Changes in Circulating ProAMH and Total AMH during Healthy Pregnancy and Post-Partum: A Longitudinal Study

Abstract

Circulating Anti-Müllerian hormone (AMH) is derived from the gonads, and is a mixture of the prohormone (proAMH), which does not bind to AMH receptors, and receptor-competent AMH. The functions of a hormone are partially defined by the factors that control its levels. Ovarian reserve accounts for 55~75% of the woman-to-woman variation in AMH level, leaving over 25% of the biological variation to be explained. Pregnancy has been reported to decrease circulating AMH levels, but the observations are inconsistent, with the effect of pregnancy on the bioactivity of AMH being unknown. We have therefore undertaken a longitudinal study of circulating proAMH and total AMH during pregnancy. Serum samples were drawn at 6-8 gestational time-points (first trimester to post-partum) from 25 healthy women with prior uneventful pregnancies. The total AMH and proAMH levels were measured at each time-point using ELISA. The level of circulating total AMH progressively decreased during pregnancy, in all women (p<0.001). On average, the percentage decline between the first trimester and 36-39 weeks' gestation was 61.5%, with a standard deviation of 13.0% (range 30.4-81.2%). The percentage decline in total AMH levels associated with maternal age (R = -0.53, p = 0.024), but not with the women's first trimester AMH level. The postpartum total AMH levels showed no consistent relationship to the woman's first trimester values (range 31-273%). This raises the possibility that a fundamental determinant of circulating AMH levels is reset during pregnancy. The ratio of proAMH to total AMH levels exhibited little or no variation during pregnancy, indicating that the control of the cleavage/activation of AMH is distinct from the mechanisms that control the total level of AMH.

Conflict of interest statement

ISM and MWP have filed a patent pertaining to cleavage-state-specific AMH assays including the proAMH-specific assay used in this manuscript (WO 2014/204327 A1). This does not alter our adherence to PLOS ONE policies on sharing data and materials. CAC, JZ and CAL have declared that no competing interests exist.

Figures

Fig 1. Change in total AMH levels during gestation.

(A) Each woman’s individual levels are shown. (B) Box and whisker plots (medians, interquartile intervals, range) of AMH levels. (C) Percentage decline. Each woman’s values were normalized to her first trimester (6–11 weeks) sample, using log transformation. The data is the mean ± standard of error of the mean. There is a significant decline with gestational age, p

Fig 2. The magnitude of pregnancy-related decline was dissociated from a woman’s level of AMH.

Each woman’s level of total AMH during the first trimester is plotted against the extent to which her total AMH levels declined by 36–39 weeks gestation. The two measures showed no significant correlation (R = - 0.04). 1 ng/ml of AMH = 7.14 pmol/L (pM).

Fig 3. The influence of maternal age on the pregnancy-associated decline in total AMH level.

The percentage decrease in total AMH levels between the first trimester and the 36–39 week samples is plotted against the women’s ages. The two parameters were significantly correlated, R = -0.53, p = 0.024.

Fig 4. Change in the proportion of the AMH forms during gestation.

(A) ProAMH levels during pregnancy. Each woman’s values were normalized to her first trimester (6–11 weeks) sample, using log transformation. The data is the mean ± standard of error of the mean. There is a significant decline with gestational age, p