A-346 Sex-specific Diagnostic Performance of Reproductive Hormones in Central Precocious Puberty in Chinese Children

Abstract

Central precocious puberty (CPP) is a common pediatric endocrine disorder requiring accurate diagnosis to prevent adult height compromise and psychosocial complications. Although 2022 CPP Expert Consensus identifies gonadotropin-releasing hormone (GnRH) stimulation tests as the gold standard, its invasiveness and limited accessibility restrict widespread use. Basal reproductive hormone measurements may provide a non-invasive adjunct for CPP evaluation. However, their diagnostic performance and optimal decision thresholds remain insufficiently defined, particularly in large multicenter populations. This study aimed to systematically investigate the diagnostic value of single and combined reproductive hormone panels for CPP diagnosis in children, and to establish specific diagnostic cutoffs. A total of 801 children with CPP (74 boys and 727 girls) and 266 healthy control children (115 boys and 151 girls) from 31 institutions in 20 provinces across China were enrolled. Ten reproductive hormones, including anti-Müllerian hormone (AMH), inhibin B , inhibin A, follicle-stimulating hormone (FSH), luteinizing hormone (serum LH, urine LH), prolactin testosterone, progesterone, and estradiol (E2), were detected using the YHLO iFlash 3000 chemiluminescence analyzer. Between-group comparisons were performed using Mann-Whitney U test. Receiver operating characteristic (ROC) curve analysis was conducted to determine optimal diagnostic cutoffs, with area under the curve (AUC), sensitivity, and specificity calculated for individual and combined biomarker panels. Data collection is ongoing, and this report presents preliminary findings based on the current dataset. Significant differences were observed between CPP and control groups for most hormones in both genders (all P < 0.01). In boys, testosterone demonstrated the most powerful discriminator (AUC=0.853, 95% CI: 0.794-0.912; cutoff 0.145 ng/mL), followed by AMH (AUC=0.821, 95% CI: 0.751-0.891; cutoff 21.025 ng/mL), and serum LH (AUC of 0.795, 95% CI: 0.728-0.861; cutoff 1.120 mIU/mL), whereas other hormones showed limited diagnostic utility. In girls, INHB showed the strongest discrimination (AUC=0.766, 95%CI: 0.723-0.809; cutoff 26.705 pg/mL), outperforming serum LH (AUC=0.730, 95%CI: 0.698-0.761; cutoff 0.205 IU/L), FSH (AUC=0.691, cutoff 2.705 IU/L), and AMH (AUC=0.597, cutoff 3.145 ng/mL), while other markers demonstrated modest or limited utility. Multi-marker combinations improved overall discrimination compared with individual biomarkers in both sexes. The triple-marker panel (AMH + inhibin B + inhibin A) achieved AUCs of 0.809 (95%CI: 0.766-0.852) in girls and 0.857 (95%CI: 0.798 -0.916) in boys, while the comprehensive model including urinary LH demonstrated the highest performance (AUC 0.877 (95%CI: 0.812-0.941) in girls and 0.883 (95%CI: 0.804-0.961) in boys). Incremental improvements were moderate as additional markers were incorporated. In this multicenter study, reproductive hormone panels demonstrated improved diagnostic discrimination compared with single biomarkers for CPP evaluation. The proposed cutoffs, may support non-invasive screening or triage prior to GnRH stimulation testing rather than replace standard diagnostic procedures. Further prospective studies are required to validate and confirm the clinical utility.