Pregnancy blood test showdown: Revvity and Roche angiogenic assays agree on risk but not on numbers

When a pregnant woman arrives at hospital with suspicious symptoms such as headache, visual disturbance or swelling, clinicians face a diagnostic dilemma that can have serious consequences for mother and baby. Blood tests measuring the angiogenic markers soluble fms-like tyrosine kinase 1 (sFlt-1) and placental growth factor (PlGF) have become powerful tools for ruling in or ruling out preeclampsia, but the growing number of commercial assays raises a practical question: do different manufacturers’ tests give the same answer? A large evaluation from Oxford now provides one of the most detailed comparisons to date, and the findings carry important implications for laboratories and patients alike.

Researchers at Oxford University Hospitals compared the PlGF and sFlt-1 assays from Revvity with those from Roche, the latter being widely used across the United Kingdom and internationally. The study, led by Tim James and colleagues including Rebecca Marr and Alexandra Cerdeira, analysed more than 600 clinical specimens obtained from routine practice, covering the full range of concentrations encountered in daily testing. The work was published in the British Journal of Biomedical Science and represents a pragmatic head-to-head evaluation conducted in a real clinical setting rather than a controlled laboratory exercise.

The precision of the Revvity assays proved reassuring. Intermediate imprecision, assessed using quality control material during the evaluation, ranged from 3.2 to 9.0 percent coefficient of variation for PlGF and from 1.1 to 9.0 percent for sFlt-1, figures that fall comfortably within accepted analytical performance criteria for immunoassays. Even more convincing, the researchers constructed precision profiles from more than 500 clinical specimens measured in duplicate, showing median coefficients of variation of 1.9 percent for PlGF and 1.1 percent for sFlt-1 across the clinically relevant concentration ranges. As with most immunoassays, precision was somewhat poorer at low concentrations, but the overall analytical performance of the Revvity platform was solid.

Agreement between the two manufacturers’ assays, however, told a more complicated story. Using Bland-Altman analysis, the researchers found that the Revvity PlGF assay gave systematically higher results than the Roche assay, with a mean bias of approximately 20 percent across the concentration range. For sFlt-1, the direction of bias reversed: the Revvity assay gave systematically lower results than the Roche assay, with a mean bias of roughly minus 20 percent. These proportional biases were consistent across the measuring range, meaning that the discrepancy between the two platforms was not confined to extreme concentrations but persisted throughout routine clinical testing.

Passing-Bablok regression analysis confirmed these findings and revealed that the relationship between the two platforms was not perfectly proportional. For PlGF, the regression equation showed a slope greater than one with a negative intercept, indicating that the proportional bias increased at higher concentrations. For sFlt-1, the slope was less than one with a positive intercept, meaning the negative bias became more pronounced as concentrations rose. The researchers also found that the calculated sFlt-1-to-PlGF ratio, a parameter increasingly used in clinical decision-making, differed systematically between the two platforms, with the ratio derived from Revvity results being substantially lower than that derived from Roche results.

Despite these analytical differences, the clinical concordance between the two platforms proved remarkably good. When specimens were classified according to the National Institute for Health and Care Excellence (NICE) thresholds for ruling preeclampsia in or out, the two platforms agreed on the clinical category in the overwhelming majority of cases. Using the NICE rule-out threshold of a sFlt-1-to-PlGF ratio of 38 or below, the two platforms showed almost perfect agreement, with only a small proportion of specimens falling into discordant categories. The researchers calculated that the clinical sensitivity and specificity for ruling out preeclampsia were essentially identical between the two manufacturers when appropriate platform-specific thresholds were applied.

The explanation for the analytical bias lies in the different calibration systems and antibody specificities used by the two manufacturers. Roche calibrators are traceable to a specific reference material, while Revvity uses its own internal reference preparation, and the two companies employ different monoclonal antibody pairs that may recognise different epitopes on the target molecules. PlGF exists in multiple molecular forms, and sFlt-1 is a complex glycoprotein with considerable structural heterogeneity, so antibodies raised against different epitopes can yield systematically different quantitative results even when both are analytically valid. The researchers also noted that the freeze-thaw cycle used in specimen preparation may have contributed to some of the observed differences, although a subgroup analysis of fresh versus frozen specimens suggested this was not a major factor.

The practical message for laboratories is that assay-specific decision thresholds are essential. The researchers demonstrated that when the NICE rule-out threshold derived from Roche-based studies was applied directly to Revvity results, the clinical performance deteriorated, with a small but measurable loss of sensitivity. However, when platform-specific thresholds were calculated using the regression equations from this study, the Revvity assays achieved clinical performance equivalent to the Roche platform. This finding underscores a broader principle in laboratory medicine: quantitative results from different immunoassay platforms are not interchangeable, and clinical decision limits must be validated for each specific assay system before implementation.

For patients, the reassuring news is that both platforms deliver clinically reliable answers when used correctly. The angiogenic markers have established themselves as among the most powerful predictors in obstetric medicine, and the Oxford evaluation confirms that the Revvity assays, despite giving systematically different numerical results from the Roche platform, provide equivalent clinical discrimination when appropriate thresholds are used. The researchers emphasise that their findings should not be interpreted as a criticism of either manufacturer but rather as a demonstration of the challenges inherent in measuring complex biological molecules and the importance of rigorous validation before clinical implementation.

The Oxford team acknowledges some limitations of their work. The evaluation was conducted at a single centre, and the specimens analysed were obtained from a specific clinical population, so the findings may not generalise to all patient groups. The researchers also note that they did not assess the clinical performance of the assays against confirmed diagnostic outcomes, relying instead on established NICE thresholds as the reference standard. Nevertheless, the study represents one of the most comprehensive comparisons of commercial angiogenic assays conducted to date, and its findings should inform laboratory validation protocols and clinical implementation strategies for these increasingly important obstetric biomarkers. As the use of angiogenic markers continues to expand in routine practice, the message from Oxford is clear: rigorous platform-specific validation is not optional but essential for delivering safe and effective clinical care.

Subject of Research: Laboratory evaluation of angiogenic biomarker assays for preeclampsia

Article Title: Evaluation of analytical performance of the Revvity sFlt-1 and PlGF methods for the assessment of preeclampsia

Article References: Marr, R., Cerdeira, A. S., Smith, I., Vatish, M., & James, T. (2026). Evaluation of analytical performance of the Revvity sFlt-1 and PlGF methods for the assessment of preeclampsia. British Journal of Biomedical Science, 83, Article 15950. https://doi.org/10.3389/bjbs.2026.15950

Image Credits: AI Generated

DOI: 10.3389/bjbs.2026.15950

Keywords: preeclampsia, angiogenic markers, sFlt-1, PlGF, immunoassay, laboratory medicine, pregnancy, biomarkers, Evaluation, analytical, performance, Revvity