MON-040 11-Oxygenated C19 Steroids in Polycystic Ovarian Syndrome

BACKGROUND: Polycystic ovarian syndrome (PCOS), an endocrine and reproductive disorder consisting of hyperandrogenism, menstrual dysfunction and ovarian changes, affects 6–20% of reproductive aged women worldwide. While hyperandrogenemia is traditionally determined by evidence of elevated testosterone (T), this hormone can be difficult to accurately measure in women with relatively lower circulating levels compared to men. Recent studies have suggested that four adrenal androgens known as 11-oxygenated C19 steroids (11OxyAs), specifically 11-ketotestosterone (11KT), may be good alternative markers for hyperandrogenism in PCOS. Using a multiethnic population seeking evaluation for PCOS symptomatology, we sought (1) to investigate the utility of 11OxyAs to differentiate women with and without NIH PCOS relative to classical androgens such as T, androstenedione (A4) and DHEAS levels, and (2) to evaluate the relationship of 11OxyAs to clinical findings of androgen excess. Methods: Using the University of California, San Francisco PCOS Tissue Bank, serum samples from 131 women seen for a PCOS evaluation were selected sequentially and identified as PCOS or non-PCOS (controls) based on meeting NIH criteria at the time of evaluation. In addition to obtaining gonadotropin and metabolic profiles, classical androgens and 11OxyAs were measured using mass spectrometry. The relationship of these androgens to modified Ferriman-Gallwey (mFG) scores and ovarian morphology were also assessed. Results: Out of 131 women selected, 83 met NIH PCOS criteria at the time of evaluation and 48 did not (controls). Age and BMI did not differ among the two groups. As expected, total T, A4 and LH were all significantly higher in NIH PCOS. A trend towards higher HOMA-IR levels was also seen in NIH PCOS, but this did not reach statistical significance (3±3.9 mg/dL vs. 1.9±1.7 mg/dL, p = 0.12). No difference was seen in all four 11OxyAs between NIH PCOS and controls. Unlike previous studies, we also did not find mean 11KT levels to exceed that of T in both controls (T 393±143 pg/mL vs. 11KT 389±206 pg/mL) and PCOS (T 530±245 pg/mL vs. 11KT 388±201 pg/mL). In addition, no relationship was seen between HOMA-IR and 11β-hydroxyandrostenedione (11OHA4) or 11-ketoandrostenedione (11KA4) levels. Within PCOS, DHEAS and A4 were noted to have a weak but inverse relationship to BMI (r(2) 0.05 p = 0.05; r(2) 0.08 p = 0.007), whereas no correlation was seen between any of the four 11OxyAs or T and BMI. Lastly, 11OxyAs, T, and A4 levels did not predict mFG scores or polycystic ovarian morphology. Conclusions: 11OxyAs levels were not statistically higher among women with NIH PCOS compared to at risk women who did not meet NIH criteria. There was no significant relationship between these androgens and mFG scores or ovarian morphology. Further studies are necessary to show the utility of 11OxyAs levels as a marker for hyperandrogenism or metabolic risk.