Anti-Müllerian Hormone (AMH) is a 140 kDa homodimeric glycoprotein consisting of two identical subunits linked by disulphide bonds and is synthesised by the testes and ovaries. Its clinical applications are prediction of ovarian response and gonadotropin dose selection upon in vitro fertilization. In males, AMH is used to investigate sexual developmental disorders and gonadal function. AMH is commonly assayed by enzyme-linked immunosorbent assay or automated immunoassay formats that show variation between methods. This review applies fundamental chemical pathology concepts to explain the observed analytical variation of AMH measurement. We examine the lack of standardisation between AMH assays, the impact of antibody design on variable measurements, consider the analytical detection of AMH isoforms, review analytical interference in AMH measurement, and briefly assess systematic bias between AMH assays. The improved attempt at standardising AMH measurement by the recent approval of a WHO Reference Reagent offers promise for harmonising immunoassay results and establishing consensus medical cut-off points for AMH in disease. Standardisation, however, will need to redress the issue of poor commutability of standard reference material and further assign a standard reference procedure to quantify AMH standard reference material. The improvement of the analytical phase of AMH testing will support harmonised method development and patient care.
Müllerian Inhibiting Substance (MIS), also known as Anti-Müllerian Hormone (AMH), is a dimeric glycoprotein that is a member of the transforming growth factor-beta superfamily. AMH is synthesised by Sertoli and granulosa cells of the testes and ovaries, respectively. AMH regulates antral follicles in the human ovary before final selection and is the gatekeeper of follicular oestrogen production and selection of the dominant follicle that will undergo ovulation. In the testes, AMH production triggers the regression of Müellerian ducts in the male foetus and is also involved in testicular development and function.
The clinical applications of AMH are ovarian reserve testing, prediction of ovarian response after controlled ovarian stimulation, prediction of the menopause onset, monitoring of ovarian effects of medication and surgical procedures, and evaluating the risk of a variety of ovarian disorders, for example, polycystic ovary disease and primary ovarian failure. It is also used in the investigation of male sexual developmental disorders and gonadal function.
Clinical applications require analysis along an analytical range that can accommodate accurate and reliable measurement at high and low levels of the reference interval to enable accurate diagnosis and disease monitoring. Currently, multiple immunoassays are available to quantify AMH and variability in measuring AMH is commonly reported.
An electronic literature search was performed for the present review using PubMed and Google Scholar databases. The search terms used were: “AMH”, “MIS”, “Measurement”, “Method comparison”, “Method evaluation”, “Analytical interference”. The inclusion criteria were studies related to humans and studies which investigated AMH measurement. The exclusion criteria were articles for which full text was not available, not written in English, were grey literature, or original articles published before the last 20 years. Core textbooks in chemical pathology prescribed by the College of Medicine of South Africa were also consulted. Studies retrieved from initial searches were supplemented with additional references that were identified by manual search among the cited references.
This review will examine issues that account for variation in serum AMH measurement, emphasising the analytical phase of the total testing laboratory process. Basic concepts in chemical pathology that underly measurement variation of AMH will also be explained to provide non-laboratory trained readers with essential laboratory knowledge to rationalise the observed variations in AMH measurement. The critical issues we will explore are standardisation of AMH assays, antibody design and detection employed in AMH measurement, analytical interference, limits of detection for manual and automated AMH testing, and systematic error between AMH assays.
The measurement of AMH immunoassays has evolved over the last two decades. Assay formulations have utilized various calibrators and antibody pairs for quantification of AMH. Consequently, the bias, imprecision, and limit of detection (LoD) demonstrate variability. Newer assays utilize automated formats with improved analytical sensitivity. Table 1 summarizes key analytical characteristics of current assays commonly utilized for AMH measurement.
Analytical characteristics of immunoassays currently used to measure AMH.
CV, coefficient of variation; ELISA, enzyme-linked immunosorbent assay; HRP, horseradish peroxidase; rhAMH, recombinant human AMH; SHRP, streptavidin horseradishperoxidase; TMB, tetramethylbenzidine.
Immunoassays, including enzyme-linked immunosorbent assay (ELISA) format, are calibrated by defined analyte material (calibrators) to allow quantification of analyte concentration in unknown patient samples and quality control materials. A calibration curve can be constructed for both manual and automated immunoassays using manufacturer-assigned values to calibrators. Multiple calibrators (two or more) are used to establish a dynamic range of measurement from low to high concentration levels, which span relevant medical decision points. Therefore, the standard curve defines the assay’s measuring range and links the concentration and signal output of the calibrators to the analytical signal from unknown samples or controls. Patients’ samples and control material are quantified by extrapolating the assay signal to the concentration using a calibration curve (or its straight-line equation). A concentration value is then assigned to the unknown sample. The use of a reference procedure and reference material defines the metrological principle of standardisation. It links the patient’s test result to the pure material, ensuring a line of traceability.
AMH measurement suffers from variable assay standardisation between commercial immunoassays. Manufacturers of AMH assays have used assay-defined proprietary calibrators derived from various sources with variably assigned values. This causes variation of standard curves between AMH assays and has contributed to the observed variation of AMH measurement by immunoassays. Reference material either from the International Standards by the WHO or Standard Reference Materials (SRM) certified by the National Institute of Standards and Technology (NIST) exist only for a limited number of measurands. This is because standardisation is expensive, demanding human resources and onerous to establish as it consists of multi-stage processes. The designation of appropriate higher-order methods to quantify pure standard material enables manufacturers to assign values to their calibrators. Thus patients results are directly traceable to the high order reference material.
The establishment of an international standard to calibrate AMH immunoassays will enable harmonisation of AMH testing across multiple commercial assays, standardise proficiency testing schemes, allow clinical laboratories to calibrate and control immunoassays, and support continued research and development in the measurement of AMH. The WHO Expert Committee approved the development of an international standard for AMH immunoassays in 2014. Recombinant AMH trial preparation SS-581 showed stability in lyophilised form and retained biological activity in the Müllerian duct regression assay.
Following on this study, a stable, lyophilised preparation of recombinant, CHO-derived AMH, encoded 16/190, was assessed by a collaborative international panel by multiple AMH immunoassays. Most assays detected the AMH preparation; however, unsurprisingly, considerable variation between assays was noted, ascribed to the variable method calibrators used by the individual assays. The various manufacturer calibrators are derived from native, recombinant, and non-human sources. Some assays exhibited low variability of AMH content in a comparator sample to the 16/190 preparation content estimate. The WHO assigned a mass to 16/190 based on a mean consensus amongst agreeable immunoassays.
The commutability of the AMH 16/190 material, however, was unsatisfactory. Commutability is a property of reference material that describes the equivalence of the mathematical relationships between the results of different measurement procedures for reference material and representative samples from healthy and diseased individuals. Thus, for routine diagnostic laboratories, commutability requires validation across all assays and methods that use the reference material. This ensures that patients’ results by routine measurement procedures, for example, AMH measured by automated or manual immunoassays, have equivalent values regardless of the AMH immunoassay used for the measurement. Commutability is, therefore, an essential requirement when a reference material is to be used as a common calibrator for clinical laboratory assays or in proficiency testing schemes or by commercial manufacturers as part of their internal traceability procedures to assign value to their product calibrators. Ultimately, commutability allows evaluation of the agreement amongst various measurement procedures.
