Comparison of two enzyme immunometric assays to measure tumor necrosis factor-alpha in human serum

Background: Tumor necrosis factor-alpha is a 17.5 kDa, 157 amino acid protein that is a potent lymphoid factor, which exerts cytotoxic effects on a wide range of tumor cells and other target cells. TNF-alpha has been suggested to play a pro-inflammatory role by influencing transendothelial migration of monocytes and elicits the expression of proteolytic enzymes by macrophages and smooth muscle cells within the atherosclerotic plaque. Methods: We compared two methods for the quantitative determination of human tumor necrosis factor-alpha in serum samples. Either kit follows the same assay procedure. Serum samples do not need to be diluted before sampling. Standard is provided lyophilized and serial dilutions after reconstitution generate the standard point curves. The tests are enzyme immunometric assays based on a standard 96-well microtiter plate. The wells are coated with anti-human TNF-alpha antibody. Results: The range of the standard curve is similar in both kits. It spans from 1000 through 15.6 pg/mL. The median TNF-alpha concentration in samples measured by Pierce assay (n = 368) was 4.23 pg/mL, (range, 1.34-77.2 pg/mL). A very different median was obtained for the same specimen measured with the Titerzyme EIA (median, 176.96 pg/mL; range, 54.7-283.9 pg/mL; n = 364). Substantial significant differences were observed between the two methods. Conclusion: Our results show that the two kits are unmatchable for results they can give when TNF-alpha concentrations are measured in serum samples. One reason of this disagreement could be the matrix effect or a cross-reactivity of one of the two methods. This study shows that the determination of human serum TNF-alpha needs to be standardized, especially when a comparison of results is required. (C) 2005 Elsevier B.V. All rights reserved.