Tuberculosis vaccine pipeline

Claire Wingfield

A vaccine that can safely and effectively protect infants, children, and adults, regardless of HIV status, against pulmonary and extrapulmonary tuberculosis (TB) will be required if we are ever to eliminate tuberculosis as a global public health threat. A 2009 study on TB incidence using mathematical modeling estimated the impact of new TB vaccines, diagnostics, and treatments found that a novel preexposure vaccine given to infants that was 60% effective would have the most significant impact—reducing TB incidence approximately 80% by 2050 (Abu-Raddad 2009).

BCG: the current TB vaccine strategy

Mycobacterium bovis, descended from Mycobacterium tuberculosis (MTB), causes a TB-like disease in cows and humans. Starting in 1908, professors Albert Calmette and Camille Guérin began culturing M. bovis in order to weaken the bacteria to the point at which it was unable to cause disease but could stimulate an immune response in humans against MTB. The idea was to train the immune system to produce cells that would fight TB by introducing an attenuated (non-disease-causing) strain of a similar mycobacterium. The smallpox vaccine is, in fact, attenuated cowpox (vaccinia). In 1921—after 11 years of attenuation—the first human received the Calmette-Guérin (BCG) vaccine, and since that time it has become the most widely administered vaccine in the world. BCG provides protection against tuberculous meningitis and military TB in infants and young children up to perhaps five years of age, but its efficacy wanes over time and most vaccine-induced immunity appears to be gone by adolescence; revaccination later in life provides no benefit. It has been hypothesized, but never proven, that the natural presence of non-tuberculosis mycobacteria in tropical environments may decrease the effectiveness of BCG.

BCG’s use is even more challenging in high-HIV-prevalence settings because it can cause a severe immune reaction in HIV-infected infants. BCGitis (local infection) or BCGosis (systemic disease) are not well characterized complications of BCG vaccination that cause significant morbidity in multiple organs among HIV-infected infants and young children. The incidence of BCG disease is unclear, but it is a leading cause of death in TB/HIV-coinfected infants in South Africa (Zar 2007). In addition to the risks for developing BCG disease, evidence suggests that BCG provides little to no protection for HIV-infected infants (Mansoor 2009). The World Health Organization (WHO) revised its guidelines to recommend that infants with a confirmed HIV diagnosis should not receive BCG vaccination. This recommendation is impossible to implement in many high-burden settings because of limited capacity to rapidly confirm diagnosis using HIV RNA testing.

Despite its variability and limitations, it is estimated that the BCG vaccine saves the lives of over 40,000 children annually (of over 100 million vaccinated). This makes the decision of whether or not to vaccinate HIV-exposed children where HIV RNA testing is not available a challenge for both parents and health care providers. Without being able to confirm an HIV diagnosis using RNA testing soon after birth, parents and their health care providers are forced to weigh the risks (BCG disease) versus the benefits (protection against severe forms of pediatric TB) of BCG vaccination.

Policy makers, clinicians, and researchers are struggling with how to use BCG more effectively and safely and what alternative strategies can be used in its place. There are a few vaccines in preclinical and early-phase clinical studies that may be replacements for BCG. But these constructs are years away from efficacy testing. Delaying BCG vaccination by providing isoniazid preventive therapy (IPT) to HIV-exposed infants as preexposure prophylaxis until HIV status can be confirmed, as well as early initiation of antiretroviral treatment (ART), may be ways to address the challenges of BCG in HIV-exposed infants, but each strategy comes with implementation challenges. HIV RNA testing is not part of regular clinical practice in many high-TB-burden countries and it is unclear how long BCG vaccination can be delayed; studies are underway that hope to answer this question. Because it is difficult to bacteriologically confirm TB diagnosis in infants and young children, and the fact that pediatric TB is often an indication that a close adult contact is sick, IPT is a significantly underused strategy due to the fear of promoting isoniazid resistance. While great progress has been made in scaling up access to HIV treatments, a majority of those in need are still waiting for treatment; therefore, getting infants into treatment earlier may be a significant challenge in places where many people have been waiting for years to get ART.

Because BCG is part of the WHO’s Expanded Program on Immunisation schedule of vaccines it is administered in conjunction with a host of other vaccines (including those for measles, polio, and tetanus) throughout the world. Little explanation is given to parents about the vaccines, and many people misunderstand the limitations of protection that BCG provides. Some may incorrectly believe that because they were vaccinated against TB that they are protected against all forms of the disease for their entire lifetime. The lack of community understanding of the limitations of BCG is a major obstacle in creating community demand for a newer, better, and safer TB vaccine.

Challenges for TB vaccine research

TB has three phases—infection, latency, and disease—with different host (human) and pathogen (TB bacterium) factors influencing each phase. It is unclear what factors are associated with the establishment of latent infection and reactivation. The interactions among host, environment, and pathogen are dynamic and the contribution of each factor to the persistence of the bacteria is not well understood (Dye and Williams 2010). The TB bacterium exists in different metabolic states depending on whether it is infecting, latent, reactivating, or spreading disease throughout the body. During acute infection the bacterial load is high due to rapid replication. Once the infection becomes latent the bacterial load remains relatively stable and is confined within tubercules or granulomas—immunological prisons—in immunocompetent hosts. Attacking TB in latency likely requires a different mechanism of action than what would be used in early infection. When TB enters latency, it changes its metabolism and gene expression and therefore requires a different vaccine-induced immune response to prevent reactivation (Beresford and Sadoff 2010; Russell 2010).

One of the major challenges of TB vaccine research—in addition to our failure to fully understand the full spectrum of the disease—is our inability to predict the level and durability of vaccine-induced immunity. Defined correlates of immunity—the level of protection provided by vaccines —are critical for measuring vaccine efficacy and getting regulatory approval. There are no validated correlates of protection for TB vaccines (Beresford and Sadoff 2010; Wallis 2010). Because clinical signs and symptoms for TB can be difficult to assess and may not be TB-specific, it is challenging in infants and young children to rely on clinical endpoints for assessing efficacy of a TB vaccine (Hanekom 2010). However, it is certain that until we better understand the disease, clinical endpoints will be required.

Much of the data used to determine which vaccines to test in humans is based on what is observed in animals. There are a variety of models—mouse, guinea pig, rabbit and nonhuman primate—used to assess the impact of experimental drugs and vaccines before testing them in humans. Each has their advantages (e.g., cost, ability to manipulate the animal’s immune system) and disadvantages (e.g., generalization to humans). The most commonly used model is the mouse because of its relative cheap cost and, like the guinea pig, it can be inbred to emphasize certain genetic characteristics (Neurmberger 2010). Other models such as the rabbit and nonhuman primates may exhibit a more complete spectrum of TB disease (Beresford and Sadoff 2010; Neurmberger 2010; Russell 2010). But none of these animals exactly replicate TB disease in humans, so extrapolation of the data to humans is limited.

Another major challenge that threatens to delay the approval of a new TB vaccine is the lack of capacity to conduct large-scale phase III efficacy trials (Kaufmann 2010). Because a vaccine trial must show impact on the population level, thousands of study volunteers are required to demonstrate that vaccination with the experimental vaccine significantly reduces TB incidence in the community and that the reduction is durable. As a result, these studies require large sample sizes and longer follow-up than clinical trials that evaluate new drugs. Conducting these studies is labor and resource intensive and few research institutions have experience conducting studies of this magnitude.

There are efforts to build vaccine site infrastructure, but currently only the South African Tuberculosis Vaccine Initiative has the capacity to carry out phase III vaccine studies (Kaufmann 2010). The Aeras Global Vaccine Foundation (Aeras) and the European and Developing Countries Clinical Trials Partnership are supporting capacity building at sites in Africa and Asia, but it is likely that only one or two sites will be capable of conducting phase III studies before any of the current vaccines in phase II are ready to enter later-stage studies (Hanekom 2010; Kaufmann 2010).

The TB pipeline

Current TB vaccine candidates are designed to contain the TB bacillus by enabling the immune system to get a head start when exposed, reducing bacterial load and preventing progression to clinical disease (Kaufmann 2010; Russell 2010). No current TB vaccine candidate is designed to produce sterilizing immunity or, in other words, to prevent infection altogether. Rather, current constructs are designed to stimulate immune cell response to, and memory of, TB to prevent disease.

The vaccine candidates farthest along in the pipeline aim to replace BCG or strengthen BCG-induced immunity. Some “boost” or strengthen the initial immunity induced by BCG (and perhaps eventually a superior BCG alternative) and prevent progression to TB disease. The prime-boost strategy involves an initial immunization with a priming vaccine (currently the only prime being used is BCG but others are in the pipeline) that introduces the immune system to TB. A booster vaccine that broadens and strengthens the TB-specific immune response then follows the prime. There are live mycobacterial vaccines that improve BCG by adding genes or are attenuated MTB strains that have the genes deleted that are responsible for virulence (ability of a pathogen to cause disease (Kaufmann 2010; Russell 2010). Viral vectored vaccines are viruses modified so that they are unable to cause disease but are recombined to express TB-specific proteins. Immune cells recognize the TB genetic material and mount an immune response. Viral vectors have been used safely and effectively in many vaccines, including hepatitis B and human papilloma virus. The other vaccine strategy that has been evaluated is a therapeutic vaccine that is meant to improve response to TB treatment in people with active TB disease.

While a number of vaccine constructs will be entering phase II studies in the coming year, this year’s vaccine pipeline report is focused on constructs that have already entered phase II studies. It is expected that in the next few years this report will grow if TB vaccine research is adequately resourced to enable the conduct of later stage efficacy studies.

TB Vaccine Constructs in Phase II Clinical Trials

Agent Strategy Type Sponsors Status
M72 Prime boost Recombinant protein GSK Biologicals/Aeras Phase II
AERAS-402/Crucell Ad35 Prime boost Viral vector Crucell N.V./Aeras Phase IIb
MVA85A/AERAS-485 Prime boost Viral vector University of Oxford/Aeras Phase IIb

GSK Biologicals, a subsidiary of GlaxoSmithKline, is working with Aeras to conduct phase II studies of GSK M72, a recombinant protein vaccine. The vaccine is made up of an adjuvant—a molecule that stimulates an immune response—and two recombinant TB proteins meant to strengthen the immune response to two highly immunogenic fragments of the TB bacillus. To date, GSK has conducted phase I and II trials of the candidate vaccine in TB-naive, TB-infected, BCG-vaccinated, and HIV-positive adults. Safety and immunogenicity trials have been conducted in the United States, Europe, South Africa, and the Philippines. Early results suggest that the vaccine is clinically well tolerated and produces a measurable immune response. Subsequent clinical trials are now planned for adolescents and infants in TB-endemic regions (Ofori-Anyinam 2010).

AERAS-402/Crucell Ad35 from Crucell NV and Aeras is an adenovirus 35 (Ad35) modified to include specific TB antigens to trigger an immune response. A series of phase I studies have demonstrated TB-specific CD4 and CD8 responses in BCG-naive and BCG-vaccinated adult volunteers after receiving the vaccine. A phase II clinical trial in adults recently treated for pulmonary TB and a phase I study in infants are ongoing. A phase IIb randomized, placebo-controlled proof-of-concept study in HIV-positive adults with CD4 counts above 350 was recently initiated to evaluate the safety and efficacy of AERAS-402/Crucell Ad35 (Wooley 2010). There is a concern that for individuals who have preexisting antibodies to Ad35 the adenoviral vaccine may be less effective. The prevalence of antibodies varies geographically from approximately 5% to 20% (Hanekom 2010; Kaufmann 2010).

MVA85A/AERAS-485 is a live viral-vectored vaccine that is an attenuated version of the vaccinia virus—the cowpox virus that confers immunity to smallpox—combined with TB antigen 85A. The first infant received a dose in July 2009 as part of a phase IIb proof-of-concept study. This is the first time in over 80 years that a vaccine has been tested for efficacy in infants (Beresford and Sadoff 2010). The trial is comparing MVA85A versus placebo in BCG-vaccinated, HIV-negative infants. The first results are expected in 2012 (McShane 2010).

Mycobacterium vaccae is a mycobacterium which has been evaluated as an immuno-therapeutic vaccine for people with TB infection. In the 2009 pipeline report, it was reported that Aeras’s external Vaccine Selection Advisory Committee had reviewed the data from the Dar Dar study—a trial that evaluated M. vaccae in HIV-positive adults who had been vaccinated with BCG—and recommended that Aeras determine if new M. vaccae vaccine could be manufactured since the trial depleted the existing supply. Aeras has undertaken some limited process-development work to produce more vaccine and this work is almost complete. At this time Aeras does not have any immediate plans for further involvement (Willingham 2010). The limited data on M. vaccae are uninspiring. A 2003 Cochrane Review review concluded that M. vaccae provided no immunotherapeutic benefit for people with TB and therefore that no further trials were warranted (de Bruyn 2003). However, evidence from the Dar Dar study has suggested that a multidose M. vaccae vaccination was associated with protection against TB disease in people with HIV with CD4 counts above 200 (von Reyn 2010). The Dar Dar study results would need to be confirmed via additional studies before any conclusions could be made about effectiveness (Kaufmann 2010). M. vaccae is the only vaccine candidate to make it to phase III, but it appears that there is neither supply of the construct nor any research institution evaluating it at this time.

What is needed?

There are still many unanswered questions about the TB life cycle, the spectrum of TB infection and disease, and the impact of host genetics on the immune response that hamper vaccine development. More attention and resources need to be focused on basic scientific research. This is critical to keeping the pipeline full of new candidates, improving existing prevention tools, and identifying novel strategies to induce safe and durable immunity to TB infection and disease. A clear understanding of the differing characteristics of TB in its latent and disease state could lead to the development of a vaccine construct that could prevent infection and thereby significantly lower future cases of TB disease. The identification and validation of correlates of immunity will be vital to expediting the evaluation of any new vaccine candidate. Without the ability to predict whether a vaccine is able to induce an adequate protective immune response and to measure the quality of that response, massive resources—which are not currently available—will need to be dedicated to conducting long-term, large-scale epidemiological and efficacy studies that could significantly delay the approval of new vaccines for years. At the same time, a better understanding of the limits of BCG protection would help to identify alternatives to its use in HIV-exposed infants.

In addition to these basic and clinical research questions, a number of operational research issues need to be evaluated. Health systems need to be strengthened to provide access to comprehensive diagnostic and treatment options, including HIV RNA testing and IPT. Implementation research would provide examples of how programs could scale up HIV and TB diagnostic and treatment services to reduce the risk of BCG disease in HIV-infected infants and provide alternatives for the prevention of latent TB infection. These studies could also ensure that HIV-exposed, uninfected infants can benefit from BCG vaccination. For too long vaccine research has been the exclusive domain of immunologists and vaccinologists; social scientists and operational researchers should be included in setting research priorities and providing evidence to policy makers and clinicians on how to implement new vaccines.

In this spirit, researchers need to work collaboratively and share data. Aeras—a nonprofit product development partnership that works with vaccine manufacturers from the private sector to test and bring constructs to licensure—and a consortium of 31 research sites in Europe and Africa called the Tuberculosis Vaccine Initiative are organized to bring different institutions working on vaccine development together. However, this may be challenging for individual researchers or small-scale research institutions conducting basic science research that may not be connected to the broader vaccine research community. In order to build upon one another’s work and to avoid overlap, more opportunities for partnership and data sharing must be created.

Vaccine developers and researchers need to collaborate with communities and policy makers to create demand for a better vaccine and improved prevention strategies. Communities and policy makers’ understanding of BCG and vaccine research is limited at best. Efforts need to be directed toward increasing the awareness of the vaccine research process, the limitations of BCG, and the need for a new vaccine. Without the support of these stakeholders it is unlikely that any new vaccine will be scaled up rapidly—if at all.

As part of these collaborative activities, experienced researchers and vaccine networks should prioritize building the research capacity and infrastructure in high-TB-burden countries to conduct clinical trials and operational research. This will require regulatory authorities to provide guidance on development pathways to ensure that they are consistent with global regulation. Establishing clear criteria for the evidence set required to license a new vaccine would establish a standard by which all trials must adhere, making the process more efficient and allowing the harmonization of data collected across studies.

Finally, without a great increase in funding, TB vaccine research will stagnate. Not only are more funds needed to support basic science, to conduct efficacy studies and operational research, and to encourage young scientists to take on TB research but there needs to be a diversification of funders. The Bill and Melinda Gates Foundation (BMGF) has consistently accounted for the great majority of funding for TB vaccine research, contributing 61% of all vaccine research and development funding in 2008. In fact, a boost in funding for TB vaccine research in 2008 was almost entirely accounted for by a grant from the BMGF to Aeras (Treatment Action Group 2010). Reliance on one or a few funders may result in donor fatigue and a shrinking pool of institutions and researchers able to contribute to vaccine development. The pool of funding must not only increase; the number of funders also needs to increase to get a newer, better, and safer vaccine that can protect everyone from all forms of TB.


Abu-Raddad LJ, Sabatelli L, Achterberg JT, Sugimoto JD, Longini IM Jr., Dye C, Halloran ME. Epidemiological benefits of more-effective tuberculosis vaccines, drugs, and diagnostics. Proceedings of the National Academy of Sciences USA 2009;106(33):13980–85.

Beresford B, Sadoff JC. Update on research and development pipeline: Tuberculosis vaccines. Clincial Infectious Diseases 2010;50(3):S178–83.

De Bruyn G, Garner, P. Mycobacterium vaccae immunotherapy for treating tuberculosis. Cochrane Database of Systematic Reviews 2003, no. 1. Art. No.: CD001166. DOI: 10.1002/14651858.CD001166.

Dye C and Williams BG. The population dynamics and control of tuberculosis. Science, 14 May 2010.

Hanekom W. Personal communication, 1 June 2010.

Kaufman SHE, Hussey G, Lambert P. New vaccines for tuberculosis. Lancet 2010; 375(9731) 2110-2119 doi:10.1016/S0140-6736(10)60359-9

Mansoor N, Scriba TJ, De Kock M, Tameris M, Abel B, Keyser A, Little F, Soares, A, Gelderbloem S, Mienjen S, Denation L, Hawkridge A, Bloom WH, Kaplan G, Hussey G, Hanekom, WA . HIV-1 infection in infants severely impairs the immune response induced by Bacille Calmette-Guerin vaccine. Journal of Infectious Diseases 2009;199(7):982–90.

McShane H. Personal communication, 14 June 2010.

Neuremberger E. Animal models, persistence and latency. Paper presented at the Workshop on Fundamental Research on Tuberculosis, Bethesda, MD, 18 March 2010.

Ofori-Anyinam O. Personal communication, 15 June 2010.

Russell DG, Barry CE, Flynn JL. Tuberculosis: What we don’t know can, and does, hurt us. Science, 14 May 2010.

Treatment Action Group. 2009 Report on Tuberculosis Research Funding Trends, 2005–2008. New York: Treatment Action Group, 2009.

von Reyn CF, Mtei L, Arbeit RD, Waddell R, Cole B, Mackenzie T, Matee M, Bakari M, Tvaroha S, Adams LV, Horsburgh CR, Pallangyo K. Prevention of tuberculosis in Calmette-Guérin-primed, HIV-infected adults boosted with an inactivated whole-cell mycobacterial vaccine. AIDS 2010;24(5):675–85.

Wallis RS, Pai M, Menzies D, Doherty TM, Walzl G, Perkins M, Zumia A. Biomarkers and diagnostics for tuberculosis: Progress, needs, and translation into practice. Lancet 2010; 375(9729)1920 – 1937; doi:10.1016/S0140-6736(10)60359-9

Willingham P. Personal communication, 7 June 2010.

Wooley J. Personal communication, 1 June 2010.

Zar HJ, Cotton MF, Strauss S, Karpakis J, Hussey G, Schaaf HS, Rabie H, Lombard CJ. Effect of isoniazid prophylaxis on mortality and incidence of tuberculosis in children with HIV: Randomised controlled trial. British Medical Journal 2007;334:136.

Links to other websites are current at date of posting but not maintained.