OR21-3 Anti-Mullerian Hormone (AMH) in Obese Adolescent Girls with Polycystic Ovary Syndrome (PCOS): Cross-Sectional and Treatment-Associated Longitudinal Changes

Program: Abstracts - Orals, Poster Previews, and Posters
Session: OR21-Markers and Consequences of Ovulatory (dys)Function
Saturday, April 2, 2016: 11:45 AM-1:15 PM
Presentation Start Time: 12:15 PM
Room 162 (BCEC)
Joon Young Kim*1, Hala Mounir Tfayli2, Sara F. Michaliszyn3, Sojung Lee4, Alexis Nasr4 and Silva A Arslanian5
1University of Pittsburgh Medical Center, Pittsburgh, PA, 2American University of Beirut Medical Center, Beirut, Lebanon, 3Youngstown State University, 4University of Pittsburgh Medical Center, 5Children's Hosp of Pittsburgh of UPMC, Pittsburgh, PA
AMH, a possible surrogate for ovarian antral follicle number, is proposed as a biomarker of PCOS. A variety of cut-off values of AMH have been suggested in adult women with differing sensitivity and specificity. This study investigated: 1) AMH levels cross-sectionally in obese adolescents with PCOS in comparison to obese non-PCOS girls, 2) treatment-associated longitudinal changes with either drospirenone/EE (DEE) or rosiglitazone (Rosi) in adolescents with PCOS, 3) the relationship of AMH to adiposity measures and insulin sensitivity, and 4) the optimal value to determine PCOS status in adolescents. AMH levels were measured in 46 obese girls with PCOS (age 14.9 ± 0.2 yrs; BMI 37.7 ± 1.1 kg/m2) and 43 obese non-PCOS girls (age 14.4 ± 0.2 yrs; BMI 33.1 ± 1.1 kg/m2). The change in AMH was examined in 37 PCOS girls randomly assigned to DEE (n=20) or Rosi (n=17) in a double-blinded 6 month trial (1). Body composition (DEXA), abdominal adiposity (MRI or CT at L4-5), in vivo insulin sensitivity (hyperinsulinemic-euglycemic clamp), and total and free testosterone were evaluated. Logistic regression analysis and receiver operating characteristic (ROC) curve were used to test the utility of serum AMH levels for the diagnosis of PCOS. AMH levels were higher in obese PCOS vs. non-PCOS girls (8.3 ± 0.6 vs. 4.3 ± 0.4 ng/ml, p<0.0001), with comparable age. AMH levels decreased with DEE treatment (pre 8.7 ± 0.7 vs. post 7.0 ± 0.8 ng/ml, p=0.001), but not with Rosi. AMH concentrations correlated positively with age (r=0.48), total (r=0.58) and free (r=0.47) testosterone, abdominal visceral (r=0.31), subcutaneous (r=0.33) and total (r=0.33) adipose tissue (all p<0.05), with no correlation to in vivo insulin sensitivity. Logistic regression analysis showed that AMH level was a predictor of PCOS diagnosis independent of age and BMI (odds ratio 1.47, p<0.0001). The optimal AMH value for the diagnosis of PCOS was 6.26 ng/ml (sensitivity 65% and specificity 81%) with the ROC area under the curve of 0.779 (95% CI 0.684-0.873, p<0.0001). In conclusion, AMH concentrations are higher in obese adolescent girls with PCOS and correlate with hyperandrogenemia and abdominal adiposity but not insulin sensitivity. AMH levels decreased with DEE treatment but not with Rosi. AMH may be a useful biomarker for the diagnosis and therapeutic follow up of PCOS in obese adolescent girls.

(1) Tfayli et al., J Clin Endocrinol Metab. 2011 May;96(5):1311-9.

Nothing to Disclose: JYK, HMT, SFM, SL, AN, SAA

*Please take note of The Endocrine Society's News Embargo Policy at https://www.endocrine.org/news-room/endo-annual-meeting/pr-resources-for-endo

Sources of Research Support: K24 HD01357 to SA, Richard L. Day Endowed Chair to SA, HT and JYK, Children’s Hospital of Pittsburgh PCTRC UL1TR000005 to CTSA and RR024153 to GCRC.