P-590 Discriminatory value of steroid hormones on polycystic ovary syndrome in women with infertility and obesity

Abstract

Abstract Study question Which steroid hormone (ratio) has the highest discriminatory value in the biochemical diagnosis of polycystic ovary syndrome (PCOS) in women with infertility and obesity? Summary answer Androstenedione (A4) is a potentially useful discriminatory biomarker for PCOS in women with obesity, whereas Testosterone (T)/Dihydrotestosterone (DHT) and 11-oxygenated androgens are not. What is known already The best biomarker to assess biochemical hyperandrogenism in PCOS is still under discussion. Serum total or free T, followed by Free Androgen Index (FAI), and Dehydroepiandrosterone Sulfate (DHEA-S) are the most widely used biomarkers. T/DHT ratio has been proposed as a useful biomarker for PCOS. Moreover, a recent study demonstrated 11-oxygenated androgens represent the majority of circulating androgens in PCOS. Whether this is also relevant in women with obesity is unclear. We aimed to assess the discriminatory value of different steroid hormones and their ratios for PCOS as well as adverse metabolic phenotype within PCOS in women with obesity. Study design, size, duration We used the baseline data and samples originating from an RCT that examined whether a six-month lifestyle intervention prior infertility treatment in women with obesity improved live birth rate, compared to prompt infertility treatment. In total of 577 women with obesity were randomized between 2009 and 2012. Women were diagnosed with PCOS according to the Rotterdam criteria. Metabolic syndrome was diagnosed based on the revised criteria of the National Cholesterol Education Program. Participants/materials, setting, methods Women with PCOS comprised the study group (N = 132). Ovulatory women with idiopathic, tubal or male factor infertility were the control group (N = 83). Women were excluded when blood sampling was performed in the luteal phase or when baseline samples were unavailable. The steroid hormones were measured using LC-MS/MS. 11β-hydroxyandrostenedione (11OHA4) and 11-ketotestosterone (11KT) were analyzed using LC-MS/MS in a separate method. The discriminatory value was based on receiver operator characteristic (ROC) curves with adjustment for age. Main results and the role of chance The mean age was 28.1 years in the PCOS group and 30.9 years in the control group (p < 0.001). BMI was not significantly different (36.0±3.2 versus 35.7±3.5, p = 0.51). Among the tested steroid hormones (ratios), SHBG, A4, T, DHT, 11KT, FAI, T/DHT ratio, and T/A4 ratio showed statistically significant differences between the PCOS and the control group. DHEA, DHEA-S, 11OHA4 did not differ between the groups. The crude AUCs of FAI, T, A4, T/DHT ratio, SHBG, 11KT, T/A4 ratio, and DHT in ROC analysis for representing PCOS were 0.85, 0.84, 0.81, 0.70, 0.62, 0.59, 0.58, and 0.57 respectively. Among the markers with crude AUCs≥0.70, the incremental value of the marker A4 (AUC difference: 0.14, 95%CI: 0.07 to 0.21), T (0.18, 95%CI: 0.10 to 0.25), and FAI (0.18, 95%CI: 0.10 to 0.25) in addition to age showed significant improvement, but T/DHT ratio (0.02, 95%CI: –0.01 to 0.04) did not. However, the combination of A4 together with T or FAI did improve discriminatory value of individual T or FAI (AUC difference: –0.001, 95%CI: –0.004 to 0.003 and 0.007, –0.007 to 0.02, respectively). None of the measured steroid hormones (ratios) have promising discriminatory value for adverse metabolic phenotype within PCOS in women with obesity. Limitations, reasons for caution This is a post-hoc analysis and no power calculation was performed. There was a significant difference in age between women with and without PCOS, and residual confounding might exist. All PCOS patients in this analysis were infertile and obese, and it is conceivable that milder PCOS phenotypes are not represented. Wider implications of the findings T/DHT ratio and 11-oxygenated androgens might not serve as a promising discriminatory biomarker for PCOS in women with obesity, nor for adverse metabolic phenotype within PCOS in these women. A4 is a potential biomarker for PCOS diagnosis in women with infertility and obesity, however, its utility potency needs further validation. Trial registration number NTR 1530