Model analysis of bidirectional interference in two-stage labeled-ligand immunoassays

Immunoassays involving sample incubation followed by a wash step prior to introduction of labeled analyte are potentially subject to both positive and negative interference (bidirectional interference) by a competing ligand. We examine this phenomenon from a theoretical standpoint using a mathematical model for sequential-step immunoassays in the presence of interferent. Competitive binding to antibody between analyte and interferent was modeled for sequential-step immunoassays. A primary assumption was that the ratio of affinity constants between the intended analyte and the interferent reflected the ratio of dissociation rate constants, with the higher dissociation rate constant for the lesser affinity ligand. Relationships of parameters (relative affinity constants, relative concentrations) for analyte and interferent were determined for conditions in which bidirectional interference can occur, for both steady-state and non-steady-state sample incubation conditions. Non-steady state sample incubation conditions can enhance the effects of an interferent. Homogeneous assay formats utilizing labeled ligand without a wash step can also demonstrate bidirectional interference, but positive interference is favored under such formats. Model calculations demonstrate the theoretical basis for bidirectional interference in two-stage immunoassays. Results delineate constraints on conditions in which bidirectional interference can occur. •A mass balance model was used to model competitive binding in two-stage immunoassays.•Interferent was assumed to have a lesser dissociation constant (Kd) relative to analyte.•Interferent was assumed to have a higher dissociation rate constant (=1/time) relative to analyte.•Model calculations demonstrate theoretical basis of bidirectional interference.•Example calculations are made for steady-state, non-steady-state, and homogeneous immunoassays.