Cold-adapted live influenza vaccine versus inactivated vaccine: systemic vaccine reactions, local and systemic antibody response, and vaccine efficacy. A meta-analysis.

Since the 1940s, influenza vaccines are inactivated and purified virus or virus subunit preparations (IIV) administered by the intramuscular route. Since decades, attempts have been made to construct, as an alternative, attenuated live influenza vaccines (LIV) for intranasal administration. Presently, the most successful LIV is derived from the cold-adapted master strains A/Ann Arbor/6/60 (H2N2) and B/Ann Arbor/1/66 (AA-LIV, for Ann-Arbor-derived live influenza vaccine). It has been claimed that AA-LIV is more efficacious than IIV. In order to assess differences between the two vaccines with respect to systemic reactogenicity, antibody response, and efficacy, we performed a meta-analysis on eighteen randomised comparative clinical trials involving a total of 5000 vaccinees of all ages. Pooled odds ratios (AA-LIV versus IIV) were calculated according to the random effects model. The two vaccines were associated with similarly low frequencies of systemic vaccine reactions (pooled odds ratio: 0.96, 95% confidence interval: 0.74-1.24). AA-LIV induced significantly lower levels of serum haemagglutination inhibiting antibody and significantly greater levels of local IgA antibody (influenza virus-specific respiratory IgA assayed by ELISA in nasal wash specimens) than IIV. Yet, although they predominantly stimulate different antibody compartments, the two vaccines were similarly efficacious in preventing culture-positive influenza illness. In all trials assessing clinical efficacy, the odds ratios were not significantly different from one (point of equivalence). The pooled odds ratio for influenza A-H3N2 was 1.50 (95% CI: 0.80-2.82), and for A-H1N1, 1.03 (95% CI: 0.58-1.82). The choice between the two vaccine types should be based on weighing the advantage of the attractive non-invasive mode of administration of AA-LIV, against serious concerns about the biological risks inherent to large-scale use of infectious influenza virus, in particular the hazard of gene reassortment with non-human influenza virus strains.