Inter-manufacturer comparison of automated immunoassays for the measurement of soluble FMS-like tyrosine kinase-1 and placental growth factor

Objective: To assess inter-manufacturer automated immunoassays for soluble FMS-like tyrosine kinase 1 (sFlt-1) to placental growth factor (PlGF).

Methods: sFlt-1 and PlGF levels were measured using the AutoDelfia PlGF1-2-3 (PerkinElmer Inc. Turku, Finland), BRAHMS Kryptor sFlt-1, PlGF plus and PlGF-2 (BRAHMS ThermoFisher, Germany) and Cobas e411 Elecsys® sFlt-1 and PLGF (Roche Diagnostics GmbH, Mannheim, Germany) in 965 asymptomatic pregnancies between 20 and 39 weeks of gestation and in in-vitro samples with predefined levels of glycosylated PlGF isomers (1, 2 and 3), sFlt-1 in human male serum. Percentage PlGF isoform recovery and cross-reactivities were determined. Paired Bland-Altman and Passing-Bablok analyses were performed to determine bias, precision and accuracy. Inter-manufacturer sFlt-1:PlGF ratio were compared.

Results: PlGF-1 isomer recovery ranged from 36 to 39% for Elecsys® to 52-60% for PlGF plus and PlGF-1-2-3 assays. PlGF-2 and PlGF-3 isoform cross-reactivity was assay dependent, ranging from 10 to 21% and 16-36% respectively. BRAHMS PlGF-2 assay had high cross-reactivity to PlGF-1 (37-41%) and PlGF-3 isomers (48-65%). Elecsys® recovery of sFlt-1 was 13% vs 6% for BRAHMS. Passing-Bablok indicated significant proportional and systematic differences between all paired PlGF assay comparisons. PlGF Bland-Altman percentage biases ranged from 12 to 37% for PlGF and 18% for sFlt-1. A linear relationship existed between log transformed sFlt-1:PlGF ratios. The clinical equivalent of the BRAHMS sFlt-1:PlGF plus to the Elecsys® sFlt-1:PlGF ratios of 38 and 110 are 55 and 188 respectively.

Conclusion: Inter-manufacture immunoassay differences are significantly different. sFlt-1:PlGF rule in/rule out criteria are manufacturer specific, not interchangeable and require separate clinical validation.