Androstenedione response to recombinant human FSH is the most valid predictor of the number of selected follicles in polycystic ovarian syndrome: (a case-control study)

BACKGROUND: We aimed to test the hypothesis that the correlation of the changes in the blood Androstenedione (A(4)) levels to the number of selected follicles during ovulation induction with low-dose recombinant human follicle stimulating hormone (rhFSH) is as strong as the correlation to changes in the blood Estradiol (E(2)) levels in polycystic ovary syndrome (PCOS). METHODS: Prospective Case-control study conducted from October 2014 to January 2016. 61 non-PCOS control (Group I) and 46 PCOS (Group II) patients treated with the chronic low-dose step up protocosl with rhFSH. A(4), E(2), progesterone blood levels and follicular growth were monitored.. Univariate and hierarchical multivariable analysis were performed for age, BMI, HOMA-IR, A(4) and E(2) (with the number of selected follicles as the dependent variable in both groups). ROC analysis was performed to define threshold values for the significant determinants of the number of selected follicles to predict cyle cancellations due to excessive ovarian response. RESULTS: The control group (Group I) was comprised of 61 cycles from a group of primary infertile non-PCOS patients, and the study group (Group II) of 46 cycles of PCOS patients. The analysis revealed that the strongest independent predictor of the total number of selected follicles in Group I was the E(2)(AUC) (B = 0.0006[0.0003-0.001]; P < 0.001); whereas for Group II, it was the A(4) (AUC) (B = 0.114[0.04-0.25]; P = 0.01). Optimum thresholds for the A(4) related parameters were defined to predict excessive response within Group II were 88.7%, 3.1 ng/mL and 5.4 ng*days for the percentage increase in A(4), the maximum A(4) value and area under the curve values for A(4), respectively. CONCLUSION: A(4) response to low-dose rhFSH in PCOS has a stronger association with the number of follicles selected than the E(2) reponse. A(4) response preceding the E(2) response is essential for progressive follicle development. Monitoring A(4) rather than E(2) may be more preemptive to define the initial ovarian response and accurate titration of the rhFSH doses. TRIAL REGISTRATION: The study was registered as a prospective case-control study in the ClinicalTrials.gov registry with the identifier NCT02329483.