Presence of defective viral genes in H1N1 live attenuated influenza vaccine strains is not associated with reduced human cell fitness or vaccine effectiveness

Abstract

In the 2013–2014 and 2015–2016 seasons, quadrivalent live attenuated influenza vaccine (LAIV) showed reduced pandemic 2009 H1N1 (A/H1N1pdm09) vaccine effectiveness (VE) in the U.S. Impaired fitness of A/H1N1pdm09 LAIV strains in primary human nasal epithelial cells (hNEC) was subsequently shown to be associated with reduced VE. As defective viral genes (DVG) have been detected in QLAIV, it was hypothesised that these might play a role in reduced A/H1N1pdm09 fitness. By applying RT-PCR based approaches, DVG for PB2, PB1 and PA segments were detected in all influenza A LAIV strains tested. Absolute quantification of PA vRNA as a biomarker, using a novel digital RT-PCR assay (RT-dPCR), showed that DVG were a minority population (between 10.2 and 27.8 % of PA vRNA) in LAIV, irrespective of subtype or VE. Importantly, no difference was observed between the fitter pre-2009 H1N1 A/New Caledonia/20/1999 (A/NC99) and less fit A/H1N1pdm09 A/Bolivia/509/2013 (A/BOL13), containing medians of 16.0 % and 10.2 % PA DVG, respectively. Manipulating propagation conditions and minimising A/BOL13 PA DVG to 5.2 % failed to improve viral replication in hNEC, suggesting DVG were not limiting A/BOL13 fitness. Conversely, DVG were able to reduce A/NC99 replication in hNEC to A/BOL13-like levels, but only by enrichment of PA DVG to 66.5 % of vRNA. Notably, this required LAIV propagation under conditions markedly different to those used for vaccine production. We conclude from these data that abundance of DVG in QLAIV is not associated with variations in influenza A VE and that the reduced fitness of A/BOL13 previously described was not driven by the presence of DVG.