The search for a test that can predict vaccine efficacy is an important part of any vaccine development program. Although regulators hesitate to acknowledge any test as a true 'correlate of protection', there are many precedents for defining 'surrogate' assays. Surrogates can be powerful tools for vaccine optimization, licensure, comparisons between products and development of improved products. When such tests achieve 'reference' status however, they can inadvertently become barriers to new technologies that do not work the same way as existing vaccines. This is particularly true when these tests are based upon circularly-defined 'reference' or, even worse, proprietary reagents. The situation with inactivated influenza vaccines is a good example of this phenomenon. The most frequently used tests to define vaccine-induced immunity are all serologic assays: hemagglutination inhibition (HI), single radial hemolysis (SRH) and microneutralization (MN). The first two, and particularly the HI assay, have achieved reference status and criteria have been established in many jurisdictions for their use in licensing new vaccines and to compare the performance of different vaccines. However, all of these assays are based on biological reagents that are notoriously difficult to standardize and can vary substantially by geography, by chance (i.e. developing reagents in eggs that may not antigenitically match wild-type viruses) and by intention (ie: choosing reagents that yield the most favorable results). This review describes attempts to standardize these assays to improve their performance as surrogates, the dangers of over-reliance on 'reference' serologic assays, the ways that manufacturers can exploit the existing regulatory framework to make their products 'look good' and the implications of this long-established system for the introduction of novel influenza vaccines.
Keywords: Influenza virus; correlates of protection; reference assays; reference reagents; standardization; vaccine.