Validation of an Anti-Müllerian Hormone Cutoff for Polycystic Ovarian Morphology in the Diagnosis of Polycystic Ovary Syndrome in the HARMONIA Study: Protocol for a Prospective, Noninterventional Study.

BACKGROUND: Polycystic ovary syndrome (PCOS) is one of the most common endocrine disorders in women and is diagnosed using the Rotterdam criteria, including diagnosis of polycystic ovarian morphology (PCOM) by transvaginal ultrasound (TVUS). Due to high cost, availability, and the impact of the operator and ultrasound equipment on the reliability of the antral follicle count (AFC) by TVUS, an unmet need exists for a diagnostic test to determine PCOM without TVUS. A strong positive correlation between elevated anti-Müllerian hormone (AMH) levels and AFCs has been demonstrated in women with PCOS. In addition, recent updates to the international evidence-based PCOS guidelines state that serum AMH can be used as an alternative to TVUS-determined AFC, in the diagnosis of PCOM. The retrospective APHRODITE study derived and validated an AMH cutoff of 3.2 ng/mL for the Elecsys AMH Plus or Elecsys AMH assays (Roche) to diagnose PCOM in patients with PCOS. OBJECTIVE: This study aims to further validate, in an independent prospective cohort, the AMH cutoff (3.2 ng/mL) for PCOM determination, which was previously derived and validated in the APHRODITE study. METHODS: This large, prospective, multicenter, population-based, noninterventional study will evaluate the previously established AMH cutoff for the determination of PCOM during the diagnosis of PCOS using the Elecsys AMH Plus immunoassay in an independent population. Participants were women born between July 1985 and December 1987 in Northern Finland; the study partially links to the Northern Finland Birth Cohort 1986. We assessed the enrolled women, determined with the 2023 PCOS Guidelines, for current PCOS status and divided them by phenotype if positive. Each participant had 1 study visit to collect serum samples, record clinical data, and undergo a gynecological examination including TVUS. All data were collected by highly trained midwives or trained gynecologists. Sensitivity, specificity, and agreement measures were used to validate the previously determined cutoff in the whole population and in subpopulations based on phenotype and relevant demographic or clinical factors. The minimum target sample size was approximately 1800 women, including approximately 10% with PCOS. RESULTS: At the time of manuscript submission, participant recruitment had concluded, and 1803 women were enrolled into the study. Data collection is complete and biostatistical analysis is planned for 2023. CONCLUSIONS: To limit variability, there were few TVUS operators and only 2 TVUS machines of the same type. Additionally, all women who were taking oral contraceptives were excluded from the primary analysis population. Selection bias was limited as this was a population-based study and participants were not seeking treatment for PCOS symptoms. Validating the AMH cutoff in a large, population-based study will provide further evidence on the utility of the Elecsys AMH Plus or Elecsys AMH assays in PCOM diagnosis as an alternative to TVUS. Measuring AMH for PCOM diagnosis could reduce delayed or missed diagnoses due to operator-dependent TVUS examinations. TRIAL REGISTRATION: ClinicalTrials.gov NCT05527353; http://tinyurl.com/2f3ffbdz. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/48854.

Antimüllerian hormone to determine polycystic ovarian morphology.

OBJECTIVE: To determine a cutoff for the Elecsys AMH Plus immunoassay (Roche Diagnostics International Ltd, Rotkreuz, Switzerland) to identify polycystic ovarian morphology (PCOM), a polycystic ovary syndrome (PCOS) criterion. DESIGN: The AMH Protein in Humans for polycystic ovaRian mOrphology DIagnostic TEsting (APHRODITE) study was a retrospective, multicenter, case-control study. The serum antimüllerian hormone (AMH) level was measured using the Elecsys AMH Plus immunoassay. The antral follicle count was determined using transvaginal ultrasound. An AMH cutoff was derived and validated in separate cohorts with cases of PCOS with full phenotype A (oligo/anovulation, hyperandrogenism, and PCOM) versus that with controls. Exploratory analyses of age and PCOS phenotype were performed. SETTING: Not applicable. PATIENT(S): Polycystic ovary syndrome-positive (PCOS A-D per the Rotterdam criteria) and PCOS-negative women aged 25-45 years. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): A validated cutoff for AMH using the Elecsys AMH Plus assay for PCOM. RESULT(S): In the validation cohort (455 cases and 500 controls), an AMH cutoff of 3.2 ng/mL (23 pmol/L) resulted in a sensitivity of 88.6% (95% confidence interval [CI] 85.3-91.3) and specificity of 84.6% (95% CI 81.1-87.7) for PCOM diagnosis as well as an area under the receiver-operator characteristic curve of 93.6% (95% CI 92.2-95.1). In women aged 25-35 years, the sensitivity and specificity for the cutoff were 88.5% and 80.3%, respectively, versus 77.8% and 90.1%, respectively, in women aged 36-45 years. The results were consistent across PCOS phenotypes A-D. CONCLUSION(S): The Elecsys AMH Plus immunoassay, with a cutoff of 3.2 ng/mL (23 pmol/L), is a robust method for identifying PCOM to aid in PCOS diagnosis.

Gestational Age-Specific Reference Ranges for the sFlt-1/PlGF Immunoassay Ratio in Twin Pregnancies.

OBJECTIVE: Establish reference ranges for the Elecsys® soluble fms-like tyrosine kinase-1 (sFlt-1)/placental growth factor (PlGF) immunoassay ratio in twin pregnancies. METHODS: Data analyzed were from 3 prospective studies: Prediction of Short-Term Outcome in Pregnant Women with Suspected Preeclampsia (PE) (PROGNOSIS), Study of Early-onset PE in Spain (STEPS), and a multicenter case-control study. Median, 5th, and 95th percentiles for sFlt-1, PlGF, and the sFlt-1/PlGF ratios were determined for normal twin pregnancies for 7 gestational windows and compared with the previous data for singleton pregnancies. RESULTS: The reference range analysis included 269 women with normal twin pregnancies. Before 29 weeks' gestation, median, 5th, and 95th percentiles for sFlt-1/PlGF ratios did not differ between twin and singleton pregnancies. From 29 weeks' gestation to delivery, median, 5th, and 95th percentiles for sFlt-1/PlGF ratios were substantially higher in twin versus singleton pregnancies. sFlt-1 values were higher in women with twin pregnancies across all gestational windows. PlGF values were similar or higher in twin versus singleton pregnancies; PlGF concentrations increased from 10 weeks + 0 days to 28 weeks + 6 days' gestation. CONCLUSIONS: Reference ranges for the sFlt-1/PlGF ratio are similar in women with twin and singleton pregnancies until 29 weeks' gestation but appear higher in twin pregnancies thereafter.

Elecsys® and Kryptor immunoassays for the measurement of sFlt-1 and PlGF to aid preeclampsia diagnosis: are they comparable?

Background For pregnant women with suspected preeclampsia, the soluble fms-like tyrosine-kinase 1 (sFlt-1)/placental growth factor (PlGF) ratio is a biomarker to aid diagnosis. We performed method comparisons between Elecsys® and Kryptor sFlt-1 and PlGF immunoassays and assessed the diagnostic performance for preeclampsia. Methods Serum samples from a case-control study involving 113 pregnant women with preeclampsia/elevated liver enzymes and low platelet count (HELLP) and 270 controls were analyzed. sFlt-1 and PlGF were measured using Roche Elecsys® and BRAHMS Kryptor sFlt-1/PlGF immunoassays. The sFlt-1/PlGF ratios were calculated, and Passing-Bablok regression/Bland-Altman plots were performed. Gestation-specific cut-offs, ≤33 and ≥85/≥110, were assessed. Results Mean (±2 standard deviation [SD]) differences between the Elecsys® and Kryptor values were: sFlt-1, 173.13 pg/mL (6237.66, -5891.40); PlGF, -102.71 pg/mL (186.06, -391.48); and sFlt-1/PlGF, 151.74 (1085.11, -781.63). The Elecsys® and Kryptor immunoassays showed high correlation: Pearson's correlation coefficients were 0.913 (sFlt-1) and 0.945 (PlGF). Slopes were 1.06 (sFlt-1) and 0.79 (PlGF), resulting in ~20% lower values for Kryptor PlGF. Sensitivities and specificities using the sFlt-1/PlGF ≥85 cut-off for early-onset preeclampsia (20 + 0 to 33 + 6 weeks) were 88.1%/100.0% (Elecsys®) and 90.5%/96.2% (Kryptor), respectively, and using the ≥110 cut-off for late-onset preeclampsia (≥34 + 0 weeks) were 51.3%/96.5% (Elecsys®) and 78.9%/90.1% (Kryptor), respectively. Using Elecsys® and Kryptor sFlt-1/PlGF, 0% and 3.8% of women, respectively, were falsely ruled-in for early-onset, and 3.5% and 9.9%, respectively, for late-onset preeclampsia. Conclusions Despite high correlation between the Elecsys® and Kryptor immunoassays, we observed significant differences between sFlt-1/PlGF and PlGF results. Therefore, sFlt-1/PlGF cut-offs validated for Elecsys® immunoassays are not transferable to Kryptor immunoassays.