Theoretical Framework and Experimental Validation of Multiplexed Analyte Quantification Using Cross-Reactive Affinity Reagents

Gold standard immunoassays depend on specific affinity reagents for accurate molecular quantification. Any cross-reactivity of affinity reagents, wherein the reagent nonspecifically binds to unintended molecules, can create false positive binding signals and result in inaccurate quantification of analytes. Mitigating cross-reactivity represents one of the greatest challenges in molecular diagnostics, and remains an unsolved problem. To mitigate the burden of cross-reactive reagents, we present a mathematical framework that uses generalized binding equations and noise estimation to enable the accurate use of multiple cross-reactive reagents for molecular quantification. As a proof of concept, we experimentally demonstrate accurate 2-plex quantification of a small molecule for which no specific affinity reagents are available, even in the presence of high concentrations of a cross-reactive molecule in 10% human serum. This robust schema yields well-defined bounds of quantification that make it easier to assess the quality of assay results, and predicts conditions under which assay performance is likely to break down. This work turns cross-reactive affinity reagents, which are conventionally unused, into a tool for achieving accurate quantification of analytes.