.
Finally, results of anti-mullerian hormone analysis and reproductive outcomes after a mean follow-up of 9 years were available for 135 women. They were classified into three groups: BRCA mutation-negative women (n=66), BRCA1 mutation-positive women (n=32), and BRCA2 mutation-positive women (n=37). As shown in Table 1, there are no statistically significant differences in the distribution of demographic and clinical variables among women in the BRCA-negative, BRCA1-positive, and BRCA2-positive groups. Results of local weighted polynomial regression analysis of anti-mullerian hormone levels are shown in Figure 1. Age accounted for 22.4% of the variability of serum anti-mullerian hormone levels. Anti-mullerian hormone curves according to BRCA status and adjusted by age showed that BRCA2-positive mutation patients accounted for a lower variability of anti-mullerian hormone levels (20.2%) than BRCA-negative (28.4%) and BRCA1-positive (23.5%). As shown in Figure 2, reduced anti-mullerian hormone values in the BRCA2-positive group occurred at the expense of lower levels in the 50th percentile.
The gynecological and reproductive data of the women according to the study groups are show in Table 2. There were no statistically significant differences in the distribution of variables among BRCA-negative, BRCA1-positive, and BRCA2-positive women. The mean age at first birth was 30 years. The percentage of nulliparous women was 30.3% in the BRCA-negative group, 21.9% in the BRCA1-positive group, and 35.1% in the BRCA2-positive group, with “not tried to conceive” as the most frequent cause in 90%, 71.4%, and 84.6% of the cases, respectively.
Among the women who tried to conceive, infertility was recorded in 18.7% of BRCA-negative women, in 22.2% of BRCA1-positive women, and in 30.8% of BRCA2-positive women. Differences in the cause of infertility or in the percentage of women undergoing infertility treatment were not observed (Table 2). In the multivariable analysis to assess risk factors for infertility (excluding male factor), there were no factors independently associated with infertility (Figure 2, online supplementary material). BRCA1 showed an OR of 1.67 (95% CI 0.51 to 5.30; p=0.383) and BRCA2 showed an OR of 1.69 (95% CI 0.55 to 5.16; p=0.352). The comparison of clinical characteristics of women according to presence or absence of infertility is shown in Table 1 of the online supplementary material.
Our hypothesis that decreased anti-mullerian hormone levels would be associated with worse reproductive outcomes in women with either BRCA1 or BRCA2 mutations was not observed. Samples for anti-mullerian hormone measurement were collected at a mean age of participants of 32 years, but epidemiological data on fertility and reproductive outcomes were recorded at a mean age of 41 years. This mean interval of 9 years allowed assessing the role of anti-mullerian hormone levels as a predictive prognostic factor of reproductive outcome in the same cohort of women with known BRCA mutation status.
We found lower anti-mullerian hormone values in BRCA2-positive women as compared with the groups of BRCA-negative and BRCA1-positive women. This finding agrees with data reported by Johnson et al11 in a prospective cohort of 195 women. BRCA2 carriers (n=50) had anti-mullerian hormone levels that were 33% lower than low-risk controls (n=64) after adjusting for age, body mass index, and regular menses, and excluding patients with polycystic ovary syndrome. In this study, however, a relationship between anti-mullerian hormone levels and reproductive outcome was not observed.
In this study, after 9 years of follow-up, significant differences in age at first birth, percentage of women with infertility, and the percentage of women undergoing infertility treatment among the groups of BRCA-negative and BCRA1 and BCRA2 carriers were not observed. This comparable reproductive prognosis in the three study groups is consistent with data from large epidemiological studies. In a matched case-control study of 2254 BRCA carriers and 764 non-carrier controls with a mean age of 42 years, differences in nulliparity, parity, age at first birth, age at last birth, fertility problems, and fertility treatments were not found.14 However, in our study, it was possible to determine that 90% of non-carriers, 71.4% of BRCA1 carriers, and 84.6% of BRCA2 carriers were nulliparous because they never tried to conceive, allowing the exclusion of women in whom fertility remained unconfirmed and reducing the possibility of selection bias. In the overall group of female infertility, the cause of infertility was unknown in five (10.4%), zero, and four (15.4%) of BRCA-negative, BRCA1-positive, and BRCA2-positive women, respectively. Also, seven (14.6%) versus three (11.1%) versus five (19.2%) women, respectively, required fertility treatment. Similar epidemiological results were reported in a survey about reproductive history in 908 matched pairs of BRCA carriers and controls, in which carrying a BRCA1 or BRCA2 mutation does not appear to affect fertility.15
In contrast to the present results, some studies have shown that anti-mullerian hormone is not reduced in BRCA carriers versus non-carriers. Van Tilborg et al16 compared anti-mullerian hormone levels between 124 BCRA1/2 mutation carriers and 131 non-carriers, and the adjusted linear regression analysis revealed no reduction in anti-mullerian hormone level in the carriers (relative change 0.98, 95% CI 0.7 to 1.22; p=0.76). In the study of Michaelson-Cohen et al12 of 41 healthy BRCA1/2 mutation carriers, the anti-mullerian hormone levels were similar to those of non-carrier women matched for age. In a recent retrospective cohort study, BRCA carriers with and without malignancy exhibited comparable ovarian reserve and responses to ovarian stimulation compared with women with BRCA-negative cancers and cancer-free controls.17
Other studies, however, have reported differences in anti-mullerian hormone levels between BRCA mutation carriers and non-carriers. Giordano et al18 evaluated 124 women aged 18–45 years participating in the Northwestern Ovarian Cancer Early Detection and Prevention Program and found that BRCA1-positive women >35 years old had 10 times the odds of a low anti-mullerian hormone (<0.5 ng/mL) compared with women 35 years. With adjustment for body mass index, duration of birth control, smoking, gravity, parity, and age >35, BRCA1 was still strongly associated with a potential decrease in ovarian reserve. In a cross-sectional study of anti-mullerian hormone concentration in 693 women, anti-mullerian hormone was measured at the time of enrolment in a familial breast cancer cohort study.10 Mutation carriers had on average 25% (95% CI 5% to 41%; p=0.02) lower anti-mullerian hormone concentrations than non-carriers, but there was no evidence of an association between anti-mullerian hormone concentration and BRCA2 mutation status. Wang et al19 carried out a cross-sectional study in 143 women, aged 18–45, who underwent clinical gene testing for BRCA mutations due to family history of cancer. They found that BRCA1 carriers had fourfold higher odds of having anti-mullerian hormone levels <1 ng/mL compared with controls, whereas there was no difference in anti-mullerian hormone levels between BRCA2 carriers and controls. In human oocytes from reproductive-age women, Titus et al20 reported a significant age-related decline in BRCA1 expression, although this was not noted for BRCA2 expression, suggesting that the decline may occur in later years. This may be related to a difference in age penetrance between the two BRCA mutations, with BRCA1-associated ovarian cancers occurring over a decade earlier than BRCA2-associated ovarian cancers.19
Our study has some limitations. First, it is important to note that our findings are based on a subset of women with regular and irregular menstrual cycles. Irregular menses have been suggested as an exclusion criterion because of the possibility of masking higher anti-mullerian hormone values secondary to polycystic ovarian syndrome. However, this possibility seems unlikely given the similar distribution of women with regular menses in the three study groups and the low prevalence of patients diagnosed with polycystic ovarian syndrome. Moreover, excluding women with irregular cycles may lead to the exclusion of those with early menopause. Second, the exclusion of patients with cancer may be a further limitation. It has been hypothesized that in women with cancer, anti-mullerian hormone values could be more diminished compared with “healthy carriers” due to a greater deleterious effect caused by the mutation in the BRCA genes.6 To assess this effect, it would be necessary to determine anti-mullerian hormone levels of those patients who subsequently presented with cancer. However, BRCA status has been tested in each patient and it is the only main difference between cases and controls. Therefore, belonging to the same families confers on patients and controls a very similar genetic background that mostly overcomes the possible bias. Also, a control group of “healthy women who are not relatives of patients with mutated BR” was not available. It has been suggested that “non-carrier relatives” have a higher cancer risk as compared with the healthy population.21 This would imply that “non-carrier relatives” may have similar genetic factors that influence the ovarian reserve, which would reduce the difference in anti-mullerian hormone values when compared with their “carrier relatives”. Further studies with such control groups in a larger sample size may help to answer these questions.
To conclude, the present study shows that serum anti-mullerian hormone levels in carriers of BRCA2 mutation were lower as compared with carriers of BRCA1 mutation and BRCA-negative women. However, these differences in anti-mullerian hormone levels did not appear to have a negative impact on reproductive outcome as the percentages of infertility and use of infertility treatment were unrelated to BRCA mutation status.
JP and SF-G are joint first authors.
Presented at This study was presented as a poster at the 21st European Congress on Gynaecological Oncology, Athens, 2019.
Furthermore, it was presented as an oral communication at the 34th National Congress of the Gynaecology Oncology section of the Spanish Society of Gynecology and Obstetrics, Barcelona, 2018.
Contributors JP: development, supervision, edition. SF-G: data collection, development, edition, supervision. ICL, e-mail: data collection, development. MTC, e-mail: data collection, development. JPM, e-mail: data collection, statistics analysis. LFE: data collection, patients selection. ATV: data collection, supervision. CL: data collection, supervision. JMBV: edition, supervision. BC-E: AMH analysis, data collection. MDR: edition, supervision.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.