Metabolic and hormonal biomarkers in pregnancies with polyendocrine metabolic ovarian syndrome (formerly polycystic ovary syndrome): a nested prospective cohort study with longitudinal sub-analysis

Abstract

Do metabolic and hormonal biomarker profiles in early pregnancy and their longitudinal variations across gestation differ among women at high risk of gestational diabetes mellitus (GDM) with and without polyendocrine metabolic ovarian syndrome (PMOS, formerly polycystic ovary syndrome [PCOS]), and are they associated with maternal and neonatal complications? Among pregnant women at high risk of GDM, PMOS was associated with persistently elevated free androgen index (FAI) throughout pregnancy, whereas no interaction between elevated FAI and PMOS status on maternal and neonatal complications was identified. Insulin resistance and hyperandrogenism are important pathophysiologic features of PMOS and are also part of the physiological changes during normal pregnancy. Pregnancies with PMOS have been associated with a higher risk of maternal and neonatal complications than those without PMOS. Nested prospective cohort study with longitudinal sub-analysis using pooled data from two international multicentre clinical trials (TOBOGM study and B2B&Me trial). A total of 151 pregnancies at high risk for GDM were included from January 2018 to December 2021. At baseline (before 24 weeks' gestation), 151 pregnancies at high risk for GDM were included (65 PMOS and 86 non-PMOS). Longitudinal sub-analysis included 75 pregnancies (15 PMOS and 60 non-PMOS). Metabolic biomarkers (fasting plasma glucose [FPG], insulin, C-peptide) and hormonal biomarkers (testosterone, dihydrotestosterone [DHT], androstenedione [A4], sex hormone-binding globulin [SHBG], and anti-Mullerian hormone [AMH]) were measured to derive indices of insulin resistance (homeostatic model assessment for insulin resistance [HOMA-IR], quantitative insulin sensitivity check index [QUICKI], fasting glucose to insulin ratio [FGIR]) and FAI. These were assessed cross-sectionally and in exploratory longitudinal sub-analyses of a subset of participants across gestation using linear mixed-effects models. Interaction analyses explored whether biomarkers modified associations between PMOS and adverse maternal or neonatal outcomes. PMOS was associated with higher pre-pregnancy and baseline BMI and greater insulin resistance (elevated fasting insulin, HOMA-IR, and FGIR, and lower QUICKI). Androgen profiles differed significantly, with PMOS associated with higher DHT (167.8 vs. 121.2 pg/mL, p = 0.003) and FAI (0.9 vs. 0.5, p < 0.001) and lower SHBG (159.1 vs. 215.3 nmol/L, p = 0.012), compared with non-PMOS controls. FAI in early pregnancy was independently associated with PMOS (geometric mean ratio 1.9, p = 0.003) and, in exploratory longitudinal sub-analyses, remained elevated throughout pregnancy in women with PMOS. Maternal and neonatal complications did not differ between groups, and no interactions between FAI, PMOS status, and composite adverse outcomes were identified. The selective inclusion of a high-risk cohort limited generalisation. Minimised or neutralised differences in maternal and neonatal outcomes between groups may be observed due to timely screening, diagnosis, and intervention within the parent RCT protocols. Since this was a sub-study, no formal a priori power calculation was conducted. Consequently, the study may possess limited statistical power to identify smaller effect sizes, longitudinal changes, and interaction effects; therefore, the findings should be interpreted with caution. Persistently elevated FAI throughout pregnancy, potentially reflecting an inadequate protective SHBG response to pregnancy-induced androgen excess. Whether elevated FAI contributes to adverse pregnancy outcomes in PMOS warrants further longitudinal evaluation. TOBOGM was funded by the National Health and Medical Research Council (NMHRC; grants 1104231 and 2009326), the Region Örebro Research Committee (grants Dnr OLL-970566 and OLL-942177), Medical Scientific Fund of the Mayor of Vienna (project 15205), the South-Western Sydney Local Health District Academic Unit (grant 2016), and a Western Sydney University Ainsworth Trust Grant (2019). B2B&Me was funded by the European Union Commission Horizon 2020 grant entitled ‘Implementation Action to Prevent Diabetes from Bump 2 Baby (IMPACT DIABETES B2B&Me)’ under grant agreement 847984, with collaborative NHMRC, Australia co-funding under grant number APP1194234. The project was sponsored by the University College Dublin. SA, HCHN, and TM are supported by Graduate Research Scholarships provided by Monash University. KR is supported by a scholarship from the NHMRC. SJE, HT, and AM are supported by fellowships from the NHMRC. The authors declare no conflicts of interest. ACTRN12616000924459 and ACTRN12620001240932.