Abstract
BACKGROUND: Reliable exposure information is crucial for assessing health outcomes of influenza infection and vaccination. Current serological methods are unable to distinguish between anti-hemagglutinin (HA) antibodies induced by infection or vaccination. OBJECTIVES: We aimed to explore an alternative method for differentiating influenza infection and vaccination. METHODS: Sera from animals inoculated with influenza viruses or purified H1N1pdm09 HA were obtained. Human samples were selected from a pregnancy cohort established during the 2009 H1N1 pandemic. Unvaccinated, laboratory-confirmed cases (N = 18), vaccinated cases without influenza-like-illness (N = 18) and uninfected, unvaccinated controls (N = 18) were identified based on exposure data from questionnaires, national registries and maternal hemagglutination inhibition (HI) titres at delivery. Animal and human samples were tested for antibodies against the non-structural protein 1 (NS1) and HA from H1N1pdm09, using a Luciferase Immunoprecipitation System (LIPS). RESULTS: Anti-NS1 H1N1pdm09 antibodies were detected in sera from experimentally infected, but not from vaccinated, animals. Anti-HA H1N1pdm09 antibodies were detectable after either of these exposures. In human samples, 28% of individuals with laboratory-confirmed influenza were seropositive for H1N1pdm09 NS1, whereas vaccinated cases and controls were seronegative. There was a trend for H1N1pdm09 NS1 seropositive cases reporting more severe and longer duration of symptomatic illness than seronegative cases. Anti-HA H1N1pdm09 antibodies were detected in all cases and in 61% of controls. CONCLUSIONS: The LIPS method could differentiate between sera from experimentally infected and vaccinated animals. However, in human samples obtained more than 6 months after the pandemic, LIPS was specific, but not sufficiently sensitive for ascertaining cases by exposure.