2556. Retrospective Evaluation of Mismatch from Egg-based Isolation of Influenza Strains Compared to Cell-based Isolation and the Possible Implications for Vaccine Effectiveness
Session: Oral Abstract Session: Flu and other Vaccines in Children
Saturday, October 6, 2018: 2:45 PM
Room: W 2002

Background: Lower influenza vaccine effectiveness (VE) against circulating H3N2 strains compared to other influenza viruses, is partly explained by antigenic mismatch between circulating strains and the vaccine strain (Belongia 2016).  This mismatch has recently been linked to a new glycosylation site introduced in the egg-adaptation step (Zost 2017) and HA L194P substitution (Wu 2017) for H3N2. Vaccine manufactured using seed virus wholly grown in mammalian (e.g. Madin-Darby Canine Kidney – MDCK) cells, as with the NH17-18 version of Flucelvax®, avoids these mutations.  Preliminary reports suggest this cell-based vaccine showed greater VE than did similar egg-based vaccines [FDA Statement]. This study aimed to compile existing data on antigenic similarity to measure the degree of match with circulating wild-type isolates of egg- and MDCK-propagated versions of the vaccine H3N2 virus over multiple seasons.

Methods: Using publicly available reports from the Worldwide Influenza Centre, London (Crick), we compiled data on antigenic similarity, defined as H3N2 circulating wild-type virus isolates showing no more than a four-fold reduction in titer to antisera raised against wholly MDCK- or egg-propagated versions of the vaccine H3N2 viruses. Titers were compared using hemagglutination inhibition (HI) assays and/or plaque reduction neutralization assays (PRNA).

Results: Data from Northern Hemisphere influenza seasons of 201112 to 2017–18, show a substantially higher proportion of tested circulating influenza H3N2 viruses matched the MDCK-propagated reference viruses than did corresponding egg-propagated reference vaccine viruses (Figures 1 & 2).  In half of the seasons evaluated, there was little to no antigenic similarity between circulating viruses and the egg-based vaccine viral seed. 

Conclusion: These data suggest higher levels of mismatch have occurred consistently with egg-propagated H3N2 reference viruses compared to MDCK-propagated reference viruses when measured against circulating wild-type isolates and may further explain the potential for lower VE observed against H3N2 historically. Further, these data point to the importance of continuing to utilize cell-derived seeds in creating seasonal influenza vaccines for this strain.


S Rajaram, MD, MRCPCH, MFPM, MBA, Seqirus, Maidenhead, United Kingdom, Josephine Van Boxmeer, MPhil, Clinical Epidemiology, Seqirus, Amsterdam, Netherlands, Brett Leav, MD, Seqirus, Inc., Cambridge, MA, Pirada Suphaphiphat, PhD, Seqirus, Cambridge, MA, Ike Iheanacho, MBBS, BSc, Evidera, London, United Kingdom and Kristin Kistler, PhD, Evidera, Waltham, MA


S. Rajaram, Seqirus: Employee , Salary .

J. Van Boxmeer, Seqirus: Employee , Salary .

B. Leav, Seqirus: Employee and Shareholder , Salary .

P. Suphaphiphat, Seqirus: Employee , Salary .

I. Iheanacho, Seqirus: Consultant , Research support .

K. Kistler, Seqirus: Consultant , Research support .

Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 3rd with the exception of research findings presented at the IDWeek press conferences.