International epidemiology of human pre-existing adenovirus (Ad) type-5, type-6, type-26 and type-36 neutralizing antibodies: Correlates of high Ad5 titers and implications for potential HIV vaccine trials

Abstract

Replication-defective adenoviruses have been utilized as candidate HIV vaccine vectors. Few studies have described the international epidemiology of pre-existing immunity to adenoviruses. We enrolled 1904 participants in a cross-sectional serological survey at seven sites in Africa, Brazil, and Thailand to assess neutralizing antibodies (NA) for adenovirus types Ad5, Ad6, Ad26 and Ad36. Clinical trial samples were used to assess NA titers from the US and Europe. The proportions of participants that were negative were 14.8% (Ad5), 31.5% (Ad6); 41.2% (Ad26) and 53.6% (Ad36). Adenovirus NA titers varied by geographic location and were higher in non-US and non-European settings, especially Thailand. In multivariate logistic regression analysis, geographic setting (non-US and non-European settings) was statistically significantly associated with having higher Ad5 titers; participants from Thailand had the highest odds of having high Ad5 titers (adjusted OR = 3.53, 95% CI: 2.24, 5.57). Regardless of location, titers of Ad5NA were the highest and Ad36 NA were the lowest. Coincident Ad5/6 titers were lower than either Ad5 or Ad6 titers alone. Understanding pre-existing immunity to candidate vaccine vectors may contribute to the evaluation of vaccines in international populations.

Introduction

Adenoviruses are nonenveloped DNA viruses that have been extensively studied and used as vectors for gene delivery. Because adenoviruses elicit potent immune responses, recombinant adenoviruses have been used as vectors for several potential vaccines [1]. Diseases for which adenovirus vectors are being considered include HIV, tuberculosis, malaria, and cancer.

There are 6 subfamilies (A–F) of adenovirus serotypes. Adenovirus type-5 (Ad5) from subfamily C is the best-studied serotype and replication-defective Ad5 viruses are presently being studied as vectors for human immunodeficiency virus (HIV) vaccines [2]. Recombinant Ad5 vectors are efficient at priming cellular immune responses [3]. A potential limitation of Ad5 vector-based vaccines, however, is pre-existing immunity against adenoviruses resulting from natural exposure [4]. Naturally occurring neutralizing antibodies (NA) to the Ad5 vector may impair its ability to elicit desired immune responses after vaccination [5]. The presence of pre-existing immunity to Ad5 has been reported to have an effect on the immunogenicity of Ad5 vectors in mice [6], [7], rhesus monkeys [8], [9], and humans [10], [11], [12].

Although Ad5 is more commonly used as a therapeutic vector, the more rare adenovirus type-6 (Ad6), also from subfamily C, is an additional candidate vector that possibly could be used as an alternative or adjunct to Ad5. Limited published data suggest that NA against Ad6 are less prevalent and present at lower titers than NA against Ad5 [13], [14]. Furthermore, there appears to be no cross-reactivity between NA against Ad5 and Ad6 [14]. Additional rare serotypes from subfamily D offer conceptually interesting vaccine vector candidates because of reported low prevalence of antibodies [15], [16].

The prevalence of naturally occurring NA to the Ad5 vector may vary among populations. Approximately 30–60% of the United States population has Ad5 NA as a result of natural infection, based on the few available epidemiologic studies, whereas higher frequencies and titers may be found in other parts of the world [5], [17], [18], [19], [20]. Although published data exist on the international prevalence of antibodies to potential HIV vaccine viral vectors [14], [17], [18], [19], [20], [21], few large scale studies have been conducted and little is known about the factors that may be associated with Ad NA prevalence among populations likely to be enrolled in HIV vaccine efficacy studies.

Results from Merck's phase II adenovirus type 5 (Ad5) gag/pol/nef test-of-concept trial showed that the vaccine lacked efficacy against human immunodeficiency virus (HIV) infection in a high-risk population [22]. Although the role of pre-existing Ad5 immunity in the lack of efficacy is not well understood, the trial results highlight the importance of understanding vector-specific immunity and its impact on qualitative and quantitative parameters of the HIV-specific immune response [23].

In this study, we performed a standardized, international collection of serum samples to assess the seroprevalence, epidemiology and correlates of NA for selected adenovirus types in countries and sites where HIV vaccines, in general, either have already been assessed or where they may eventually be evaluated or implemented.