Comparing GDF9 in mature follicles and clinical outcomes across different PCOS phenotype

Background: Polycystic ovary syndrome (PCOS) is main cause of anovulatory infertility in women with gestational age. There are currently four distinct phenotypes associated with individualized endocrinology and metabolism. Growth differentiation factor 9 (GDF9) is a candidate as potential biomarker for the assessment of oocyte competence. The effect on oocyte capacity has not been evaluated and analyzed in PCOS phenotypes.

Objective: We aimed to screen the expression levels of GDF9 in mature follicles of women with controlled ovarian hyperstimulation (COS) with different PCOS phenotypes. To determine the correlation between the expression level of GDF9 and oocyte development ability.

Methods: In Part 1, we conducted a retrospective study comparing the clinical outcomes and endocrine characteristics of patients with PCOS according to different subgroups (depending on the presence or absence of the main features of polycystic ovarian morphology (PCOM), hyperandrogenism (HA), and oligo-anovulation (OA)) and non-PCOS control group. We stratified PCOS as phenotype A (n = 29), phenotype B (n = 18) and phenotype D (n = 24). In Part 2, the expression of GDF9 in follicular fluid (FF) and cumulus cells (CCs) were detected by enzyme-linked immunosorbent assay (ELISA) and immunohistochemistry, respectively.

Results: In Part 1, the baseline clinical, hormonal, and ultrasonographic characteristics of the study population were matched with the presence or absence of the cardinal features of each PCOS phenotypes showed a clear difference. Phenotypes A and D had statistically significant associations with blastocyst formation and clinical pregnancy compared with phenotypes B (p < 0.001). In Part 2, the levels of GDF9 in FF and CCs for phenotype A and B were significantly were higher than those of phenotype D (P = 0.019, P = 0.0015, respectively). Multivariate logistic regression analysis showed that GDF9 was an important independent predictor of blastocyst formation (P＜0.001). The blastocyst formation rate of phenotype A was higher than that of phenotype B and D (P＜0.001). Combining the results of the two parts, GDF9 appears to play a powerful role in the development of embryos into blastocysts.

Conclusions: GDF9 expression varies with different PCOS phenotypes. Phenotype A had higher GDF9 levels and blastocyst formation ability.