No Serological Evidence of Influenza A H1N1pdm09 Virus Infection as a Contributing Factor in Childhood Narcolepsy after Pandemrix™ Vaccination Campaign in Finland

​Melén K,  Partinen M, Tynell J, Sillanpää M, Himanen S-L, Saarenpää-Heikkilä O, Hublin C, Olsen P, Ilonen J, Nohynek H, Syrjänen R, Kilpi T, Vuorela A, Kirjavainen T, Vaarala O, Julkunen I

PLoS ONE 8(8): e68402. doi:10.1371/journal.pone.0068402 (published August 8, 2013)

An epidemiological association between vaccination with AS03-adjuvanted influenza A(H1N1)pdm09 pandemic vaccine Pandemrix™ and development of narcolepsy has been reported from six EU Member States: Finland, France, Ireland, Norway, Sweden, and United Kingdom (1-8).   Regulatory activities and subsequent epidemiological studies related to the reported adverse event conducted so far are listed in the ECDC timeline (9). Following the epidemiological confirmation, several studies were initiated to increase the understanding of possible biological mechanisms leading to this unexpected event. In this study Finnish investigators have investigated whether infection with the influenza A(H1N1)pdm09 virus itself could have contributed to the sudden and significant increase in the incidence of childhood narcolepsy observed in Finland in 2010 (10).

 

Sera from vaccinated children with confirmed narcolepsy (n=45), acute and convalescent sera from age-matched individuals naturally infected with a laboratory-confirmed influenza A (H1N1)pdm09 virus (n=28) and control sera from age-matched control-individuals (n=50) collected before the 2009 pandemic were evaluated for presence of influenza non-structural protein 1 (NS1) - specific antibodies using recombinant influenza A virus NS1 proteins and Western blot technology. Antibodies to the influenza A(H1N1)pdm09 NS1 protein were selected for analysis since these proteins were not part of the components of the pandemic vaccine, produced on the backbone of the vaccine PR8 strain and only expressing the surface proteins from the influenza A(H1N1)pdm09 virus. Influenza A (H1N1)pdm09 NS1-specific antibody presence in sera therefore is indicative of a previous natural infection at some time-point either before or after vaccination but during the winter season of 2009/2010. Antibody reactions to influenza A(H3N2) NS1 proteins were used as a control.

 

In vaccinated children and adolescents high antibody reactivity to the influenza A(H3N2) NS1 subtype was observed while the reactivity to the influenza A(H1N1)pdm09 NS1 subtype was low, indicating a low likelihood of a previous infection with influenza A(H1N1)pdm09 virus. Geometric mean antibody titers (GMT) against influenza A(H1N1)pdm09 subtype was low in narcoleptic children and adolescents (GMT 156; 95% CI: 111-218), high in naturally infected (GMT 8,414; 95% CI: 4295-16,482) and also rather low in age-matched control children and adolescents (GMT 418;  95% CI: 300-582). Among the vaccinated children and adolescents with narcolepsy only two out of 45 sera showed specific antibody response against the influenza A(H1N1)pdm09 NS1 proteins. Sera from all three groups of children and adolescents displayed high levels of A(H3N2) subtype NS1-specific antibodies.

 

The Finnish authors conclude that the results strongly suggest that pandemic influenza  A(H1N1)pdm09 virus infection was not a main contributing factor to the increased incidence of narcolepsy observed in Finland in the beginning of 2010.

ECDC Comment (15 August 2013):

This is the first published study where an attempt has been made to understand the possible biological mechanisms behind development of narcolepsy following vaccination with Pandemrix™. Earlier epidemiological studies have been able to confirm an association.

To evaluate an adverse event following immunization (AEFI) may be challenging. This has certainly been the case for the investigations of the unexpected reports of narcolepsy following vaccination with Pandemrix™.

 

To guide assessments, a standardized algorithm to assist in collecting and interpreting data and help assess causality after individual AEFIs has recently been published by Halsey et al (11).  However, the algorithm requires a case definition for the reported adverse event, a proper diagnosis and preferably a known causal association. When narcolepsy initially was reported as a possible adverse event, there was no agreed universal case definition or diagnostic criteria.  The diagnostic capacity differed significantly between clinical centres and Member States and the event had never been reported following vaccination and therefore no AEFI case definition had been developed along the Brighton Collaboration criteria (12). In addition, no biological mechanisms were established and therefore causality assessment was difficult. However, there were indications from the scientific literature that infections could play a role in triggering narcolepsy (14-16). Also, a biologically plausible and evidence-based risk interval post-vaccination was and still is unknown (17), although the epidemiological studies now provide some evidence for when narcolepsy developed following vaccination (1). From these the risk interval seem long and will make clinical and causality interpretations complex.

 

This study found no serological evidence of influenza A(H1N1)pdm09 virus infection as a contributing factor to development of childhood narcolepsy after the conducted pandemic vaccination campaign using the AS03-adjuvanted influenza A(H1N1)pdm09 pandemic vaccine Pandemrix™.

 

These results are supported by earlier Finnish published data that showed that no increased incidence of narcolepsy was observed in unvaccinated children (2).  Similar findings have been observed in all countries that assessed narcolepsy following Pandemrix vaccination (3-7). In addition, the nationwide Finnish Infectious Disease Registry, which collects data on microbial infections in Finland does not provide evidence that any of the 15,000 confirmed A(H1N1)pdm09 infection cases (all ages) diagnosed in 2009–2011 have developed narcolepsy (11). Therefore the authors are suggesting that in genetically susceptible children Pandemrix vaccination may alone be sufficient to initiate/ accelerate narcolepsy.

 

Seasonality in the diagnosis of narcolepsy has been described from China and the same investigators interestingly reported an increase following the 2009 influenza A(H1N1) pandemic in China (15-16). This prompted the Finnish investigators to investigate whether influenza A(H1N1)pdm09 virus infections had contributed to the sudden increase in incidence of childhood narcolepsy in Finland.

 

Today, more than ten research groups mainly from the Nordic countries, France and the US (US investigators at Stanford University engaged by GlaxoSmithKline) have initiated studies with the aim to study possible biological mechanisms. First preliminary results were reported in the First Nordic Symposium on narcolepsy held in Finland on February 7-8, 2013 (18) and now the first scientific publication is made available. Further publications are expected when results evolve.

 

To support the immunization programmes and maintain the public and professional confidence in the EU, an increased preparedness to conduct epidemiological studies should safety signals arise is needed, including scientific competence and funding.  All data published so far on the epidemiological association between Pandemrix and development of narcolepsy has been funded through often limited public sources.

 

Meanwhile, we face the emergence of new influenza viruses with pandemic potential such as influenza A(H7N9), and the need of developing new vaccine candidates at least to be used for first clinical trials.  Ideally the development of future influenza vaccines should benefit from a greater understanding of this event than is currently known.

 

Comments or questions on this Scientific Advance are welcome and should be addressed to: influenza@ecdc.europa.eu.