DOI: 10.1177/20420188261494047 ISSN: 2042-0188

Serum testosterone and uric acid: Key biomarkers for predicting response to drospirenone/ethinyl estradiol tablets (II) in adolescents polycystic ovary syndrome

Xuyong Chen, Yuanyuan Teng, Yue He, Jiaqi Gao, Mingyang Zhao, Hao Yan, Mengyu Li, Min Wang

Background

Adolescent polycystic ovary syndrome (PCOS) poses diagnostic and management challenges. Although combined oral contraceptives (COCs) can improve symptoms, treatment response varies.

Objectives

To evaluate the efficacy of drospirenone/ethinyl estradiol tablets (II) (DRSP/EE (II)) and explore predictors of therapeutic response.

Design

Single-center retrospective study.

Methods

Adolescents with PCOS and healthy controls were retrospectively included. Among patients treated with DRSP/EE (II), clinical, biochemical, and sex-hormone indices were assessed at baseline and at 3 and 6 months, including total testosterone (TT), sex hormone-binding globulin (SHBG), free androgen index (FAI), uric acid (UA), and lipid profile. Binary logistic regression was used to identify factors associated with treatment response, and receiver operating characteristic (ROC) curves were used to assess discriminative performance.

Results

The PCOS group had higher androgen levels and lower SHBG than controls. TT and FAI decreased significantly after 3 and 6 months of DRSP/EE (II), while SHBG increased. Triglycerides, total cholesterol, LDL-C, and HDL-C increased. Responders had higher baseline TT (2.03±0.59 vs. 1.55±0.46 nmol/L; FDR P=0.009) and lower UA (314.43±61.33 vs. 372.16±77.16 μmol/L; FDR P=0.023) than non-responders. Multivariable logistic regression showed that TT (OR=6.893; P=0.004) and UA (OR=0.984; P=0.002) were independently associated with treatment response. ROC analyses showed AUCs of 0.716 for TT and 0.713 for UA, while the combined model yielded an AUC of 0.815, with 87.8% sensitivity and 75.0% specificity.

Conclusion

DRSP/EE (II) effectively improves hyperandrogenism in adolescent PCOS but may worsen lipid profiles. Baseline TT and UA are promising predictors of clinical response and may facilitate individualized management.