Challenge of developing new tuberculosis vaccines to generate life-long protective immunity

A major hurdle in the eradication of important human pathogens is the development of vaccines to combat the three big killers: HIV, tuberculosis (TB) and malaria. While several factors could be attributed to the difficulty of developing vaccines against these diseases, a common theme is the notion that effective control of all three diseases is dependent upon sustained generation of T-cell immunity. In the case of TB, this is quite a task. TB is a chronic disease; the causative agent Mycobacterium tuberculosis is highly persistent and possesses multiple strategies to evade immune clearance, and individuals can be exposed many years after vaccination. Furthermore, while generating long-lived humoral immunity to protect against acute viruses and extracellular bacteria has proven successful, antibodies appear to play a limited role in controling TB, and sustained protective T-cell responses (ideally both CD4⁺ and CD8⁺) may be required to protect against this pathogen. In order to achieve this, most vaccine approaches aim to replicate, in a safe form, the T-cell immunity generated by natural infection. While this strategy seems to work well for other viral and bacterial pathogens, TB-induced immunity is unusual, as primary infection does not necessarily protect against reinfection [1]. Furthermore, the bacillus Calmette–Guérin (BCG) vaccine, the only approved TB vaccine in use, probably does not facilitate the generation of sufficient memory T-cell responses to protect against long-term infection (see later). Therefore, the challenge is to develop new vaccines that deliver superior protective immunity than that induced by either M. tuberculosis or the BCG vaccine. In this editorial, we aim not to summarize the current vaccine strategies being employed against TB, but rather to offer our perspective on what we consider to be the best way of addressing the issues concerning the generation of long-lived anti-TB immunity.

Requirements for the generation of sustained memory T-cell responses

The generation of immunological memory to infectious pathogens is a major function of the adaptive immune response. Memory T cells are generated by the encounter of naive cells with their cognate antigen in the lymph node, which initiates the clonal expansion and differentiation of these cells [2]. These antigen-specific memory cells are then able to migrate to peripheral tissue and respond rapidly to secondary infection by acquiring effector functions and producing effector molecules. The capacity to induce T-cell responses to mycobacterial pathogens is dependent on the level of antigen present at the time of T-cell priming [3,4], and continual antigenic stimulation is most likely required to maintain memory CD4⁺ T cells, which are the most critical subset required for anti-TB immunity. How then does the generation of T-cell memory relate to the efficacy of the BCG vaccine? First, field trial data indicate that, in certain situations, BCG can induce good levels of protective immunity, which seems to be dependent on the age at which M. tuberculosis is encountered postvaccination and other factors such as geographical location and genetic makeup of the vaccinated population [5]. It has also been shown in mice that the antigens shared by BCG and M. tuberculosis induce a similar level of protective immunity in short-term studies of vaccine efficacy [6]. These data suggest that BCG expresses the necessary antigenic repertoire to induce protective immune responses, yet it is the quality and longevity of the memory T-cell response induced by BCG that is not sufficient to maintain protection long term. Below we present our thoughts on how BCG-induced immunity might be improved to confer such long-term protection.

Giving BCG a boost

In animal models, BCG is cleared relatively quickly from mouse organs after vaccination, in contrast to the persistent nature of M. tuberculosis[7]. This suggests that BCG may not persist long enough in the host to stimulate an adequate population of memory T cells to confer long-term protection. This experimental finding reflects clinical trial data, where BCG can protect children against TB up to approximately 15 years of age, after which protective immunity appears to wane [8]. The obvious approach to overcome this problem would be to ‘boost’ the immunity imparted by BCG given at birth. Delivering a second dose of BCG later in life would appear the most likely way to do this; however, data from field trials indicate that repeat BCG immunization does not significantly improve efficacy, and repeat vaccination is not recommended by the WHO [9]. One issue here is that the immune response generated after the initial BCG vaccination may be sufficient to clear the second BCG inoculum, rather than boosting immunity. This dilemma may be overcome by using a heterologous vector to deliver protective antigens. Indeed, a boosting vaccine comprising a modified vaccinia virus expressing M. tuberculosis Ag85A (MVA-85A) is the most well-developed TB vaccine currently in human trials [10,101], and other boosting vaccines have shown some promise in animal models and have entered Phase I trials [11]. One crucial issue relates to the efficacy of the boosting agent in maintaining memory T-cell responses. While MVA is a safe vaccine vehicle, it is rapidly cleared from the host [12], and it is unclear how this may influence the longevity of the boosting effect. New-generation vaccines may include viral delivery vectors that result in persistent antigen production in the lung, such as recombinant adenovirus vectors [13]. Nonetheless, ongoing clinical trials with the current boosting agents have certainly set the platform to facilitate and accelerate the assessment of future vaccine candidates.

Turning on the immune system

An alternative approach to boosting BCG immunity is to manipulate the response at the time of vaccination, in order to prolong the quality and longevity of memory T-cell responses. As one could consider BCG an effective ‘pool’ of protective antigens, modifying the vaccine to effectively deliver these protective antigens to the immune system appears to be a suitable strategy. Promising vaccines include BCG engineered to target dendritic cells in order to enhance T-cell priming [14], or BCG secreting the cytokine IL-15 that improves the generation of memory CD8⁺ T-cell populations [15]. Modifying BCG to alter the cellular processes of apoptosis or autophagy to induce T-cell activation are other strategies that have shown some potential [16,17]. As these vaccines function by manipulating immune processes, which could potentially result in detrimental immunopathology, the present challenge is to determine whether vaccine efficacy and safety can be transferred from the encouraging results in animal studies to use in human populations.

For T-cell vaccines, the key to affording optimal protective efficacy is to find the ideal balance of antigenic stimulation to generate effective memory T-cell populations. Antigen dose appears to dictate the ‘fitness’ of memory CD4⁺ T cells, with high or low levels of antigen resulting in cells displaying limited survival following secondary expansion after antigenic restimulation, while intermediate levels of stimulation results in memory T cells with high expansion potential [18]. Extrapolated to the TB situation, it is possible that high levels of persisting antigen, such as that occurring during TB infection, may compromise the effective generation of CD4⁺ T-cell memory responses. Conversely, low-level nonpersisting antigen, which could be considered the case after BCG vaccination, may not provide adequate levels of antigen to generate the right ‘quality’ of memory T-cell populations. So do we need ‘persistent’ vaccines to achieve the right type of immunity to control TB? Evidence from the literature suggests this may indeed be the case. BCG expressing the RD1 region from M. tuberculosis, which is more virulent than parental BCG in mice, was a more effective vaccine in both mice and guinea pigs [19,20]. Likewise, M. tuberculosis strains which are modified to be less virulent, yet persist in the host, display improved protective efficacy compared with BCG [7,21]. Therefore, modifying the immunizing dose and persistence of TB vaccines may be crucial in determining their long-term effectiveness. However, while persisting vaccines may be a solution in terms of fulfilling the immunological requirements for optimal protective immunity, their capacity to induce protection in the absence of overt pathology is the major obstacle that must be overcome before consideration for clinical trials in humans.

Acknowledgements

The authors would like to thank Nicholas West for helpful discussion.

Financial & competing interests disclosure

James A Triccas is supported by the NHMRC and is a Career Development Award holder. Jonathan K Nambiar is supported by an Australian Postgraduate Award. 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.