Investigation of the role of X chromosome inactivation and androgen receptor CAG repeat polymorphisms in patients with recurrent pregnancy loss: a prospective case–control study

Abstract

Abstract Background Previous research has revealed that skewed X chromosome inactivation (SXCI) and androgen receptor (AR) CAG polymorphisms are associated with increased risk of recurrent pregnancy loss (RPL); however, the results are conflicting, and the underlying mechanisms remain unclear. This study investigated the role of SXCI and AR CAG polymorphisms in patients with RPL and explored whether the underlying mechanisms were related to the ovarian reserve and preimplantation embryo aneuploidy. Methods This was a prospective case-control study carried out in a tertiary hospital-based reproductive medicine center. An external validation RPL cohort was recruited during the study period. Data on baseline and cycle characteristics were collected. X-chromosome inactivation (XCI) was measured using a human AR assay. AR polymorphisms were assessed using quantitative fluorescent polymerase chain reactions and direct sequencing. Blastocysts of the patients with RPL were tested by single nucleotide polymorphism microarray based preimplantation genetic testing for aneuploidy. Results In total, 131 patients with idiopathic RPL and 126 controls were included for the case-control study. Patients with RPL exhibited a significantly more skewed XCI distribution pattern (67.71 ± 10.50 vs. 64.22 ± 10.62, p = 0.011), as well as significantly shorter bi-allelic mean (18.56 ± 1.97 vs. 19.34 ± 2.38, p = 0.005) and X-weighted bi-allelic mean (18.46 ± 2.02 vs. 19.38 ± 2.53, p = 0.001) of AR CAG repeats. Multivariate logistic regression models indicated that CAG repeat < 20, SXCI, and duration of stimulation were independently associated with the risk of RPL. However, SXCI and AR CAG polymorphisms were not associated with ovarian reserve or preimplantation embryo aneuploidy in the RPL group, and the same results were attained in a separate validation cohort of 363 patients with RPL. Conclusion SXCI and AR CAG polymorphisms are related to RPL; however, these two factors do not lead to RPL by affecting the ovarian reserve or increasing embryo aneuploidy. The roles of SXCI and AR CAG in RPL may involve other mechanisms that require further investigation. Trial registration: NCT02504281, https://www.clinicaltrials.gov (Date of registration, 21/07/2015; date of enrolment of the first subject, 30/07/2015).