Reduced reproductive potential in young healthy women with hereditary breast and/or ovarian cancer syndrome

Abstract

Background

Young, healthy women carrying a pathogenic or likely pathogenic variant (P/LPV) in genes configuring the hereditary breast and/or ovarian cancer (HBOC) syndrome face several non-oncological issues. Among these, the implications on fertility are not yet entirely understood.

Methods

Aiming to explore the ovarian reserve in young, healthy women with HBOC syndrome, we conducted a monocentric, prospective, observational cohort trial between January 2020 and September 2023. Eighty-seven healthy women aged less than 42 years with a P/LPV in HBOC predisposition genes were enrolled: 32 BRCA1 P/LPV carriers, 47 BRCA2 P/LPV carriers, and 8 carriers of P/LPV in other genes (TP53, RAD50, CHECK2, RAD51D, PALB2, ATM). AMH levels and antral follicular count (AFC) were evaluated as fertility biomarkers.

Results

No significant differences in demographic characteristics or mean levels of AMH or in AFC are observed between BRCA1 and BRCA2 P/LPV carriers. The distribution of AMH values is significantly lower compared to the general population (p = 0.019). The significant decrease in AMH levels is mostly ascribable to BRCA1 P/LPV carriers (p = 0.03). Both in the overall population and in BRCA1/2 P/LPV carriers, AFC decreases faster compared to those reported in the nomogram.

Conclusions

A consistent trend for reduced reproductive potential in young, healthy women with HBOC syndrome is observed, particularly in BRCA1 P/LPV carriers. These findings need to be confirmed by larger studies including also women carrying P/LPV in other HBOC syndrome-related genes.

Plain language summary

Some young, healthy women have an increased chance of developing breast and/or ovarian cancer because they have inherited mutated genes from their parents. This is known as hereditary breast and/or ovarian cancer (HBOC) syndrome. We investigated whether these women also had issues with fertility. We measured two indicators of ovarian reserve. The ovarian reserve is the number of healthy eggs that a woman has in her ovaries. We found a trend for reduced reproductive potential in young, healthy women with HBOC syndrome. If these findings are confirmed by larger studies, fertility counseling could be helpful for this specific population of women.

Sighinolfi et al. evaluate the ovarian reserve in young healthy women with hereditary breast and/or ovarian cancer (HBOC) syndrome. A consistent trend for reduced reproductive potential in young, healthy women with HBOC syndrome is observed, particularly in BRCA1 pathogenic/likely-pathogenic variant carriers.

Introduction

Harboring a germline pathogenic or likely pathogenic variant (P/LPV) in BRCA1/2 is associated with an increased lifetime risk of breast and ovarian cancer. More specifically, the cumulative risk of breast and ovarian cancer by 80 years of age is 72% and 44% for BRCA1 P/LPV carriers and 69% and 17% for BRCA2 P/LPV carriers, respectively. In BRCA1/2 P/LPV carriers, breast cancer often occurs during reproductive age, while ovarian cancer is rare before 40–45 years of age¹. Besides these well-defined high penetrance genes, other genes can increase the susceptibility to breast and/or ovarian cancers. In detail, the most prevalent high and moderate penetrant genes are ATM, BARD1, BRIP1, CHEK2, CDH1, NF1, PALB2, PTEN, RAD51C, RAD51D, STK11, and TP53. Low penetrant genes are FANCC, MRE11A, NBN, RECQL4, RAD50, RINT1, SLX4, SMARCA4, and XRCC2². Germline P/LPVs in one of these genes configure hereditary breast and/or ovarian cancer (HBOC) syndrome. The identification of HBOC syndrome plays a crucial role in the management of hereditary cancer prevention, diagnosis and treatment^3–20.

Importantly, besides concerns for increased cancer risk, young women carrying a germline P/LPV in HBOC predisposition genes face several non-oncological issues that affect the quality of their reproductive life. First of all, the premature menopause induced by risk-reducing salpingo-oophorectomy may cause hot flashes, cognitive changes, vaginal dryness, sleep disturbances, and increased risk for osteoporosis and cardiovascular disorders^21–23. Moreover, the risk-reducing bilateral mastectomy in BRCA-carriers and the breast surgery in BRCA-cancer patients may affect body image perception and sexual functioning^24,25. Finally, previous researches indicated that female carriers of a BRCA P/LPV may display lower ovarian reserve, tendency to experience premature menopause, increased ovarian aging, decreased anti mullerian hormone (AMH) level, and lower ovarian response to ovarian stimulation than non-carriers^26–32. Nevertheless, the implications of germline P/LPVs on women’s fertility are not yet entirely understood.

The anti-mullerian hormone (AMH) is a dimeric glycoprotein solely produced from granulosa cells of preantral (primary and secondary) and small antral follicles³³. The number of small antral follicles directly correlates with the primordial follicle pool. As the amount of antral follicles diminishes with age, AMH serum levels decrease to the point of becoming almost undetectable at menopause^34,35. Moreover, AMH levels reflect the ovarian reserve, thus predicting menopause age^36,37. On the other hand, the quantity of primordial follicles corresponds with the number of antral follicles at every age. With ageing, the primordial follicle pool parallelly decreases with the amount of antral follicles sensitive to FSH³⁸. The correlation between antral follicle count (AFC) and the onset of menopausal transition has been investigated in two studies, thus demonstrating the close relationship with the quantitative aspect of ovarian reserve^39–41.

The aim of this study is to explore ovarian reserve in young, healthy women carrying germline P/LPVs in HBOC predisposition genes, evaluating AMH and AFC as fertility biomarkers. We observe a consistent trend for reduced reproductive potential in young, healthy women with HBOC syndrome, particularly in BRCA1 pathogenic/likely-pathogenic variant carriers.

Methods

Study participants

This is a monocentric, prospective, observational, cohort trial conducted at the Modena Family Cancer Clinic (MFCC), Italy, between January 2020 and September 2023. In the Emilia-Romagna region (Italy), women with a family history of breast and/or ovarian cancer are currently invited to a first-level evaluation at one of the 13 Spoke Centers. On the basis of their lifetime breast cancer risk assessed by risk prediction models such as the Gail and Tyrer-Cuzick models⁴², they are offered a chance to join a personalized surveillance program. After the first evaluation, some of them move on to a second-level evaluation at a HUB Center to assess the need for a genetic testing in the family. The MFCC is one of the four HUB Centers in Emilia-Romagna for the identification of families at increased familial or hereditary cancer risk. To date, more than 900 women who carry a genetic P/LPV predisposing to breast and/or ovarian cancer have attended the MFCC for their screening programs.

We enrolled healthy women carrying a germline P/LPV in one of the breast and/or ovarian cancer predisposition genes, aged less than 42 years. Women with previous ovarian surgery, history of cystectomy or detection of current ovarian cyst at the time of ultrasonography, personal history of cancer, current or past use of hormonal contraceptive methods within the last 6 months, and pregnancy or breastfeeding were excluded from the study.

Procedure

At study entry, participants filled out an epidemiologic questionnaire assessing body mass index (BMI), menstrual history, parity, use of hormonal contraceptive methods, fertility treatment, and smoking status. After informed consent was provided, blood samples were obtained to determine AMH level, and transvaginal ultrasound (TV-US) was performed by an experienced operator (G.S.) to evaluate AFC. Blood samples were collected and TV-US was performed on the day of patient recruitment, regardless of the last menstrual cycle. The blood was centrifuged at 2000 g for 10 min, and the serum was stored in polypropylene tubes at −80 °C. Serum AMH was measured through the new automated Access AMH assay (Beckman Coulter). AMH values were then compared with the Roche Diagnostics (Elecsys) nomogram. This nomogram was based on a control population of 887 women not taking contraceptives from 20 to 50 years of age (Roche study No. RD001727). This population included 150 women aged between 20–24 years, 150 aged 25–29 years, 138 aged 30–34 years, 138 aged 35–39 years, 142 aged 40–44 years and, 169 aged 45–50 years. The participants’ AFC values were compared with those of one of the most recent nomogram for AFC in the general population. This nomogram was based on the same control population used in the AMH nomogram by Bozdag: 381 women aged between 20–50 years with regular menstrual bleeding and without hirsutism, menstrual irregularity, diagnosis of current/history of endometrioma, and hormonal drug use within the last 6 months⁴³.

Institutional ethical approval by the Area Vasta Emilia Nord Ethics Committee was granted for the study protocol according to the Declaration of Helsinki principles. Informed consent was obtained from all patients recruited.

Statistics

The primary objective of the study was to evaluate ovarian reserve for healthy carriers of germline P/LPVs in genes predisposing to HBOC syndrome, defined by AMH levels in ng/mL and AFC. Descriptive statistics were produced for the characteristics of the subjects under analysis. Numerical variables were reported as mean and standard deviation (SD), whereas categorical variables were expressed through absolute frequencies and percentages. The correlation between outcome measures and patient characteristics was investigated through univariable and multivariable linear regression models. Age, BMI, and smoking status were included in the models as covariates, in order to adjust the estimation of the relationship between genetic P/LPVs and ovarian reserve.

The association between AMH and AFC was also investigated. Univariable linear regression models were fitted to study this association and assess how it is affected by BRCA1 and BRCA2 P/LPVs. Results were reported in terms of mean difference (MD), with 95% confidence interval (95%CI) and p-values. AMH was measured through the Beckman Coulter Access assay, while the distribution of values was compared with that of AMH in the general population, obtained by Roche Elecsys assay using the conversion factor from Access to Elecsys (Elecsys = −0.06 þ 0.88 x Access)⁴⁴. The analysis of AMH and AFC against the available nomogram curves was performed by comparing the observed frequencies of subjects in each percentile with their expected proportions, stratifying by age. Pearson’s test was used to assess the difference between these proportions. All results were considered significant when associated with a p-value < 0.05. The analyses were carried out through the statistical software R version 4.2.2 (The R Foundation for Statistical Computing, 2022).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Results

Study population

Eighty-seven women were enrolled in the study. Thirty-two women carried a BRCA1 P/LPV, and 47 carried a BRCA2 P/LPV, whereas 8 women had a P/LPV in genes other than BRCA1/2 (two in ATM, two in PALB2, one in CHECK2, one in RAD50, one in RAD51D, and one in TP53).

No significant differences were observed between the epidemiologic characteristics of BRCA1 and BRCA2 P/LPV carriers (Table 1). The mean age at time of enrollment of the whole HBOC cohort was 32.3 years (32.3 in BRCA1 and 32.2 in BRCA2), the mean BMI was 23.5 (24 in BRCA1 and 22.9 in BRCA2), and the mean age at menarche was 12.3 years (11.9 in BRCA1 and 12.4 in BRCA2). Overall, seventy-eight women (89.7%) reported regular menstrual cycles and 49 women (56.3%) reported previous use of hormonal contraceptive methods with a mean duration of previous exposure of 5.8 years (5.2 in BRCA1 and 5.7 in BRCA2). However, none of the patients enrolled were under treatment at the time of recruitment and study. Forty-eight women (55.2%) were nulliparous, and forty-four (50.5%) never had a pregnancy. Seventeen women (19.5%) reported at least one spontaneous termination of pregnancy during the first trimester.

Table 1.

Characteristics of the overall population (HBOC), BRCA1 P/LPV carriers and BRCA2 P/LPV carriers

	HBOC 87 patients			BRCA1 P/LPV carriers 32 patients			BRCA2 P/LPV carriers 47 patients			BRCA1 Vs BRCA2
	Mean	SD	Miss.	Mean	SD	Miss.	Mean	SD	Miss.	p-value
Age	32.31	5.73	0	32.31	5.33	0	32.17	6.32	0	0.914
Hormonal contraceptive use (years)	5.76	5.72	40 (46%)	5.25	5.22	12 (38%)	5.67	6.50	25 (53%)	0.819
Menarche (years)	12.26	1.48	0	11.95	1.40	0	12.40	1.54	0	0.182
Length of menstrual cycle (days)	28.18	2.35	9 (10%)	28.10	2.11	3 (9%)	28.14	2.66	5 (11%)	0.945
BMI	23.52	4.39	1 (1%)	23.98	3.88	0	22.94	4.81	1 (2%)	0.292
AMH (ng/ml)	2.62	1.86	1 (1%)	2.92	2.06	0	2.43	1.75	1 (2%)	0.278
AFC (n)	16.20	9.97	2 (2%)	16.06	9.02	0	15.98	10.96	2 (4%)	0.971
	N	Perc.	Miss.	N	Perc.	Miss.	N	Perc.	Miss.
Previous pregnancies (n)			0			0			0	0.946
0	44	50.57		16	50.00		25	53.19
1	13	14.94		5	15.63		7	14.89
2	21	24.14		9	28.13		10	21.28
3	5	5.75		1	3.13		3	6.38
4	4	4.60		1	3.13		2	4.26
Previous deliveries (n)			0			0			0	0.936
0	48	55.17		18	56.25		26	55.32
1	16	18.39		6	18.75		9	19.15
2	19	21.84		7	21.88		9	19.15
3	3	3,45		1	3.13		2	4.26
4	1	1.15		0	0.00		1	2.13
Previous termination of pregnancy (n)			0			0			0	0.883
0	70	80.46		26	81.25		40	85.11
1	12	13.79		5	15.63		5	10.64
2	5	5.75		1	3.13		2	4.26
Current smoking			0			0			0	0.518
No	71	81.61		24	75.00		40	85.11
Yes	16	18.39		8	25.00		7	14.89
Family history of premature menopause			0			0			0	0,178
No	80	91.95		29	90.63		43	91.49
Yes	7	8.05		3	9.38		4	8.51
Prophylactic surgery			0			0			0	1,000
None	79	90.80		28	87.50		43	91.49
Mastectomy	6	6.90		3	9.38		3	6.38
Salpingectomy	1	1.15		0	0.00		1	2.13
Mastectomy and salpingectomy	1	1,.15		1	3.13		0	0.00

AFC antral follicle count, AMH anti-mullerian hormone, BMI body mass index, HBOC hereditary breast and/or ovarian cancer, Miss missing, Per percentage, P/LPV pathogenic/likely pathogenic variants. SD standard deviation.

Sixteen women (18.4%) were current smokers (8 BRCA1 and 7 BRCA2), and 6 (6.7%) were past smokers. Seven women (8%) reported a positive family history of premature menopause (before the age of 40). Eight women underwent risk-reducing surgery: 6 had risk-reducing mastectomy, 1 had risk-reducing salpingectomy, and 1 had both risk-reducing mastectomy and salpingectomy (3 risk-reducing mastectomy in BRCA1, 3 risk-reducing mastectomy in BRCA2, 1 risk-reducing salpingectomy in BRCA2, and 1 both mastectomy and salpingectomy in BRCA1).

For the evaluation of AMH levels and AFC compared to the nomogram, patients were therefore divided into three groups according to age at enrollment. 29 women (33.3%) belonged to the 20–29 years group, 23 (26.4%) to the 30–34 years group and, 35 (40.2%) to the ≥35 years group.

Serum AMH Levels

Eighty-six patients (32 BRCA1, 46 BRCA2, 8 other genes) had an available AMH level for the analysis. The overall mean AMH level was 2.6 ng/mL (SD 1.85, median 2.32), with 2.92 ng/mL (SD 2.06, median 2.06) in BRCA1 P/LPV carriers and 2.43 ng/mL in BRCA2 P/LPV carriers (SD 1.73, median of 2.18). For AMH mean levels, no significant difference was observed between BRCA1 and BRCA2 P/LPV carriers (p = 0.263). Considering overall the HBOC cohort, the distribution of AMH values in our study population was significantly lower than that reported for the general population (p = 0.019) (Table 2, Fig. 1A). In the 20–29 and ≥35 years of age group, AMH levels were significantly lower compared to the nomogram while, for the 30–34 age group, differences were not statistically significant despite a trend towards lower values.

Table 2.

Distribution of AMH values in the HBOC population, BRCA1 P/LPV carriers and BRCA2 P/LPV carriers compared with the expected nomogram for the general population

	Percentile Distribution in Study Population (HBOC)					Nomogram
Age	0–5	5–50	50–95	95–100	Total	0–5	5–50	50–95	95–100	p-value
20–29	5 (18%)	11 (39%)	12 (43%)	0 (0%)	28 (100%)	5%	45%	45%	5%	0.012
30–34	0 (0%)	13 (57%)	9 (39%)	1 (4%)	23 (100%)	5%	45%	45%	5%	0.567
≥35	5 (14%)	11 (31%)	19 (54%)	0 (0%)	35 (100%)	5%	45%	45%	5%	0.020
Total	10 (12%)	35 (41%)	40 (47%)	1 (1%)	86 (100%)	5%	45%	45%	5%	0.019
	Percentile Distribution in BRCA1 P/LPV carriers					Nomogram
20–29	3 (27%)	4 (36%)	4 (36%)	0 (0%)	11 (100%)	5%	45%	45%	5%	0.008
30–34	0 (0%)	4 (44%)	4 (44%)	1 (11%)	9 (100%)	5%	45%	45%	5%	0.771
≥35	2 (17%)	2 (17%)	8 (67%)	0 (0%)	18 (100%)	5%	45%	45%	5%	0.064
Total	5 (16%)	10 (31%)	16 (50%)	1 (3%)	32 (100%)	5%	45%	45%	5%	0.030
	Percentile Distribution in BRCA2 P/LPV carriers					Nomogram
20–29	2 (13%)	6 (38%)	8 (50%)	0 (0%)	16 (100%)	5%	45%	45%	5%	0.409
30–34	0 (0%)	7 (70%)	3 (30%)	0 (0%)	10 (100%)	5%	45%	45%	5%	0.409
≥35	2 (10%)	8 (40%)	10 (50%)	0 (0%)	20 (100%)	5%	45%	45%	5%	0.528
Total	4 (9%)	21 (46%)	21 (46%)	0 (0%)	46 (100%)	5%	45%	45%	5%	0.312

HBOC hereditary breast and/or ovarian cancer, P/LPV pathogenic/likely pathogenic variants.

Fig. 1. Fertility biomarkers in the overall population.

A AMH values (32 BRCA1, 46 BRCA2, 8 other genes) and (B) AFC (32 BRCA1, 45 BRCA2, 8 other genes) observed vs nomogram curves in the overall population. AMH anti mullerian hormone. AFC antral follicular count.

Consistent findings were observed when considering only women carrying BRCA1/2 P/LPVs (overall p = 0.024, 25–29 years group p = 0.008, shown in Table S1 of Supplementary Materials). Nevertheless, when splitting the BRCA1/2 population into BRCA1 and BRCA2 P/LPV carries, the significant decrease in AMH levels was mostly ascribable to BRCA1 P/LPV carriers (overall p = 0.03), with a significantly lower ovarian reserve in the 20–29 years age group (p = 0.008). In BRCA2 P/LPV carriers, on the other hand, no significant difference was observed compared to the nomogram (p = 0.312).

The distribution of AMH levels in women carrying a P/LPV in genes other than BRCA1/2 is reported in Fig. 2A and Table S2 of Supplementary Materials. However, the sample size is too limited to draw any conclusion.

Fig. 2. Fertility biomarkers in women with P/LPV in genes other than BRCA1/2.

A AMH and (B) AFC observed vs nomogram curves in women with P/LPV in genes other than BRCA1/2. AMH: anti mullerian hormone. AFC antral follicular count.

AFC Evaluation

Eighty-five patients (32 BRCA1, 45 BRCA2, 8 other genes) had available AFC for the analysis. The overall mean AFC was 16.16 (SD 9.86, median 15), with 16.06 (SD 9.02, median 15) in BRCA1 P/LPV carriers and 15.98 in BRCA2 P/LPV carriers (SD 10.73, median 15). No significant difference in mean AFC was identified between BRCA1 and BRCA2 P/LPV carriers (p = 0.971). In the overall HBOC population, AFC was observed to be significantly higher compared to the general population (p < 0.001), particularly between 20 and 34 years of age (p < 0.001). From 35 years onwards, HBOC women showed an overdispersion in terms of AFC, with higher occurrence of extreme values, both higher and lower than the expected distribution in the general population (p = 0.001) (Table 3, Fig. 1B).

Table 3.

Distribution of AFC in the HBOC population, BRCA1 P/LPV carriers and BRCA2 P/LPV carriers compared with the expected nomogram for the general population

	Percentile Distribution in Study Population (HBOC)					Nomogram
Age	0–5	5–50	50–95	95–100	Total	0–10	10–50	50–90	90–100	p-value
20–29	0 (0%)	8 (29%)	9 (32%)	11 (39%)	28 (100%)	10%	40%	40%	10%	3.36e-6
30–34	0 (0%)	0 (0%)	13 (59%)	9 (41%)	22 (100%)	10%	40%	40%	10%	1.96e-7
≥35	9 (26%)	7 (20%)	12 (34%)	7 (20%)	35 (100%)	10%	40%	40%	10%	0.001
Total	9 (11%)	15 (18%)	34 (40%)	27 (32%)	85 (100%)	10%	40%	40%	10%	5.11e-11
	Percentile Distribution in BRCA1 P/LPV carriers					Nomogram
20–29	0 (0%)	3 (27%)	5 (45%)	3 (27%)	11 (100%)	10%	40%	40%	10%	0.178
30–34	0 (0%)	0 (0%)	4 (44%)	5 (56%)	9 (100%)	10%	40%	40%	10%	3.63e-5
≥35	3 (25%)	1 (8%)	6 (50%)	2 (17%)	12 (100%)	10%	40%	40%	10%	0.088
Total	3 (9%)	4 (13%)	15 (47%)	10 (31%)	32 (100%)	10%	40%	40%	10%	0.0001
	Percentile Distribution in BRCA2 P/LPV carriers					Nomogram
20–29	0 (0%)	5 (31%)	4 (25%)	7 (44%)	16 (100%)	10%	40%	40%	10%	0.0001
30–34	0 (0%)	0 (0%)	7 (78%)	2 (22%)	9 (100%)	10%	40%	40%	10%	0.029
≥35	6 (30%)	5 (25%)	4 (20%)	5 (25%)	20 (100%)	10%	40%	40%	10%	0.001
Total	6 (13%)	10 (22%)	15 (33%)	14 (31%)	45 (100%)	10%	40%	40%	10%	1.89e-5

HBOC hereditary breast and/or ovarian cancer, P/LPV pathogenic/likely pathogenic variants.

Among 77 BRCA1/2 P/LPV carriers, AFC was significantly higher up to 34 years (p < 0.001). After 35 years of age, in contrast, AFC values were more variable than expected (p < 0.001) and decreased faster compared to those reported in the nomogram (shown in Table S3 of Supplementary Materials). Splitting by BRCA1 and BRCA2 P/LPV, AFC was significantly higher in both BRCA1 and BRCA2 P/LPV carriers (p < 0.001 in both BRCA1 and BRCA2).

AFC distribution in women carrying a P/LPV in genes other than BRCA1/2 is reported in Fig. 2B and Table S3 of Supplementary Materials. However, the sample size is too limited to draw any conclusion.

For both AMH levels and AFC, no significant correlation was observed with BMI or smoking habits. AMH levels and AFC positively correlated both in the whole sample (MD = 0.13, 95%CI:0.11–0.16, p < 0.001) and in the BRCA1/2 subpopulations (BRCA1: MD = 0.16, 95%CI:0.10–0.22, p < 0.001; BRCA2: MD = 0.13, 95%CI:0.10–0.16, p < 0.001). Sufficient evidence has not emerged to draw conclusions on the possible effects of BRCA1/2 P/LPVs on the relationship between AFC and AMH levels.

Discussion

In the last decades, while oncological risks associated with germline P/LPVs in HBOC syndrome-related genes have been extensively studied, little has become known about the non-oncological implications of these genetic alterations, particularly in terms of how these disorders can affect fertility. Moreover, whether patients with BRCA1/2 germline P/LPVs have a lower ovarian reserve than non-carriers is still an unsolved issue^{26–28,45–52}. This hypothesis was firstly put forward based on the observation that BRCA1/2 P/LPV carriers seemed to have a lower response to gonadotropins during ovarian stimulation for fertility preservation^46,53. Alterations to these genes can lead to oocyte apoptosis and premature follicular depletion in vitro²⁸. In addition, some authors reported that female carriers tend to experience premature menopause, increased ovarian aging, decreased ovarian reserve, and lower ovarian response to ovarian stimulation than non-carriers^27,45.

AMH is currently one of the best biomarkers to predict menopausal age and reproductive life span^37,40,54,55. AMH is solely produced from granulosa cells of antral follicles. AMH production begins during intrauterine life. It reaches a peak in the mid-20s and then decreases progressively until menopause, when it becomes undetectable⁵⁶. In our study, healthy carriers of HBOC syndrome showed lower AMH levels compared to the nomogram. The decrease in AMH levels was significant starting at 20 to 29 and over 35 of age, but was also present in women aged 30–34 years (lack of statistical significance in this age group could be due to the smaller sample size). Interestingly, splitting the study population into BRCA1 and BRCA2 P/LPV carries, these lower values were mostly ascribable to BRCA1 P/LPV carriers, starting from 20 years of age. Although the available literature is conflicting, these results are in line with two recent meta-analysis showing that women with germline BRCA1/2 P/LPV had significantly lower AMH levels compared to controls. In these metanalysis as well, AMH levels were lower in women with BRCA1 P/LPV, but not in BRCA2 P/LPV carriers, compared to controls^57,58.

Our study also included AFC evaluation. In our analysis, AFC was significantly higher up to 34 years, but it decreased faster than that in the general population after 35 years of age. Higher overall AFC may reflect a higher sensitivity of current instruments for TV-US, compared to those available when the nomogram was defined⁴³. Even taking this bias into account, however, the significant decrease in AFC after 35 years of age corroborates the hypothesis that ovarian reserve in HBOC syndrome decreases significantly and faster than in the general population.

As in the general population, AMH and AFC were found to be positively correlated both in the whole sample and in the BRCA1/2 subpopulations. Nevertheless, the mean difference between the two was lower than in the previous literature^59,60. Whether this means that the relationship between AFC and AMH is less strong in BRCA1/2P/LPV carriers should be investigated in future research. BRCA1 and BRCA2 play a key role in DNA double-strand break (DSB) repair through homologous recombination, as well as in chromosomal stability, apoptosis, and cell cycle⁶¹. Moreover, other genes belonging to the homologous recombination pathway have been identified as predisposing factors to premature ovarian insufficiency, such as ATM bi-allelic mutation⁶². In humans, oocyte meiosis starts during embryogenesis and concludes many years later. Oocyte meiosis has a long stop at the stage of meiosis I, ranging from fetal life to even several decades after ovulation. In this context, the accumulation of DSBs and the role of BRCA1 and BRCA2 in maintaining chromosome telomere integrity can be associated with ovarian aging and reproductive senescence^48,63–66. Therefore, the role of BRCA1 in the DSB repair mechanism could be at the basis of the lower ovarian reserve in BRCA1 P/LPV carriers. It remains to be fully elucidated why the same process seems not to affect BRCA2 P/LPV carriers.

Several studies indicated that BRCA1/2P/LPVs carriers experience early menopause^27,45,47, nevertheless, menstruation in not the finest indicator of ovarian reserve and may be influenced by other conditions. Overall, the possible modest reduction in ovarian reserve in BRCA1/2P/LPVs carriers and the likely earlier onset of menopause seems not to impact on natural fertility. Indeed, no difference was found regarding parity and the use of assisted reproductive techniques between BRCA1/2 P/LPVs carriers and general populations, indicating normal fertility in BRCA1/2 P/LPVs carriers^50,67–70. Finally, the rate of spontaneous termination of pregnancy in our HBOC population was in line with the previous literature⁷¹.

To our knowledge, this is one of the largest studies including healthy women with P/LPVs in breast and/or ovarian cancer predisposition genes. At the same time, it is the first study to also include high and moderate penetrant genes defining HBOC syndrome, other than BRCA1/2. Moreover, all women in our study were prospectively examined by the same gynecologist (G.S.), AMH was assayed at the same laboratory, and fertility evaluation also included AFC, along with AMH. In addition, all women enrolled were under no hormonal contraceptive methods at the time of evaluation. Nonetheless, our analysis presents some limitations. In particular, our study lacks a control group. Our results have, therefore, been compared with previously published nomograms. In addition, the sample size of women with P/LPV in genes other than BRCA1/2 is still too limited to drive any conclusion from this subset.

It is important to underline that lower AMH values in BRCA1/2 P/LPV carriers do not translate into less fertility. Such values, however, result in a shorter reproductive window in a lifetime. Our study underlines the importance of counseling, so that young healthy carriers do not delay pregnancy, especially with a BRCA1 P/LPV. Furthermore, it is crucial to evaluate the possibility of oocyte cryopreservation at a younger age and in elective settings, when ovarian stimulation may yield more better-quality oocytes. In addition, due to the possible existence of poor ovarian reserve as well as DNA damage-induced follicle death, it can be hypothesized that BRCA1/2 P/LPV carriers may be at a higher risk of chemo-induced amenorrhea during chemotherapy⁴⁶. It has already been reported that chemotherapy induces DNA DSBs both in mice^72–74 and in human oocytes^75–77, therefore, BRCA1/2 P/LPVs carriers are potentially more sensitive to gonadotoxicity of chemotherapy compared to non-carriers. Particularly, higher gonadotoxicity has been reported with alkylating and platinum-based agents, taxanes, anthracyclines, topoisomerase inhibitors and vinca alkaloids⁷⁸. The impact of endocrine therapies (including tamoxifen) on fertility in BRCA1/2 P/LPVs carriers are still conflicting^79–81, whereas data on new therapies developed for BRCA1/2 P/LPVs patients (including PARP inhibitors) are substantially lacking³². As a result, BRCA1/2 P/LPV carriers should be counseled for ovarian suppression and preservation before the start of any oncological treatment.

To conclude, as previously highlighted by Peccatori et al.³⁰, the fertility issues that BRCA patients may face, lead to some considerations: (1) the evidence on a shortening of reproductive life and a possible premature diminishing of ovarian reserve in BRCA P/LPVs carriers are increasing; (2) when the risk-reducing bilateral salpingo-oophorectomy is chosen at 35 years old, oocyte cryopreservation prior to surgery should be considered; (3) a policy of oocyte cryopreservation in young carriers before cancer development deserves consideration. Particularly, fertility preservation strategies should be considered at a young age and in an elective setting, when ovarian stimulation may yield more oocytes of better quality. (4) Preimplantation genetic diagnosis (PGD) may be considered since it could avoid the possibility of transmitting the mutation to the offspring^30,31. Our study confirms a consistent trend for reduced reproductive potential in young healthy women carrying germline P/LPVs in genes associated with HBOC syndrome, particularly in BRCA1 P/LPV carriers. These findings need to be confirmed by larger studies that also include carriers of germline P/LPVs in other tumor suppressor genes, since these may influence fertility counseling for this specific women population.

Author contributions

Concept and design: G.S., G.G., A.T. Acquisition, analysis, or interpretation of data: G.S., E.B., M.V., C.M., R.L.B., R.C.C., R.D., E.T. Drafting of the manuscript: G.S., A.T., C.P. Critical review of the manuscript for important intellectual content: M.L., F.A.P., M.D., L.C., A.L.M. Statistical analysis: R.C.C., R.D. Obtained funding: A.T. Administrative, technical, or material support: G.S., C.M., R.L.B. Supervision: G.S., A.T.

Peer review

Peer review information

Communications Medicine thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Data availability

The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request. Data are located in controlled access data storage at the Azienda Ospedaliero-Universitaria di Modena. The source data for Figs. 1, 2 can be found in Supplementary Data 1 (for Fig. 1) and Supplementary Data 2 (for Fig. 2).

Competing interests

G.S., G.G., M.V., C.P., C.M., R.L.B., R.C.C., R.D., F.A.P., and M.D. reported no conflicts of interest. E.B. reported travel grants from Lilly, Novartis, Pfizer, MSD, AstraZeneca, Gilead, and Daiichi Sankyo, all outside the submitted work. M.L. reported advisory role for Roche, Lilly, Novartis, AstraZeneca, Pfizer, Seagen, Gilead, MSD, Exact Sciences; speaker honoraria from Roche, Lilly, Novartis, Pfizer, Sandoz, Libbs, Daiichi Sankyo, Takeda, Knight, Ipsen, AstraZeneca; travel grants from Gilead, Daiichi Sankyo, Roche and a research grant (to his Institution) from Gilead, all outside the submitted work. E.T. reported travel grants and speaker honoraria from MSD and SOPHiA GENETICS, all outside the submitted work. L.C. reported personal fees from AstraZeneca, Pfizer, Novartis, Gilead, and Clovis and grants from MSD outside the submitted work. A.L.M. reported grants from Merck, Ferring Pharmaceuticals, MSD, IBSA MSD, Roche Diagnostics, Beckmann Coulter, and TEVA, all outside the submitted work. A.T. reported conflicts of interest with Lilly, Novartis, Pfizer, AstraZeneca, MSD, Seagen, Gilead, and Daiichi Sankyo, all outside the submitted work.

Supplementary information

The online version contains supplementary material available at 10.1038/s43856-025-00788-9.