Comparative evaluation of three new commercial immunoassays for anti-Müllerian hormone measurement.

STUDY QUESTION: How do the three new anti-Müllerian hormone (AMH) assay methods, manufactured by Beckman Coulter, Roche and Ansh Labs compare with each other and with the Gen II assay?

SUMMARY ANSWER: The three new AMH assays are well correlated among themselves and with the Gen II assay, although differences in calibration do exist.

WHAT IS KNOWN ALREADY: The Gen II assay has been the mainstay method for AMH measurement in the past few years. Recently, a few new AMH measurement methods have come to the market.

STUDY DESIGN, SIZE, DURATION: This was a prospective assay evaluation performed on 178 human serum samples.

PARTICIPANTS/MATERIALS, SETTING, METHODS: AMH concentration was measured in residual serum samples donated by female patients in a reproductive medicine centre. The three new assay methods were tested in parallel and the numerical values obtained were compared among themselves and with those obtained by the Gen II assay. The assay stability upon different sample storage conditions, intra-assay and inter-assay precision, linearity and dilution recovery, and diagnostic performance for polycystic ovary syndrome (PCOS) of the three new AMH assay methods were also compared.

MAIN RESULTS AND THE ROLE OF CHANCE: AMH values measured by the Gen II kit and the three new assay methods have good correlations (R > 0.9 for all pairwise correlations). Values measured by the Ansh Labs assay were significantly higher, whereas those by the Roche assay were significantly lower, than those from the Gen II and Beckman-Coulter automated assays (P < 0.05). AMH values were significantly different when measured on the fresh and frozen-thawed serum sample (at -20o C and -80o C) for all three new methods (P < 0.05), but the magnitude of difference was very small with the Beckman-Coulter automated assay and Roche assay. The intra-assay coefficients of variation (CVs) were 0.7-2.2%, 0.5-1.4%, and 1.4-5.4% for the Beckman-Counter automated, Roche and Ansh Labs assays, respectively. Their inter-assay CVs were 0.9-2.5%, 0.7-1.9%, and 6.2-13.5%, respectively. All three new assay methods showed acceptable linearity, and provided excellent discrimination of PCOS from controls.

LIMITATIONS, REASONS FOR CAUTION: The precision and dilution linearity experiments involved a small sample size, although these were not the primary outcome measures and have been properly evaluated in previous publications. The study was not designed or powered for determining diagnostic cut-off values.

WIDER IMPLICATIONS OF THE FINDINGS: The results demonstrate that the three new AMH assay methods are all valid methods to be adopted in the field of reproduction and are a basis for further work on their clinical application.

STUDY FUNDING/COMPETING INTERESTS: The authors have no competing interest to declare. The execution of this study was funded by the Department of Obstetrics and Gynaecology, The University of Hong Kong.