Correction for non‐specific interference in competitive immunoassays

Abstract

Methods for correcting for non‐specific interference in immunoassays are outlined. Two main correction methods are evaluated: i) The use of antibodies to non‐naturally occurring substances, ii) curve‐fitting approaches. Curve‐fitting is based on the assumption of non‐competitive kinetics where only the maximum binding of label is affected, with no significant effect on the shape of the response curve. The latter assumption is evaluated using a simple mass‐action model, and shown to be adequate for many assays of the plant growth regulator, abscisic acid. A non‐linear curvefitting approach is presented and is shown to have wide applicability to all immunoassays beyond its use in correcting for interference. All correction methods have reduced sensitivity and require additional data compared with direct estimation of antigen. Nevertheless the additional measurements, which involve the assay of each sample with a range of added spikes of unlabelled antigen, should be routinely included in assays in order to test for interference.