Multipoint kinetic methods evaluated for quantitation of theophylline with prosthetic group label immunoassay

We describe the development and evaluation of a multipoint kinetic approach to the measurement and data-processing steps in immunoassays. Nonlinear regression methods are used to fit data obtained during the kinetic phase of the immunochemical reaction to a model that represents the kinetic behavior of the total system. This approach is evaluated for quantitation of theophylline with a new reaction system in which FAD is the label in a homogeneous reaction system. Competition for antibody binding sites between drug and labeled drug is monitored by the activity of glucose oxidase (EC 1.1.3.4) generated by the reaction of unbound label with apoglucose oxidase. The process is fit well by a parallel first-order/zero-order model and regression methods are used to predict steady-state velocities from non-steady-state portions of the response. One fitting method yields results in about 3 min after mixing, as compared with longer than 8 min for single-point fixed-time methods. Steady-state velocities vary linearly with theophylline concentration.