Serum persistent organic pollutants and diminished ovarian reserve: a single-exposure and mixture exposure approach from a French case-control study

STUDY QUESTION Are persistent organic pollutants (POPs) associated with a diminished ovarian reserve (DOR) in women of reproductive age? SUMMARY ANSWER Amongst 17 POPs detected in over 20% of serum samples, only p,p '-DDE was significantly associated with an increased risk of DOR, and beta-hexachlorocyclohexane (beta-HCH) was significantly associated with a decreased risk of DOR whilst mixture analyses yielded non-significant associations and did not detect any interactions between POPs. WHAT IS KNOWN ALREADY Animal studies have shown that several POPs can alter folliculogenesis and increase follicle depletion. However, only a few studies have been conducted in humans, with small sample sizes and inconsistent results. STUDY DESIGN, SIZE, DURATION Our study included 138 cases and 151 controls from the AROPE case-control study. Study participants were women between 18 and 40 years of age recruited amongst couples consulting for infertility in four fertility centres in western France between 2016 and 2020. PARTICIPANTS/MATERIALS, SETTING, METHODS Cases of DOR were defined as women with anti-Mullerian hormone (AMH) levels <= 1.1 ng/ml and/or antral follicle count (AFC) <7, and controls were women with AMH levels between 1.1 and 5 ng/ml and AFC >= 7, without genital malformations and with a menstrual cycle length between 26 and 35 days. A total of 43 POPs (including 15 organochlorine pesticides, 17 polychlorinated biphenyls, and 9 polybromodiphenylethers) were measured in the serum at inclusion into the study. We conducted logistic regression adjusted for potential confounders using a directed acyclic graph to study the effect of each POP on DOR as single exposures, and used Bayesian kernel machine regression (BKMR) to measure the mixture effect of POPs on DOR. MAIN RESULTS AND THE ROLE OF CHANCE Of the 43 POPs, 17 were detected in over 20% of the serum samples. In the single-exposure multivariate logistic regressions, p,p '-DDE (median 165.0 IQR 161.0 ng/l in controls) as a continuous exposure was significantly associated with an increased risk of DOR (odds ratio (OR) 1.39, 95% CI 1.10-1.77) and non-significantly associated with an increased risk of DOR for the second and third terciles (OR 1.46, 95% CI 0.74-2.87, and OR 1.72, 95% CI 0.88-3.37, respectively). beta-HCH (median 24.2 IQR 21.5 ng/l in controls) was significantly associated with a decreased risk of DOR when beta-HCH was treated as a continuous exposure (OR 0.63, 95% CI 0.44-0.89) and for the third tercile of exposure (OR 0.43, 95% CI 0.21-0.84) and non-significantly associated with a decreased risk of DOR for the second tercile (OR 0.77, 95% CI 0.42-1.42). All sensitivity analyses confirmed our results. BKMR showed similar associations for single exposures but found no significant associations for the total mixture effect. In addition, the BKMR results did not suggest any interactions between POPs. LIMITATIONS, REASONS FOR CAUTION Controls were recruited amongst infertile couples and thus may not be representative of all women of reproductive age. However, their POP concentrations were in the same range as in the general French population. WIDER IMPLICATIONS OF THE FINDINGS This study is the first to examine the associations between serum POPs and DOR. The well-recognized anti-androgenic properties of p,p '-DDE and estrogenic properties of beta-HCH could explain these associations of opposite direction. If these results are replicated elsewhere, this could have an impact on fertility prevention messages and help in understanding the impact of POPs on the female reproductive system. STUDY FUNDING/COMPETING INTEREST(S) This study was funded by the Fondation de France (grant numbers 2014-50537 and 00110196) and the French Biomedicine Agency (2016). None of the authors have any conflicts of interest to declare.