Polycystic ovary syndrome (PCOS) is a diagnosis of exclusion with unknown etiology. We recently identified autoantibodies (AAbs) to the second extracellular loop (ECL-2) of the gonadotropin releasing hormone receptor (GnRHR) in a majority of patients with PCOS compared to ovulatory controls in baseline serum measurements. The present assay may represent the desired serological test needed to effectively screen subjects for possible PCOS. Currently, no data exists regarding autoantibody activity over time or with changes in the hormonal milieu associated with infertility treatments, specifically controlled ovarian hyperstimulation (COH) with in vitro fertilization (IVF). To evaluate the levels of AAbs to GnRHR in the serum of eleven subjects with PCOS and infertility over time and observe for changes relative to hormonal fluctuations occurring over the course of infertility treatment. Sera samples from eleven women with PCOS and infertility drawn over multiple time points (0: not on medications, 1: down-regulation phase of an IVF cycle (oral contraceptives or lupron), 2: IVF stimulation day five, 3: day of hCG trigger, and 4: day of pregnancy test) during infertility treatments were assessed for GnRHR-AAbs. The de-identified samples were screened by ELISA for GnRHR-AAbs using a synthetic 28-mer peptide (LifeTein, Somerset, NJ) from the ECL2 of human GnRHR as coating antigen. Optical density (OD) values were read at 405 nm at 60 minutes. Covariates evaluated include age, body mass index (BMI), anti-Mullerian hormone (AMH) level, antral follicle count (AFC), IVF protocol (GnRH antagonist or agonist), estradiol level, and hormonal down-regulation during IVF preparation. Statistical analyses were performed using generalized estimating equations within a generalized linear modeling framework to account for the repeated measures within each patient. AAb levels did not differ significantly by patient age (range 25-40 years; p=0.85) or BMI (range 19.3-41.8; p=0.89). After taking into account inter-patient variation and inter-time point variation, there was an association between GnRHR-AAb levels and estradiol levels (higher AAb level with lower estradiol; p<0.0001). Also, the GnRH antagonist protocol was associated with lower AAb levels (p=0.019). Most inter-patient variation could not be attributed to any of the patient characteristics. However, after taking into account inter-time point variation only, inter-patient variation in AAb levels was partially attributable to an interaction between BMI and AMH (p<0.0001) but not to age (p=0.70) or AFC (p=0.97). There was significant intra-patient variation in AAb levels at both baseline (time point 0, not on meds, p=0.042) and during the down-regulation phase (time point 1, p<0.0001) that could not be attributed to any of the measured covariates. Our initial investigations have shown that most patients in our study with PCOS have activating AAbs to GnRHR compared to ovulatory controls. Our pilot study has demonstrated that there is inter-patient variation of GnRHR-AAb levels on no medications and while in the down-regulation phase in preparation for IVF. Additionally, the fluctuation in AAb levels with estradiol levels in GnRH antagonist protocol would suggest an association between the two clinical time points and potentially a relationship between estradiol and the GnRHR-AAbs.
Read full abstract