In the present study, we compared antibody profiles in response to infection and vaccination with influenza A(H1N1)pdm09 with a PA. This technology was developed to enable comparative studies using standardized assay format to reduce the problem of variability and potentially interlaboratory differences in results of HI and microneutralization testing by obviating the need for use of biological reagents that are difficult to standardize, such as red blood cells (for HI assays) or cells (for virus neutralization assays). 
We demonstrated that subjects with a history of seasonal vaccination generally exhibited higher baseline titers for the various HA antigens than subjects without such history. We also show that the response differs according to the type of exposure: following natural infection, a strong homologous response, and weaker but significant increases in antibody titers to different HA1 antigens are seen, whereas the response following vaccination was more restricted. These results need to be interpreted with caution, because the mean age of the persons in the infection group was lower, which in part could explain a more vigorous antibody response. Remarkably, in both groups, responses in persons with a history of vaccination were skewed towards older seasonal H1 antigens. These details were not evident from the kinetics in HI titer, the method routinely used for evaluation of antibody responses.
The study confirms that antibody response to infection and vaccination are both shaped by prior exposure history, where prior regular seasonal influenza virus vaccination had a clear effect on the magnitude and kinetics of response to pandemic influenza infection and vaccination. The higher GMTs at baseline were particularly clear for antibodies binding to HA1 peptide from recent seasonal influenza viruses, but slightly elevated levels were also observed for antigens not offered through vaccination within subtype H1. These reactivities may reflect the broadening of antibody response with age. 
An intriguing question is whether such antibodies influence outcome of infection. Studies during the pandemic have been inconclusive in this respect: The presence of cross-neutralizing antibodies was limited to higher age groups, and no protective effect was expected or observed in several studies. 
In contrast with this, some degree of cross protection from severe illness by prior seasonal vaccination was suggested in some studies. 
Prior infection with an influenza A virus can reduce morbidity and mortality caused by an infection with an antigenically divergent influenza A virus because of heterosubtypic immunity, both within and between subtypes. 
Both natural infection and seasonal vaccination can induce heterosubtypic neutralizing antibodies, but our data suggest skewing against such antibodies in persons with a history of seasonal vaccination. 
The immunological basis of heterosubtypic immunity is not fully understood, but B cells, CD4+
T-cells and mucosal immunity may contribute. 
Passive serum transfer showed that antibodies induced by seasonal influenza A(H1N1) virus conferred protection in naïve recipient mice against A(H1N1)pdm09 challenge. The presence or absence of HI antibodies, therefore, is not the sole indicator of the effectiveness of protective cross-reactive antibody immunity. 
In a mouse model, Hillaire et al. 
, demonstrated that induction of T cells specific for a seasonal H3N2 influenza virus led to protection against infection with the antigenically unrelated A(H1N1)pdm09. In addition, repeated infection with seasonal influenza virus improved protection and clearance of influenza A(H1N1)pdm09 in ferrets. 
In young children, a difference was observed in levels of CD8+ T cells between vaccinated and unvaccinated individuals, suggesting that the same mechanisms may apply in humans as postulated by Bodewes et al. 
The influenza pandemic of 2009 showed an unbalanced age distribution of infected individuals, with a low incidence in elderly and a high incidence in children. This could be partly explained by the lack of heterosubtypic immunity, as a proportion of young children are immunologically naïve for influenza viruses. 
However, others described that low level heterosubtypic antibody responses following seasonal influenza vaccination could offer immune protection against antigenically distinct influenza viruses to a certain extent. 
The observations above imply that seasonal influenza vaccines should also be evaluated for their capacity to mimic the balance in response triggered by wild type infection. Studies from Skowronski et al. 
suggested an increased risk of illness in persons with a history of seasonal influenza vaccination during the pandemic, which potentially could be explained by reduced levels of cross protective antibodies or T cells as a result of reduced wild type infection.
In this study, we show that a history of seasonal influenza vaccination has different effects on infection and vaccination response. In vaccinees, the level of antibodies to the homologous strain was reduced in persons with a history of vaccination, whereas the reverse was true for infected persons. In both groups, however, the antibody response was skewed against heterologous antigens. More research is needed to understand if these observations are relevant for susceptibility of the individuals for infection and disease. We also conclude that improved assessment of the quality of immune response is needed when evaluation current and potential influenza vaccines.