The common H202D variant in GDF-15 does not affect its bioactivity but can significantly interfere with measurement of its circulating levels

Genetic variants in proteins can interfere with measurement of their circulating concentrations. Given the growing biomedical importance of GDF-15, we wished to establish whether a common histidine to aspartate variant present in position 6 of the mature GDF-15 protein (H202D variant) interfered with its measurement by two commonly used immunoassays. We first examined the detectability of recombinant monomers, homodimers and heterodimers of GDF-15 by assays and/or reagents used in two widely used immunoassays (Roche Elecsys(R) GDF-15 and the R&D antibody combinations used in their Quantikine(R) and DuoSet(R) ELISAs). The Roche assay detected the H and D containing peptides similarly but the assays based on the R&D reagents consistently underreported concentrations of the D-containing variant peptide. Measurements of plasma concentrations of GDF-15 in genotyped human participants showed that the R&D reagents reported values in heterozygotes were [~]25% lower, and in homozygotes, 50% lower than the Roche assay. We finally studied the activation of the GDF-15 receptor, GFRAL-Ret, in a cell based assay and found that the activities of the HH and DD containing GDF-15 peptide were indistinguishable. These results have implications for the interpretation of genetic epidemiological studies which have used the R&D reagents to measure GDF-15, and for the emerging clinical use of GDF-15 as a diagnostic and prognostic biomarker. We provide correction equations, which may be of utility for the analysis of data generated with the R&D reagents where the genotype of the participants is known.