Ovarian Aging in Women With BRCA Germline Mutations

Abstract

Context: Recent clinical and laboratory studies suggested that women with BRCA mutations have lower ovarian reserve and their primordial follicle oocytes may be more prone to DNA damage; however, direct proof is lacking.

Objective: To determine whether women with germline BRCA mutations have reduced primordial follicle reserve and increased oocyte DNA damage.

Design: A comparative laboratory study of ovarian tissue obtained from unaffected BRCA mutation carriers (BMCs) vs age-matched organ donor cadavers.

Setting: Two academic centers.

Patients or other participants: Of the 230 ovarian specimens from BMCs, 18 met the study inclusion criteria. Healthy ovaries from 12 organ donor cadavers served as controls.

Intervention: Histology and immunohistochemical analysis on paraffin-embedded ovarian sections.

Main outcome measure(s): Primordial follicle density and the percentage of DNA double-strand break (DSB)-positive primordial follicle oocytes.

Results: Ovaries from BMCs had significantly lower primordial follicle densities than those of controls (11.2 ± 2.0 vs 44.2 ± 6.2 follicles/mm3; P = 0.0002). BRCA mutations were associated with increased DNA DSBs in primordial follicle oocytes (62% ± 5.2% vs 36% ± 3.4%; P = 0.0005). In subgroup analyses, both BRCA1 and BRCA2 mutations were associated with lower primordial follicle density (P = 0.0001 and 0.0030, respectively), and BRCA1 mutations were associated with higher DNA DSBs (P = 0.0003) than controls. The rates of follicle decline (R2 = 0.74; P = 0.0001) and DNA DSB accumulation (R2 = 0.70; P = 0.0001) appeared to be accelerated, particularly in primordial follicle oocytes of BMCs over age 30 years.

Conclusions: We provide direct evidence of diminished ovarian reserve as well as accelerated primordial follicle loss and oocyte DNA damage in women with BRCA mutations. These findings may further our understanding of ovarian aging, and be useful when counseling BMCs.

Figure 1.

Effect of BRCA mutations on primordial follicle reserve. (A) The mean primordial follicle density was lower in the ovaries of BMCs than in those of controls (11.2 ± 1.59 vs 44.18 ± 6.16 follicles/mm³; P = 0.0002). (B) Both BRCA1 (P = 0.001) and BRCA2 (P = 0.0003) mutations were associated with low primordial follicle reserve. The error bars represent standard error of the mean. Statistical significance is denoted by asterisks (*). (C and D) Representative photomicrographs of primordial follicle density in (C) a control and (D) a patient with a BRCA1 mutation. Arrows show the primordial follicles. (E) Comparison of the linear regression curves by multivariate analysis indicates that there is a sharper decline of primordial follicle reserve (R² = 0.74; P = 0.0001) in BMC ovaries (slope = −0.1270) than in control ovaries (slope = −0.0634). This difference in the slopes between BMC (slope = −0.2243) and controls (slope = −0.051) became more distinct when the linear regression was limited to those aged >30 years. The squares represent the control group, and the triangles represent the BMCs. Dotted lines represent the analyses that were limited to age >30 years in each group. Refer to the text for complete linear regression analysis results.

Figure 2.

Effect of BRCA mutations on DNA damage in primordial follicle oocytes. (A) The proportion of γH2AX-positive primordial follicle oocytes was higher in the ovaries of BMCs than in the ovaries of controls (59% ± 5% vs 36% ± 3%; P = 0.0005). (B) Compared with controls (36% ± 3%), a higher likelihood of DNA DSBs was observed in the primordial follicle oocytes of the BRCA1 subgroup (63% ± 5%; P = 0.0003) but not the BRCA2 subgroup (51% ± 11%; P = 0.24). The error bars represent standard error of the mean. Statistical significance is denoted by asterisks (*). (C) Comparison of the linear regression curves by multivariate analysis indicates that there was a faster accumulation of DNA breaks (R² = 0.57; P = 0.002) in primordial follicle oocytes of BMC ovaries (slope = 0.0530) than in control ovaries (slope = 0.0390). This difference between the BMCs (slope = 0.1263) and controls (slope = 0.081) became more distinct when the analysis was limited to those aged >30 years. The squares represent the control group, and the triangles represent the BMCs. Dotted lines represent the analyses that were limited to age >30 years in each group. Refer to the text for linear regression results. Photomicrographic representation of γH2AX expression in (D) controls and (E) BMCs. (D) Arrowheads show the γH2AX⁻ follicles in controls. (E) Arrows mark the γH2AX⁺ follicles in BMCs.