Tuberculosis Vaccine Development: Progress in Clinical Evaluation

Abstract

Tuberculosis (TB) is the leading killer among all infectious diseases worldwide despite extensive use of the Mycobacterium bovis bacille Calmette-Guérin (BCG) vaccine. A safer and more effective vaccine than BCG is urgently required. More than a dozen TB vaccine candidates are under active evaluation in clinical trials aimed to prevent infection, disease, and recurrence. After decades of extensive research, renewed promise of an effective vaccine against this ancient airborne disease has recently emerged. In two innovative phase 2b vaccine clinical trials, one for the prevention of Mycobacterium tuberculosis infection in healthy adolescents and another for the prevention of TB disease in M. tuberculosis-infected adults, efficacy signals were observed. These breakthroughs, based on the greatly expanded knowledge of the M. tuberculosis infection spectrum, immunology of TB, and vaccine platforms, have reinvigorated the TB vaccine field. Here, we review our current understanding of natural immunity to TB, limitations in BCG immunity that are guiding vaccinologists to design novel TB vaccine candidates and concepts, and the desired attributes of a modern TB vaccine. We provide an overview of the progress of TB vaccine candidates in clinical evaluation, perspectives on the challenges faced by current vaccine concepts, and potential avenues to build on recent successes and accelerate the TB vaccine research-and-development trajectory.

Keywords: clinical trials; desired attributes; developmental trajectory; natural immunity; new TB vaccines; tuberculosis.

FIG 1

Timeline of key milestones in the history of tuberculosis vaccine development and human use. BCG, bacille Calmette-Guérin; BMRC, British Medical Research Council; USPHS, U.S. Public Health Service; WHO, World Health Organization; MVA85A, modified vaccinia virus Ankara vector expressing antigen 85A of M. tuberculosis; M72:AS01_E, a recombinant fusion protein of M. tuberculosis 39a (Rv1196) and M. tuberculosis 32a (Rv0125) in the AS01_E adjuvant. *The WHO further updated guidelines on BCG vaccination of infants born to HIV-infected mothers in 2018. According to these guidelines, HIV-infected neonates should delay BCG vaccination until antiretroviral therapy (ART) has been started and they are immunologically stable. If HIV-infected individuals, including children, who are receiving antiretroviral therapy (ART) are clinically well and immunologically stable, they should be vaccinated with BCG. Furthermore, neonates with an unknown HIV status born to HIV-infected women should be vaccinated if they have no clinical evidence suggestive of HIV infection, regardless of whether the mother is receiving ART.

FIG 2

Current global clinical pipeline of TB vaccine candidates. The 2019 global clinical portfolio of TB vaccine candidates includes mycobacterial killed, whole-cell, or extract vaccine candidates (Vaccae, MIP, DAR-901, and RUTI); live-attenuated mycobacterial vaccine candidates (VPM1002, BCG revaccination, and MTBVAC); recombinant live-attenuated or replication-deficient virus-vectored candidates expressing an M. tuberculosis protein(s) (TB/FLU-04L, Ad5Ag85A, and ChAdOx1.85A/MVA85A); and a mycobacterial fusion protein(s) in an adjuvant formulation (M72:AS01_E, H56:IC31, ID93:GLA-SE, and GamTBvac). See the text for additional information on vaccine candidates. ID, intradermal; IM, intramuscular; Mtb, M. tuberculosis.

FIG 3

Tuberculosis vaccine and desired attributes. (A) Characteristics of an ideal TB vaccine. (B) An effective TB vaccine will likely need to engage multiple mechanisms and should aim to elicit a comprehensive immune response involving multiple arms of the immune system. Humoral and cell-mediated immunity may act at different points in time, or synergistically, to prevent M. tuberculosis infection and TB disease. Vaccine-elicited protective antibodies in the airways can prevent the establishment of infection or limit the acquisition of infection. Cytokines produced by airway-resident innate lymphocytes may equip alveolar macrophages with early bactericidal functions and recruit monocyte-derived macrophages to the site of infection. Yet M. tuberculosis, a “robust” intracellular pathogen, frequently succeeds in establishing a long-term infection in macrophages. Vaccine-elicited or trained innate lymphoid cells, NK cells, unconventional T cells, and T_RM cells at submucosa may act as sensory cells, recruit memory T cells, and/or act as early effectors to increase the kinetics of killing of M. tuberculosis-infected cells, leading to abortion of infection. Nevertheless, immune evasion strategies employed by M. tuberculosis likely present challenges for the prevention of infection, resulting in the establishment of infection in the interstitium. Induction of lymphoid follicles, as a local antigen presentation site, may be a desired feature of vaccination to reduce the bottleneck in delayed antigen presentation in draining LNs and impediment in the activation of T_RM and T_EM cells. Activated dendritic cells and other antigen-presenting cells during recall responses may rapidly initiate the activation of T_CM cells and memory B cells. TB vaccines will need to elicit long-lived memory T cells, and these memory T cells will need to rapidly expand and generate secondary effectors with a sustained proliferative and “functional” capacity. Primed effectors will need to be specific to critical antigens in the life cycle of M. tuberculosis, possess lung-homing potential, traffic to the infection site, recognize M. tuberculosis-infected cells, and resist terminal differentiation or exhaustion. Mucosal antibodies may prevent infection or reduce the severity of infection, host-damaging effects, and systemic dissemination. Effector T-cell responses must be capable of eliminating infection, or at least enforcing lifelong control of infection, while preserving delicate anatomical structures. This necessitates appropriately placed, tightly regulated, and highly balanced pro- and anti-inflammatory responses. Although pulmonary mucosal vaccination appears to be capable of inducing a protective local immune response, it must be safe for administration. (C) Desired attributes of immune responses (portrayed and listed in panel B) to be elicited by a “modern TB vaccine.” iNKT, invariant natural killer T cells; MAIT, mucosa-associated invariant T cells; HEV, high endothelial venules; FC, fragment crystallizable region of an antibody.

FIG 3

Tuberculosis vaccine and desired attributes. (A) Characteristics of an ideal TB vaccine. (B) An effective TB vaccine will likely need to engage multiple mechanisms and should aim to elicit a comprehensive immune response involving multiple arms of the immune system. Humoral and cell-mediated immunity may act at different points in time, or synergistically, to prevent M. tuberculosis infection and TB disease. Vaccine-elicited protective antibodies in the airways can prevent the establishment of infection or limit the acquisition of infection. Cytokines produced by airway-resident innate lymphocytes may equip alveolar macrophages with early bactericidal functions and recruit monocyte-derived macrophages to the site of infection. Yet M. tuberculosis, a “robust” intracellular pathogen, frequently succeeds in establishing a long-term infection in macrophages. Vaccine-elicited or trained innate lymphoid cells, NK cells, unconventional T cells, and T_RM cells at submucosa may act as sensory cells, recruit memory T cells, and/or act as early effectors to increase the kinetics of killing of M. tuberculosis-infected cells, leading to abortion of infection. Nevertheless, immune evasion strategies employed by M. tuberculosis likely present challenges for the prevention of infection, resulting in the establishment of infection in the interstitium. Induction of lymphoid follicles, as a local antigen presentation site, may be a desired feature of vaccination to reduce the bottleneck in delayed antigen presentation in draining LNs and impediment in the activation of T_RM and T_EM cells. Activated dendritic cells and other antigen-presenting cells during recall responses may rapidly initiate the activation of T_CM cells and memory B cells. TB vaccines will need to elicit long-lived memory T cells, and these memory T cells will need to rapidly expand and generate secondary effectors with a sustained proliferative and “functional” capacity. Primed effectors will need to be specific to critical antigens in the life cycle of M. tuberculosis, possess lung-homing potential, traffic to the infection site, recognize M. tuberculosis-infected cells, and resist terminal differentiation or exhaustion. Mucosal antibodies may prevent infection or reduce the severity of infection, host-damaging effects, and systemic dissemination. Effector T-cell responses must be capable of eliminating infection, or at least enforcing lifelong control of infection, while preserving delicate anatomical structures. This necessitates appropriately placed, tightly regulated, and highly balanced pro- and anti-inflammatory responses. Although pulmonary mucosal vaccination appears to be capable of inducing a protective local immune response, it must be safe for administration. (C) Desired attributes of immune responses (portrayed and listed in panel B) to be elicited by a “modern TB vaccine.” iNKT, invariant natural killer T cells; MAIT, mucosa-associated invariant T cells; HEV, high endothelial venules; FC, fragment crystallizable region of an antibody.

FIG 4

Distinct features of T-cell responses to TB vaccine candidates and disparities between responses in animal models and humans. TB vaccine candidates elicit CD4⁺ and CD8⁺ T-cell responses with distinct functional profiles, although these responses differ between animal studies and clinical trials. The sizes of the colored circles indicate relative magnitudes of specific T_H1, T_H17, and CD8⁺ T-cell responses induced. The quality of the response with the dominant cytokine-producing subset(s) is described. NI, not investigated; PNI, polyfunctionality not investigated; BDL, below the detection limit. In mice and NHPs, parental BCG induces polyfunctional T_H1 and T_H17 responses and affords partial protection against M. tuberculosis challenge (108, 199, 224). No correlation was found between the magnitudes or polyfunctional profiles of BCG-specific T cells and protection against pulmonary TB in South African infants (225). Interestingly, a BCG-specific IFN-γ ELISPOT response was found to be associated with a reduced risk of TB in the same settings (171). BCG-elicited immune responses and protection wane over time in settings where TB is endemic (3, 105, 107), but BCG revaccination improves protection against sustained M. tuberculosis infection in adolescents (9). The live BCG replacement vaccine candidates VPM1002, AERAS-422, and MTBVAC induce broad immune responses, including increased polyfunctional CD4⁺ and CD8⁺ T-cell responses and improved protection in mice relative to BCG (131, 133, 156, 226), but in clinical trials, polyfunctional T-cell responses did not differ significantly from those induced by BCG (8, 157–159). MTBVAC induces antigen-specific polyfunctional CD4⁺ T-cell responses at magnitudes that exceed those induced by BCG in infants, but no significant CD8⁺ T-cell responses are induced, and MTBVAC also results in IGRA conversion. BCG booster subunit candidates increase immunogen-specific CD4⁺ and/or CD8⁺ T-cell responses in mice and improve protection. However, inconsistent protection in NHPs (MVA85A, Ad35Ag85A/AERAS-402, Ad5Ag85A, and H1/H4/H56) (26, 155, 181, 182, 227, 228) and relatively low-level immune responses in BCG-vaccinated infants (MVA85A, AERAS-402, and M72) (165–167, 177, 229, 230) compared to those in adults were observed for some candidates. In clinical trials, MVA85A failed to improve protection against M. tuberculosis infection or disease despite increased polyfunctional T_H1 and T_H17 responses (6, 168), and AERAS-402-induced polyfunctional CD8⁺ and CD4⁺ T cells failed to recognize M. tuberculosis-infected target cells (180). Unlike BCG-elicited responses, adjuvanted subunit candidates predominantly induce IL-2-coexpressing polyfunctional CD4⁺ T-cell subsets that correlate with enhanced protection in mice (183, 188, 191), but the role of these subsets in human protection is unclear, and the relevance of varying immunogenicity of antigenic components in subunit vaccine candidates for protection is unknown (231, 232). MIP elicits beneficial effects in M. tuberculosis-infected animal models as a therapeutic vaccine, but in clinical trials, it provided ambiguous benefits in TB patients (196–198).