Antibody persistence and serological protection among seasonal 2007 influenza A(H1N1) infected subjects: Results from the FLUREC cohort study.

INTRODUCTION Haemagglutination-inhibition (HI) antibody titer is a correlate of protection against influenza; its persistence after infection or vaccination is important to determining susceptibility to subsequent infection. Few studies, however, have reported longitudinal data regarding the magnitude and duration of HI protection following natural seasonal influenza A infection. METHODS Using French influenza cohort study data collected from 2008 to 2010, we investigated persistence of serological protection among subjects according to influenza-like illness (ILI) and laboratory-confirmed seasonal 2007 influenza A(H1N1) infection status at inclusion in 2008 (ILI-A(H1N1) positive, ILI-A(H1N1) negative, or no-ILI). Antibody titers against seasonal 2007 A(H1N1) were determined using the HI technique for sera. Regression models for interval-censored data were used to estimate geometric mean titers (GMT) for HI assays. A logistic regression model adjusted for age group (subjects <30, 30-50 and >50 years old) was used to quantify the association between HI titer and protection against infection. RESULTS Based on 310 total subjects, influenza A(H1N1) infection was confirmed in 39 of 115 ILI subjects at inclusion. GMT associated with 50% probability of protection among ILI subjects decreased with age group (subjects <30 yo: GMT of 40.8 was associated with 50% [95CI: 29.3%; 70.7%] probability of protection, subjects 30-50 yo: 26.8 [95CI: 34.4%; 65.6%] and subjects >50 yo: 8.9 [95CI: 15.3%; 84.7%]). GMT declined after the first annual study visit among ILI-A(H1N1) positive subjects but remained higher compared to inclusion at the 2010 study visit (41.5 [95CI: 34.8; 49.5], p=0.0157). GMT remained stable among ILI-A(H1N1) negative subjects (p=0.7502), but decreased among no-ILI subjects (p<0.0001). CONCLUSION Our results confirm the positive relationship between HI titer and probability of protection among naturally infected subjects, and provides evidence that protection associated with HI titer varies with age. This longitudinal analysis suggests the rise in HI titers following seasonal 2007 influenza A(H1N1) infection may persist into subsequent influenza seasons.