Baculovirus Binding Assay for Antibody Clearance Prediction

The patent applies preliminary action by performing in silico predictions and in vitro binding assays before conducting in vivo studies. The method uses computational models to predict off-target binding and identifies high-risk antibodies early in the selection process, allowing researchers to prioritize which candidates warrant expensive in vivo testing. This preliminary screening reduces the number of animals needed and accelerates the overall drug discovery timeline.

The patent employs copying by creating in silico models that replicate in vivo pharmacokinetic behavior. The computational models simulate antibody clearance mechanisms and predict human pharmacokinetics based on non-clinical data, providing a virtual copy of the complex in vivo system. This allows multiple scenarios to be tested computationally before committing to actual animal studies.

The patent performs preliminary in silico predictions and in vitro binding assessments before expensive in vivo studies. By identifying antibodies with high predicted risk of fast clearance through computational models and off-target binding assays, the method filters out poor candidates early, ensuring that costly in vivo resources are allocated only to promising candidates with lower predicted risk.

The patent introduces in silico predictions and in vitro binding assays as intermediary steps between antibody selection and in vivo testing. These intermediate assessments provide early indicators of pharmacokinetic risk, serving as mediators that guide the selection process and reduce the number of antibodies requiring expensive in vivo validation.

The patent uses in silico models as virtual copies to predict pharmacokinetic behavior and identify fast clearance risk. The computational models replicate the complex interactions between antibodies and the human pharmacokinetic system, allowing rapid assessment of multiple candidates without wet lab experiments. This digital copying approach maintains prediction accuracy while dramatically improving throughput and efficiency.

The patent replaces traditional mechanical/wet lab-based selection methods with in silico computational predictions. Instead of relying solely on time-consuming in vitro binding assays and in vivo studies, the method uses computer-based models to predict off-target binding and clearance risk, substituting computational mechanics for traditional experimental mechanics and significantly accelerating the selection process.