Persistence of Anti-West Nile Virus-Specific Antibodies Among Asymptomatic Blood Donors in Northeastern Italy

Introduction

In the summer of 2008, the first human case of West Nile virus (WNV) neuroinvasive disease (WNND) was recognized in Italy (Rossini et al. 2008). A serosurvey conducted among blood donors (BDs) living in the WNV-affected area of Ferrara (northeastern Italy) defined WNV seroprevalence as 0.69% (Pierro et al. 2011). This study is a serial follow-up of the 68 BDs with a positive immune response to WNV identified in our first study (Pierro et al. 2011). The first objective of this study was to evaluate the variation in antibody titers over time, and the second objective was to monitor variations in immunoglobulin G (IgG) avidity.

Materials and Methods

After the first detection of a specific immune response against WNV (Pierro et al. 2011), each BD was invited to give follow-up blood samples during subsequent donations. Each series comprised a variable number of sera. The time interval between the first and the second positive samples varied between 3 and 6 months; the third samples were obtained a mean 4 months after the second. Written informed consent was obtained from each BD before sampling. WNV-specific IgG and IgM and IgG avidity were measured as previously described (Pierro et al. 2011).

Results

BDs refusing to give follow-up samples (19 out of 68 IgG-WNV positive) were not included in the study. The IgG-WNV–specific antibody titer decreased in 74% of BDs within 12 months after the initial detection. The percentage decrease was higher in females (55.5%) than in males (27.5%); particularly, 47.5% and 11.1% of the seropositive male and female BDs, respectively, reverted to complete antibody negativity. These findings were more evident among subjects aged 30–50 years: 12/21 subjects (57.1%) showed a complete reversion to IgG WNV negativity. Furthermore WNV-specific IgG titer increased in a smaller percentage (10%), whereas 16% of BDs showed no titer variation.

The IgM response was detected at first testing in 4 subjects (Pierro et al. 2011), suggesting a recent infection: three of these samples became IgM negative at the second sampling, and the last sample showed no variation in IgM WNV antibody titer throughout the follow-up. These four samples showed decreasing levels of IgG (from 1:3200 to 1:400) between the first and final samplings, and all had an increased IgG avidity.

Subjects with only one positive sample (13 of 68 IgG WNV positive) were not included in the evaluation of IgG avidity. According to modifications in antibody titer, IgG avidity increased in 20 subjects (55.5%) and remained constant in 16 (44.5%). No subjects showed a decrease in IgG avidity. The average value of IgG avidity was 59.66% at the first sampling and 83.53% at the last; this was calculated considering the whole population of BDs, including those subjects showing no variation in the IgG avidity value. The delta percentage of IgG avidity increased by a mean value of 23.87% 1 year after first detection.

Discussion

Since the recent WNV outbreaks in Europe, several seroprevalence studies have been performed (Charrel et al. 2001, Pfleiderer et al. 2008). The variation in antibody titers over time has been evaluated in patients with acute WNND (Busch et al. 2008) or among viremic BDs (Prince et al. 2007). Data from these studies differed depending on the characteristics of WNV infection (Southam et al. 1954) and the dynamics of the antibody response (Busch et al. 2008, Prince et al. 2007). These discrepancies are likely generated by the methods used and by the clinical conditions of the populations studied. To date, the persistence of WNV antibodies has not been assessed among a population of asymptomatic nonviremic BDs.

This study disclosed a decrease in antibody titer until negativity in a high percentage of cases over a 12-month period after first detection. One major bias in these results could be the imprecise identification of the seroconversion date in most subjects. Given the large proportion of BDs that showed an increased IgG avidity value along the study period, thus clearly indicating a recently acquired infection, and the circulation of WNV in the studied area, we consider that the large majority of this population was infected over a 1-year period before the first detection. Our data also showed a mean 23.87% increase in IgG avidity during the study period, suggesting that most of the first samples were associated with a likely recent infection because IgG WNV avidity has been reported to increase ∼30–60 days after onset of infection (Fox et al. 2006). Our study further suggests that in a population of asymptomatic nonviremic BDs, antibody titers against WNV decrease about 12–18 months after infection. Thus, the population geographically exposed to WNV could be at risk of repeated infections that could generate a new clinical disease.

Our findings differ from previous reports (Prince et al. 2005, Prince et al. 2007). This could be explained by the hypothesis that host factors hampered the development of clinically evident disease in this population and could also modulate the humoral immune response against WNV. This difference could also reside in virus-specific factors, but there was an elevated genetic homology between different WNV isolates obtained in 2008 and 2009 from patients living in the study area (Rossini et al. 2011). The small number of IgM- and IgG-positive subjects could be explained by the long persistence of the first class of immunoglobulins among viremic BDs (Prince et al. 2005). Our study may indicate that people seroconverting against WNV without developing clinically detectable disease are likely to be underprotected against additional exposure to infected mosquitoes within a short time frame, because many of these subjects reverted to seronegativity in about 12 months. This fact and the knowledge that both virus and host characteristics may vary over time suggest that surveillance activity should be maintained year after year to have updated epidemiologic data on WNV circulation and thereby prevent WNV disease. Our results should also be considered in the anti-WNV vaccination strategy.