Molecular Autism (Aug 2023)

Sex-specific and sex-independent steroid-related biomarkers in early second trimester maternal serum associated with autism

  • Deborah A. Bilder,
  • Whitney Worsham,
  • Scott Sullivan,
  • M. Sean Esplin,
  • Paul Burghardt,
  • Alison Fraser,
  • Amanda V. Bakian

DOI
https://doi.org/10.1186/s13229-023-00562-5
Journal volume & issue
Vol. 14, no. 1
pp. 1 – 12

Abstract

Read online

Abstract Background Prenatal exposure to maternal metabolic conditions associated with inflammation and steroid dysregulation has previously been linked to increased autism risk. Steroid-related maternal serum biomarkers have also provided insight into the in utero steroid environment for offspring who develop autism. Objective This study examines the link between autism among offspring and early second trimester maternal steroid-related serum biomarkers from pregnancies enriched for prenatal metabolic syndrome (PNMS) exposure. Study design Early second trimester maternal steroid-related serum biomarkers (i.e., estradiol, free testosterone, total testosterone, and sex hormone binding globulin) were compared between pregnancies corresponding to offspring with (N = 68) and without (N = 68) autism. Multiple logistic regression analyses were stratified by sex and gestational duration. One-way ANCOVA with post hoc tests was performed for groups defined by autism status and PNMS exposure. Results Increased estradiol was significantly associated with autism only in males (AOR = 1.13 per 100 pg/ml, 95% CI 1.01–1.27, p = 0.036) and only term pregnancies (AOR = 1.17 per 100 pg/ml, 95% CI 1.04–1.32, p = 0.010). Autism status was significantly associated with decreased sex hormone binding globulin (AOR = 0.65 per 50 nmol/L, 95% CI 0.55–0.78, p < 0.001) overall and when stratified by sex and term pregnancy status. The inverse association between sex hormone binding globulin and autism was independent of PNMS exposure. Limitations The relative racial and ethnic homogeneity of Utah’s population limits the generalizability of study results. Although significant differences by autism status were identified in concentrations of sex hormone binding globulin overall and of estradiol in participant subgroups, differences by PNMS exposure failed to reach statistical significance, which may reflect insufficient statistical power. Conclusion Both elevated maternal serum estradiol in males only and low maternal serum sex hormone binding globulin in both sexes are associated with increased autism risk. Further investigation is merited to identify how steroid, metabolic, and inflammatory processes can interact to influence neurodevelopment in early second trimester.

Keywords