Prospective validation of anti-Müllerian hormone cutoff to determine polycystic ovarian morphology: HARMONIA study

OBJECTIVE: To verify the anti-Müllerian hormone (AMH) cutoff (3.2 ng/mL) for the determination of polycystic ovarian morphology (PCOM) as part of the latest polycystic ovary syndrome (PCOS) diagnosis in an independent prospective cohort of women of reproductive age. DESIGN: A prospective, population-based, noninterventional study. SUBJECTS: Women enrolled in the Northern Finland Birth Cohort 1986 and women born in the Northern Finland region within 1.5 years of the Northern Finland Birth Cohort 1986. EXPOSURE: Antimüllerian hormone was measured in serum samples using the Roche Elecsys AMH Plus immunoassay (Roche Diagnostics International Ltd, Rotkreuz, Switzerland). Assessment of the performance of the AMH cutoff of 3.2 ng/mL to identify PCOM status determined by transvaginal ultrasound (TVUS). MAIN OUTCOME MEASURES: The number of PCOM-positive cases (participants with PCOS phenotypes A, C, or D) and negative controls according to serum AMH or TVUS. Other measures included PCOM status within selected PCOS phenotypes and demographic/clinical factors including body mass index. RESULTS: The primary analysis population (N = 948) comprised 128 PCOM-positive cases and 820 negative controls (PCOM-negative), with a mean age of 35.2 years. The median AMH levels were numerically higher in cases (6.21 ng/mL) than in controls (2.16 ng/mL), with highest levels observed for phenotype A (9.05 ng/mL) and similar levels observed for phenotypes C and D (5.67 ng/mL and 5.83 ng/mL, respectively). Applying the AMH cutoff of 3.2 ng/mL, 118 of 128 cases and 639 of 820 controls were identified in accordance with TVUS (overall percent agreement, 79.9%). Agreement of AMH with PCOM status was also observed across all PCOM-positive PCOS phenotypes and body mass index categories. The area under the receiver operating characteristic curve for prediction of case-control status was 94.0% (95% confidence interval, 91.9%-96.2%). CONCLUSION: The Elecsys AMH Plus immunoassay cutoff of 3.2 ng/mL was successfully verified, showing high accuracy to detect PCOM. On the basis of current guideline recommendations for the detection of PCOM as part of PCOS diagnosis, these data support use of the AMH test as an easy-access diagnostic workup tool compared with TVUS. TRIAL REGISTRATION NUMBER: NCT05527353.