A perfect storm: Impact of genomic variation and serial vaccination on low influenza vaccine effectiveness during the 2014-15 season.

Skowronski DM1, Chambers C2, Sabaiduc S2, De Serres G3, Winter AL4, Dickinson JA5, Krajden M6, Gubbay JB7, Drews SJ8, Martineau C9, Eshaghi A4,Kwindt TL2, Bastien N10, Li Y10.

Author information

Abstract

BACKGROUND:

The 2014-15 influenza season was distinguished by an A(H3N2) epidemic of antigenically-drifted virus and vaccine containing identical components to 2013-14. We report 2014-15 vaccine effectiveness (VE) estimates from Canada and explore contributing agent-host factors.

METHODS:

VE against medically-attended laboratory-confirmed influenza was derived by test-negative-design. Patients presenting with influenza-like-illness to sentinel outpatient clinics between November 1,2014 and April 30,2015 were eligible to participate. Influenza was diagnosed by RT-PCR. Sanger-sequencing identified amino-acid differences relative to vaccine at key antigenic-sites of the viral hemagglutinin(HA) protein.

RESULTS:

Among 1939 eligible participants, 814(42%) tested influenza-positive: 590(72%) influenza/A and 225(28%) influenza/B. Virtually all influenza/A viruses with known subtype (570/577;99%) were A(H3N2) with most sequenced viruses belonging to genetic clades 3C.2a (409/460;89%) or 3C.3b (39/460;9%). Both clades had multiple amino-acid differences from vaccine at antigenic-sites of the HA-protein, but clade-3C.2a viruses bore the pivotal mutations F159Y (cluster-transition-site) and adjacent K160T (predicted gain-of-glycosylation). VE was -16%(95%CI:-49-9%) for A(H3N2) overall: -13%(95%CI:-50-15%) for clade-3C.2a and 52%(95%CI:-16-81%) for clade-3C.3b. Among patients vaccinated in 2014-15 but not 2013-14, VE was 53%(95%CI:10-75%), significantly lower at -32%(95%CI:-75-0%) if also vaccinated in 2013-14 and -54%(95%CI:-109--14%) if serially-vaccinated each year since 2012-13. Among influenza/B detections with known lineage, most (193/199;85%) were B(Yamagata);all were clade-mismatched to vaccine with VE of 42%(95%CI:10-62%). A similar pattern of reduced VE with repeat vaccination was observed for B(Yamagata) but the effect was less pronounced compared to A(H3N2).

CONCLUSIONS:

A combination of agent-host factors, including variation in the viral genome and negative effects of repeat vaccination, likely contributed to poor influenza vaccine performance in 2014-15.