Modeling an approach to define sensitivity of viral detection in sample matrices-examples with microarray readout.

Abstract

Advances in viral detection technologies have the potential to increase the safety assurance of medicines produced in biological production systems. However, taking full advantage of these technological advances in the regulated testing environment will require protocols for standardization and performance testing. The most essential performance characteristics of detection methods for these applications include sensitivity, breadth of detection, and consistency. We have evaluated an approach to establishing the suitability of advanced nucleic-acid-based detection systems for characterization of cell substrates and production cultures. This approach is based on selecting relevant challenge viruses, their preparation and characterization, their application to a sample preparation workflow, and quantitation of their recovery by an independent means as well as the advanced detection readout. This approach helps us evaluate the suitability of the workflow for handling diverse sample matrices across manufacturing platforms for vaccines and other biological products, and also suggests a means by which technology users, developers, and regulators may consider the critical performance attributes of novel detection technologies. We have applied this workflow to a novel microarray-based viral detection system in collaboration with Lawrence Livermore National Laboratory. This system is appealing because of the rapid turnaround of results compared to some other advanced detection technologies.