Human type 5 adenovirus-based tuberculosis vaccine: is the respiratory route of delivery the future?

Abstract

Despite progress in managing TB, there were 8.6 million new cases in 2012. To control TB will require a more effective vaccine than BCG, new drugs and better diagnostic tests. Recombinant replication-defective adenoviruses expressing foreign DNA have been studied as vaccines. We developed and evaluated a recombinant replication-deficient human Ad5 vector expressing Ag85A (Ad5Ag85A) as a TB vaccine in animal models and a Phase I human study. Animal models of Ad5Ag85A show markedly improved protection over BCG alone and immunization via the respiratory route provides the best type of protection. In humans, intramuscular vaccination was safe; Ad5Ag85A was immunogenic and stimulated polyfunctional T cell responses, more potently in previously BCG-vaccinated volunteers. Pre-existing Ad5 antibodies did not dampen the response. Given its potency, Ad5-based TB vaccines are well-positioned to be delivered to the respiratory tract, induce local lung immunity to control TB, and inform innovative approaches to new TB vaccination strategies.

Keywords::

• Ad5Ag85A
• adenovirus
• CD4⁺
• CD8⁺
• Mycobacterium tuberculosis
• pulmonary
• TB
• vaccine

Despite progress being made to meet the WHO’s global TB targets, still an estimated 8.6 million new cases of TB have been reported from the year 2012, with 1.3 million deaths. The management of multidrug-resistant TB and the care available for TB-HIV co-infected persons falls far short of the Millennium Development Goals [1]. To effectively control TB, there is a requirement for more effective TB vaccines, new drugs and better diagnostic tests [2]. For almost 70 years, BCG, an attenuated strain of Mycobacterium bovis, has been the only TB vaccine in use and, while protecting children against disseminated disease, it fails to control adult pulmonary TB. Mathematical modeling has demonstrated that a new vaccine with 60% efficacy for preventing active disease and a mass vaccination strategy could lead to an 80% decrease in TB incidence by 2050 [2].

Gaps remain in our understanding of the immunology of TB infection and the correlates of immune protection, but enough is known about the immune response to Mycobacterium tuberculosis (M.tb) to inform current vaccine development [3]. Like other intracellular pathogens, T lymphocytes are essential for the immune response to M.tb. Following inhalation of M.tb, alveolar macrophages and dendritic cells engulf the organism, transporting it to the lymph nodes where T lymphocytes are activated to initiate Th1 and Th17 cytokine responses, leading to activation of macrophages and neutrophils. There is, however, a delay in the onset of adaptive immune responses to M.tb allowing M.tb to establish a foothold in the lung. In humans, skin test reactivity does not appear until 42 days after infection and in mice the appearance of anti-TB Th1 cells in the lung is delayed for up to 14 days [3]. Respiratory mucosal vaccination helps close the immunologic gap in mucosal protection in the early phases of M.tb infection.

With results reported recently of the first large-scale human trial of a TB vaccine since 1968 [4], momentum is building in the field of TB vaccine research and the TB vaccine pipeline now has at least a dozen products in clinical trials, designed either to replace BCG with an improved organism-based vaccine or boost BCG primed responses [5]. Disappointingly, despite having been shown to induce durable CD4⁺ T-cell responses, intradermal injection of MVAAg85A, a recombinant modified vaccinia virus expressing Ag85A, in BCG-immunized infants failed to improve protection against active TB over BCG alone [4].

Recombinant replication-defective viral vectors expressing foreign DNA have been developed as vaccines, with adenoviruses being particularly robust as a vaccine platform [6]. There are more than 50 human serotypes, with adenovirus type 5 (Ad5) being one of the most common and the most studied gene transfer vectors, both in cancer and infectious diseases.

Adenoviruses are highly immunogenic, activating both the innate and adaptive immune systems. The adenoviral backbone is a type 1 immune adjuvant enhancing the immune response toward the foreign antigen and of all vector platforms, it derives the strongest transgene expression with continuously raised levels of protein in vivo for up to 3 weeks. The adenoviral vector triggers potent transgene product-specific CD8⁺ T-cell responses and to a lesser extent CD4⁺ T-cell responses. Ad5-based vaccines have been developed for several infectious diseases, including malaria, HIV, influenza and Ebola virus disease, with demonstrated safety, immunogenicity and efficacy [6].

We have developed and evaluated a recombinant replication-deficient human Ad5 vector expressing Ag85A (Ad5Ag85A), an immunodominant antigen produced by all mycobacterial species, as a novel TB vaccine in different animal models and recently in a Phase I study in humans. In a murine model, intramuscular immunization with Ad5Ag85A induced robust antigen-specific T-cell responses in the spleen and lung interstitium, although did not provide protection against pulmonary M.tb challenge [7]. The peripherally distributed T cells activated by intramuscular Ad5Ag85A immunization were, however, functional as upon adoptive transfer into the airway of naïve severe combined immunodeficiency mice, they were capable of immune protection [8]. Respiratory mucosal administration of Ad5Ag85A elicited lower systemic levels of antigen-specific T-cell responses, but elevated T-cell responses in the lung, which correlated with improved protection following pulmonary challenge [7]. Mucosal immunization with Ad5Ag85A elicited a significant CD8⁺ T-cell population in the airway lumen, which was absent with intramuscular immunization [8]. These CD8⁺ T cells were of a long-lasting effector memory phenotype and were capable of self-renewing via in situ proliferation in a specific Ag-dependent manner [9] and remained immune-protective even in the absence of CD4⁺ T cells [10]. Our data support the concept that following immunization with an adenoviral TB vaccine, the geographical distribution of T cells is closely associated with the route of immunization and immune protection against pulmonary M.tb infection.

Ad5Ag85A vaccine has been evaluated in calves and intramuscular, intradermal or endobronchial administration is safe, activates both CD4⁺ and CD8⁺ cells and protects against virulent M. bovis infection [11]. In a guinea pig model, animals primed with BCG and then boosted intranasally with Ad5Ag85A significantly outlived those vaccinated with BCG or BCG with intramuscular boosting with Ad5Ag85A after pulmonary M.tb challenge [12], and goats primed with BCG and boosted with Ad5Ag85A had reduced lung pathology and significant reductions in bacterial load when challenged with Mycobacterium caprae [13].

The pre-clinical safety and immunogenicity of Ad5Ag85A formed the basis of our recent Phase I clinical trial [14]. We evaluated the safety and immunogenicity of a single intramuscular injection of 10⁸ pfu Ad5Ag85A in 12 BCG-naïve and 12 previously BCG-immunized healthy adults. Vaccination was safe and well tolerated in both groups, with most side effects related to grade 1 injection site reactions. Ad5Ag85A was immunogenic in both groups and stimulated polyfunctional T-cell responses, but it more potently boosted both CD4⁺ and CD8⁺ T-cell immunity in previously BCG-vaccinated volunteers compared with BCG-naïve volunteers, supporting its further clinical development as a boost vaccine after BCG priming.

Pre-existing immunity to Ad5 due to natural infection, observed in up to 90% of populations, may reduce uptake of the adenovirus vector and limit transgene expression leading to suboptimal immune responses [15]. Using human adenovirus of rare serotypes or non-human primate-derived adenovirus has been proposed to circumvent this problem and the results of Phase II clinical trials of a human Ad35 TB vaccine (Crucell Ad35/AERAS-402) [16] are awaited. Compared with Ad5, however, Ad35 is much less immunogenic [17]. We found low-to-medium levels of pre-existing anti-Ad5 humoral immunity in most of our study volunteers, but there was no evidence that pre-immunization anti-Ad5 immunity significantly dampened the potency of Ad5Ag85A vaccine nor was the level of Ad5 antibody associated with increased adverse effects [14]. Emerging evidence, however, suggests that the potential negative effect of Ad5 neutralizing antibodies could be avoided if the Ad5 vector was administered directly into the respiratory tract [18].

Recent HIV vaccine clinical trials using Ad5 vectors expressing three HIV proteins were stopped early because of futility [19–21] and in addition an increase in HIV infection was observed in men with high pre-existing Ad5 antibody titers in the Step study [19]. However, a similar relationship was not observed in the Phambili trial, where the same Ad5 HIV vaccine was tested in Africa [21]. It has been argued that Ad5-induced activation of CD4⁺ T cells, the specific target for HIV, may lead to increased HIV infection, but if this were the case, it may also apply to any HIV vaccine or other Ad vectors [22]. With the overlapping epidemics of HIV and TB, especially in sub-Saharan Africa, whether or not an Ad5 TB vaccine increases the risk of HIV acquisition should be investigated [22]. Although Ad5-based vaccines for HIV are no longer appropriate, the effort in developing Ad5-based vaccines against other infections needs to continue.

The lung is the primary target for M.tb infection. There is convincing evidence from animal studies that any effective immunization strategy for M.tb must induce a memory T-cell population within the lung that can inhibit the growth of M.tb immediately after the bacilli laden droplets have been inhaled [23]. We argue that given its unrivalled potency, an Ad5-based TB vaccine is well positioned to be delivered to the respiratory tract, induce the local lung immunity needed to control TB and inform rational TB vaccine design. We have characterized the properties of the aerosol droplets of Ad5Ag85A generated by nebulization in preparation for human clinical studies, where Ad5Ag85A will be administered by inhalation for evaluation of safety and immunogenicity. FluMist^® and measles vaccines have been safely administered to humans by intranasal or inhaled aerosol delivery and measles vaccine delivered by the respiratory route is at least as immunogenic as subcutaneous administration [24,25]. Animal models of Ad5Ag85A have, unlike MVAAg85A, shown significantly improved protection over that by BCG alone and provided convincing evidence that immunization via the respiratory route provides the most desirable type of immune protection. Further studies with Ad5Ag85A or a multivalent version can do much to inform innovative approaches to new TB vaccines with the ultimate goal of controlling TB.

Financial & competing interests disclosure

Studies from the authors’ laboratory were supported by funds from the Canadian Institutes of Health Research, the Natural Sciences and Engineering Research Council of Canada, the Canadian Foundation for Innovation and the Michael G DeGroote McMaster Institute for Infectious Diseases Research. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.