Vaccine 20 (2002) S27–S31 Immunopathogenesis of vaccine-enhanced RSV disease Peter J.M. Openshaw * , Fiona J. Culley, Wieslawa Olszewska Department of Respiratory Medicine, National Heart and Lung Institute, Imperial College of Science, Technology and Medicine at St. Mary’s, Norfolk Place, London W2 1PG, UK Abstract Inducing a strong immune response is an essential aim of vaccination. Although immune responses to virus infections are usually protective, they can also be harmful. The best-documented examples of an immune response increasing disease severity are with dengue, measles and respiratory syncytial virus infections. In the 1960s, administration of formalin-inactivated, tissue culture grown RSV (FI-RSV) was found to induce strong ELISA binding but poor virus-neutralising antibody. Infants given this ‘lot 100’ vaccine appeared to exhibit an increased rate of RSV infection during subsequent natural RSV outbreaks. Although it has not been possible to exactly delineate the cause of disease enhancement in man, animal models strongly suggest that it was due to strong (and perhaps unbalanced) T cell priming rather than infection-enhancing or sensitising antibody. In animal models, enhanced disease can result from over-exuberant T cell priming which recruits an abundant inflammatory infiltrate in the lung (the nature of which depends on the patterns of cytokines and chemokines produced). Formalin-treated RSV vaccination has been linked specifically to the induction of Th2 cells, which make IL-4 and IL-5 and induce a strong pulmonary eosinophilic response. The vaccine dosing regime and the interval between vaccination and challenge can be critical to the induction of protection or pathology. Defining the correlates of protection and disease enhancement in man is critical to the rational development of effective and protective vaccines against RSV. © 2001 Published by Elsevier Science Ltd. Keywords: Respiratory syncytial virus; Th1/Th2 review; Immunopathology; Animal models 1. Human RSV formalin vaccine trials Following the success of killed polio vaccine and other inactivated viral vaccines (given singly or in combination), there was hope that other viruses could be chemically inac- tivated and retain their immunogenic properties. Trials with formalin-inactivated RSV vaccine were therefore conducted in 1966–1967 in infants and children aged 2 months to 9 years. During subsequent RSV exposure, the rate of the virus infection in infants who received the vaccine was no less (and was perhaps even greater) than that in control group im- munised with parainfluenza vaccine. Most remarkably, 80% of RSV vaccinees needed hospitalisation, whereas only 5% of such infections among control parainfluenza vaccinees required admission to the hospital [1]. Illnesses among vaccinated children included pneumonia, bronchiolitis, rhinitis or bronchitis. Sadly, two infants died. The major findings at post-mortem examination were exten- sive bronchopneumonia with emphysema and pneumotho- rax. Microscopic findings indicated an intense inflammatory * Corresponding author. Tel.: +44-20-7594-3854; fax: +44-20-7262-8913. E-mail address: p.openshaw@ic.ac.uk (P.J.M. Openshaw). cellular infiltrate in the lungs comprising mononuclear cells, eosinophils and polymorphonuclear cells. RSV could be grown from the lung (at 10 4 TCID 50 per gram of tissue). Additionally, pure culture of Escherichia coli was recovered from the trachea, lung, blood and spleen of one infant and Klebsiella from the lung, trachea and nose culture from the other. The intense inflammatory infiltrate in the lungs of vacci- nated children suggested an immunopathological cause of enhanced disease. Limited data is available regarding the immunological status of vaccinees before, during and after the immunisation. At the time of the study, it was commonly believed that antibodies were responsible for protection and blood samples were collected and stored to study humoral responses. Although there was a significant rise in specific ELISA antibody titres, the levels of neutralising antibodies were poor [1]. Subsequent investigations showed that serum from vaccinated children had poor fusion inhibiting anti- body, raising the possibility that fusion protein was altered during formalin treatment [2]. Significantly, the highest incidence of serious disease oc- curred in children younger than 6 months (when maternally derived serum antibody is still present). It has been sug- gested that the altered reactivity in the RSV vaccinees was a 0264-410X/01/$ – see front matter © 2001 Published by Elsevier Science Ltd. PII:S0264-410X(01)00301-2