Article Text

Download PDFPDF

Measurement of tumour necrosis factor α
  1. Division of Gastroenterology
  2. University Hospital, Nottingham, UK andrew.austin{at}
  1. H-D VOLK
  1. Institut für Medizinische Immunologie
  2. Humboldt-Universität, Campus Charité Mitte
  3. D-10098 Berlin, Germany
  1. Professor H-D Volk.hans-dieter.volk{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Editor,—In their paper (OpenUrlPubMed), von Baehret al suggest that the high sensitivity Quantikine ELISA for tumour necrosis factor α (TNF-α) is capable of differentiating between trimeric (bioactive) TNF-α and proteolytic split products of TNF-α. Both forms are thought to be measured by the Medgenix assay. They go on to propose that this allows a measure of recently released bioactive TNF-α as opposed to an estimate of release over past hours. The authors do not include any data to substantiate such a claim.

Following discussion with R&D systems, I can confirm that they do not claim that their high sensitivity kit measures only trimeric (bioactive) TNF-α and that there are no data comparing this kit with a bioassay as they measure different things, namely immunoreactivity (mass) versus bioactivity. The apparent twofold greater level of TNF-α found using the Medgenix kit may simply reflect a calibration difference between the two kits.


Editor,—Several years ago we compared different tumour necrosis factor (TNF) assays.1-1 We observed a strong correlation between the bioassay and the R&D assay (r=0.88) whereas the correlation was poor (<0.3) between both assays and the Medgenix ELISA. Moreover, the kinetic studies after in vitro lipopolysaccharide stimulation showed that bioactive TNF is produced for a few hours only whereas the Medgenix TNF assay (in contrast with the R&D assay) detects TNF even 24 hours after stimulation (fig 4 in Asadullah and colleagues1-1). In addition, monoclonal antibodies recognising and neutralising TNF interact with the R&D assay but not with the Medgenix assay. Hence the differences between the assays are not simply due to differences in calibration. If sera that contained TNF immunoreactivity were fractionated into fractions less than or greater than 40 kDa (the trimer has about 51 kDa) it was observed that the Medgenix but not the R&D assay recognised low molecular weight as well high molecular weight fractions (unpublished data). In summary, there is strong evidence that the Medgenix assay recognises biologically inactive TNF split products in addition to the bioactive TNF trimer but the R&D assay does not.


  1. 1-1.