A-008 Development and characterisation of immunoturbidimetric NT-proBNP assay based on antibodies targeting glycosylation-free regions of NT-proBNP

Abstract Background Multiple commercial immunoassays are available for the measurement of circulating NT-proBNP levels. Most of the assays use antibodies that bind to the central region of NT-proBNP, which is usually glycosylated. Glycosylation of NT-proBNP restricts the binding of the antibodies and detection of the protein. Several studies have confirmed that NT-proBNP assays underestimate the concentration of plasma NT-proBNP measuring only a subfraction of the circulating NT-proBNP because of the negative effect of glycosylation on NT-proBNP recognition by the antibodies. Aim of this study is to develop an NT-proBNP assay which is not affected by glycosylation of the protein and comparison of its diagnostic utility with performance of other commercially available immunoassays. Methods Analytical performance of the assay in optimization phase including linearity, LoQ, precision, antigen access, interference, calibration curve- and onboard reagent stability was evaluated according to Clinical and Laboratory Standards Institute (CSLI) guidelines. Instrument variation study was performed using Mindray BS380 and Beckman Coulter DxC 700 AU instruments. To determine the total concentration of NT-proBNP, independent of glycosylation, plasma samples were either non-treated or treated with a mixture of glycosidases for 24 h at 37 °C. Following, samples were analysed by a commercially available NT-proBNP immunoassay, based on antibodies targeting a glycosylated region of NT-proBNP and the prototype of the Gentian NT-proBNP assay, based on antibodies targeting regions of NT-proBNP that are free of O-glycans. Results NT-proBNP concentrations were measured in plasma samples either treated or untreated with a mixture of glycosidases. No significant effect of the glycosidase treatment was detected on NT-proBNP values measured by a prototype of Gentian NT-proBNP assay. NT-proBNP concentrations in treated and non-treated samples were similar (slope=0.849) and highly correlated (R2=0.936). As a comparison, we used one of the commercially available assays with antibodies targeting the central region of the NT-proBNP. NT-proBNP concentrations showed a good correlation (R2=0.879), but the assay strongly underestimated concentrations in untreated samples (slope=6.34), The effect of glycosylation was also shown when results measured with Gentian NT-proBNP assay prototype were compared with results from glycosylated and non-glycosylated samples measured with a commercial assay. Results from samples treated with glycosidases and measured with a commercial assay were significantly closer to results obtained with Gentian assay (R2=0.8898) compared to results from non-treated samples. Instrument variation study showed great correlation between values measured on instruments from Mindray and Beckman Coulter (slope = 0.87 and R2=0.969). Conclusion Our results confirm the impact of glycosylation on measurement of NT-proBNP and underestimation of NT-proBNP levels by assays based on antibodies targeting glycosylated regions of NT-proBNP. Given the considerable variability in levels and site occupancy of O-glycosylated proteins, it is reasonable to anticipate that immunoassays targeting glycosylation-free regions of NT-proBNP may offer advantages in heart failure (HF) diagnosis and prognosis for specific patient groups and disease states since these assays can detect endogenous NT-proBNP regardless of its glycosylation status. Further investigations, including exploration of the clinical significance are ongoing in order to understand