BackgroundUpdated 2023 international evidence-based guidelines do not recommend anti-Müllerian hormone (AMH) for diagnosing adolescent polycystic ovary syndrome (PCOS, also newly termed polyendocrine metabolic ovarian syndrome [PMOS]) due to limited specificity. Physiological multifollicular ovaries are prevalent in teenagers, and existing studies show massive heterogeneity in participant age, BMI, AMH detection platforms and cut-off thresholds without unified adolescent-specific standards. This study aimed to analyze serum AMH correlations with core adolescent PCOS phenotypes and evaluate the auxiliary diagnostic performance of AMH for adolescent PCOS.MethodsBased on a cross-sectional school-based epidemiological survey among adolescent girls aged 14–18 years, we enrolled 211 healthy girls, 158 girls with oligo-/amenorrhea, 38 girls with isolated clinical hirsutism, 51 girls with isolated biochemical hyperandrogenemia, and 39 adolescent PCOS patients for case-control analysis. Pearson's correlation coefficients were calculated to determine the correlations between AMH and various parameters. Receiver Operating Characteristic (ROC) curve analysis was performed to examine the diagnostic performance of AMH.ResultsSerum AMH concentrations were significantly higher in girls with oligo-/amenorrhea and isolated biochemical hyperandrogenemia relative to healthy controls (P < 0.05), while isolated hirsutism presented comparable AMH levels with controls (P > 0.05). Adolescent PCOS patients had the highest AMH levels among all subgroups (P < 0.05). AMH was positively correlated with all circulating androgen markers (T, FAI, A2, fT, DHEAS) but showed no correlation with age, menarcheal age, gynecological age, BMI, WHR or SHBG. Within the PCOS cohort, only serum testosterone maintained an independent positive correlation with AMH (r = 0.450, P = 0.004). ROC analysis showed an AUC of AMH for adolescent PCOS of 0.717 (95%CI: 0.625–0.809, P < 0.001). Bootstrap-validated optimal AMH cut-off value was 4.0 ng/mL, with sensitivity 51.3% (95CI: 35.4%–67.2%) and specificity 86.1% (95CI: 81.2%–90.0%). post-hoc power analysis demonstrated moderate statistical power of 0.74 for this ROC model due to limited PCOS sample size (n = 39).ConclusionsSerum AMH is closely correlated with two core diagnostic manifestations of adolescent PCOS: ovulatory dysfunction and biochemical hyperandrogenemia. Nevertheless, the low sensitivity of the 4.0 ng/mL threshold leads to nearly half missed PCOS cases if used alone for screening. Combined with inherent incorporation bias, absence of ovarian ultrasound data and consistent 2023 guideline recommendations, AMH can only serve as an adjunctive auxiliary biomarker rather than an independent diagnostic or screening tool for adolescent PCOS. Multicenter studies with standardized ovarian ultrasonography and external validation cohorts are required to establish generalizable age-specific AMH cut-offs for adolescents.