Pitfalls in Measuring Cytokines

TO THE EDITOR:

Casey and colleagues [1] note the potential importance of measuring cytokine levels in patients with systemic sepsis, whereas Dinarello and Cannon [2] detail many of the pitfalls associated with assaying cytokine levels. We report some other difficulties inherent in measuring “immunoreactive” tumor necrosis factor (TNF)- α by enzyme-linked immunosorbent assay (ELISA). As noted by Casey and colleagues [1], the assessment of TNF-α immunoreactivity reflects the level of the biologically active and inactive forms of TNF-α. However, the level of immunoreactive TNF-α is greatly influenced by the type of monoclonal antibody used in the ELISA kit, as well as by the presence of type I and type II TNF-binding proteins (soluble receptors) in the serum during septic shock [3]. This point is illustrated in Figure 1. In the ELISA kit used by Casey and colleagues [1] (T Cell Science, Cambridge, Massachusetts), the level of TNF-α immunoreactivity was masked by 75% and 60%, respectively, in the presence of type 1 TNF-binding proteins and type 2 TNF-binding proteins. In contrast, the level of immunoreactive TNF-α measured by an ELISA kit from a second vendor (Genzyme Corp., Cambridge, Massachusetts) was not affected by the type 1 proteins, whereas TNF-α immunoreactivity was masked by 80% in the presence of the type 2 proteins. Finally, levels of immunoreactive TNF-α measured by an ELISA kit from a third vendor (R & D Systems, Minneapolis, Minnesota) were not masked by either type of protein.