Serum metabolomics study of women with different annual decline rates of anti-Müllerian hormone: an untargeted gas chromatography-mass spectrometry-based study

Study question: Which metabolites are associated with varying rates of ovarian aging, measured as annual decline rates of anti-Müllerian hormone (AMH) concentrations?

Summary answer: Higher serum concentrations of metabolites of phosphate, N-acetyl-d-glucosamine, branched chained amino acids (BCAAs), proline, urea and pyroglutamic acid were associated with higher odds of fast annual decline rate of AMH.

What is known already: Age-related rate of ovarian follicular loss varies among women, and the factors underlying such inter-individual variations are mainly unknown. The rate of ovarian aging is clinically important due to its effects on both reproduction and health of women. Metabolomics, a global investigation of metabolites in biological samples, provides an opportunity to study metabolites or metabolic pathways in relation to a physiological/pathophysiological condition. To date, no metabolomics study has been conducted regarding the differences in the rates of ovarian follicular loss.

Study design, size, duration: This prospective study was conducted on 186 reproductive-aged women with regular menstrual cycles and history of natural fertility, randomly selected using random case selection option in SPSS from the Tehran Lipid and Glucose Study.

Participants/materials, setting, methods: AMH concentrations were measured at baseline (1999-2001) and the fifth follow-up examination (2014-2017), after a median follow-up of 16 years, by immunoassay using Gen II kit. The annual decline rate of AMH was calculated by dividing the AMH decline rate by the follow-up duration (percent/year). The women were categorized based on the tertiles of the annual decline rates. Untargeted metabolomics analysis of the fasting-serum samples collected during the second follow-up examination cycle (2005-2008) was performed using gas chromatography-mass spectrometry. A combination of univariate and multivariate approaches was used to investigate the associations between metabolites and the annual decline rates of AMH.

Main results and the role of chance: After adjusting the baseline values of age, AMH and BMI, 29 metabolites were positively correlated with the annual AMH decline rates. The comparisons among the tertiles of the annual decline rate of AMH revealed an increase in the relative abundance of 15 metabolites in the women with a fast decline (tertile 3), compared to those with a slow decline (tertile 1). There was no distinct separation between women with slow and fast decline rates while considering 41 metabolites simultaneously using the principal component analysis and the partial least-squares discriminant analysis models. The odds of fast AMH decline was increased with higher serum metabolites of phosphate, N-acetyl-d-glucosamine, BCAAs, proline, urea and pyroglutamic acid. Amino sugar and nucleotide sugar metabolism, BCAAs metabolism and aminoacyl tRNA biosynthesis were among the most significant pathways associated with the fast decline rate of AMH.

Limitations, reasons for caution: Estimating the annual decline rates of AMH using the only two measures of AMH is the main limitation of the study which assumes a linear fixed reduction in AMH during the study. Since using the two-time points did not account for the variability in the decline rate of AMH, the annual decline rates estimated in this study may not accurately show the trend of the reduction in AMH. In addition, despite the longitudinal nature of the study and statistical adjustment of the participants' ages, it is difficult to distinguish the AMH-related metabolites observed in this study can accelerate ovarian aging or they are reflections of different rates of the aging process.

Wider implications of the findings: Some metabolite features related to the decline rates of AMH have been suggested in this study; further prospective studies with multiple measurements of AMH are needed to confirm the findings of this study and to better understand the molecular process underlying variations in ovarian aging.

Study funding/competing interest(s): This study, as a part of PhD thesis of Ms Nazanin Moslehi, was supported by Shahid Beheshti University of Medical Sciences (10522-4). There were no competing interests.

Trial registration number: N/A.