The experiments presented here demonstrate that the monovalent inactivated A(H1N1)pdm09 influenza vaccine elicits readily detectable CD4+ T-cell responses and that the magnitude of these responses correlates with gains in neutralizing antibody. This positive relationship between CD4+ T cells and antibody responses was most apparent when looking at CD4+ T-cell responses directed against conserved peptide epitopes within the HA protein of the A/California/07/09 virus. On the basis of these results, we postulate that activation of CD4+ T cells following vaccination may be one of the limiting factors for neutralizing antibody production following pandemic influenza vaccination.
As there is great interest in defining biomarkers that will predict success in vaccination, our studies evaluated whether prevaccination levels of influenza virus–specific CD4+
T-cell reactivity predicted the magnitude of the neutralizing antibody response. We did not observe any correlation between these parameters but instead found a correlation between CD4+
T-cell expansion and the titer of neutralizing antibody produced. This suggests that, following vaccination, only a subset of influenza virus–reactive cells are able to be recruited into the draining lymph nodes and expand, leading the cells that ultimately reenter the circulating pool to be most indicative of the help available for the antibody response. The effect this has on CD4+
T-cell memory will be important to address in future studies. Our results contrast with what was recently reported in the study by Wilkinson et al [39
], in which greater baseline circulating numbers of influenza virus–specific CD4+
T cells were protective against development of severe disease in an influenza challenge model. One potential reason for this difference is that the study by Wilkinson et al predominantly examined the role of effector CD4+
T cells that exerted their function prior to either viral clearance or the production of neutralizing antibody, possibly via cytolytic activity. This contrasts with our study, which considered the neutralizing antibody produced in response to vaccine challenge. Additionally, more CD4+
T cells may be able to be recruited into the immune response following infection, because of a greater abundance and diversity of epitopes displayed by antigen-presenting cells. Our ability to use peripheral blood CD4+
T-cell reactivity to predict future vaccine-induced B-cell responses is likely to require a more refined definition of epitopes recruited by vaccination and a better understanding of the subsets of CD4+
T cells that can participate in extrafollicular and germinal center responses to vaccine components. Further, it is possible that the observed relationship between CD4+
T-cell and neutralizing antibody responses may be the result of an overall more robust vaccine response in some individuals. Future efforts to selectively boost CD4+
T-cell responses will help to confirm the causal relationship between these parameters.
An important issue we sought to evaluate in these studies was whether the influence of CD4+
T cells on neutralizing antibody responses to vaccination was related to antigen specificity. It is interesting that the strongest correlation observed was between expansion of HA-specific CD4+
T cells and the neutralizing antibody response, although the distinction between NP/M1- and HA-specific expansion and the antibody response that we detected was modest. Both concurrent expansion of HA-reactive CD4+
T cells and cells specific for the NP/M1 pool and the inclusion of M1 peptides within the NP pool, as M1 associates with the viral surface glycoproteins [40
] and may be taken up with HA by B cells, may have lessened our ability to detect a potential linkage between B-cell and T-cell specificities. If HA-specific B cells do have preferential access to limited viral proteins, vaccine development efforts that focus the CD4+
T-cell response on HA-derived epitopes may improve the neutralizing antibody response following vaccination.
One of the challenges to vaccination against pandemic influenza is that the viral protein composition of the next pandemic strain cannot be predicted [41
]. The failure to detect CD4+
T-cell responses to novel epitopes in the current study suggests either that these epitopes are poorly immunogenic or that naive cells fail to successfully compete with the more abundant and rapidly recruited memory cells. Further studies involving individuals who are now primed with A(H1N1)pdm09 will help clarify the overall immunogenicity of these peptides and the effect of competing memory T cells on naive CD4+
T-cell expansion and response kinetics. For antibody responses to A(H1N1)pdm09, there may have been enough conserved HA epitopes to promote antibody responses, but for more distant viruses conserved HA epitopes may be quite limited, possibly resulting in a correspondingly low neutralizing antibody response. This could explain the disparity between the robust responses to a single dose of A(H1N1)pdm09 vaccine [43
] as compared to A(H5N1) vaccine, to which responses are modest [45
]. If future studies substantiate the link between HA-specific CD4+
T cells and anti-influenza virus neutralizing antibody production, efforts to enrich the CD4+
memory population with T cells specific for potentially cross-reactive HA epitopes by prepriming with peptide-based vaccines or novel HA constructs [47
] may increase the recruitment of HA-specific CD4+
T cells on challenge with divergent HA proteins. Such a strategy could promote a more broadly cross-reactive and rapid response to novel strains of influenza virus, increasing pandemic preparedness by providing stand-alone protection while allowing dose-sparing efforts to facilitate rapid deployment of limited vaccine stocks to the population in the event of a pandemic.