Despite being discovered more than a century ago, tuberculosis still lacks an effective vaccine leading to millions of deaths. Photo credits: CDC via Unsplash
Tuberculosis (TB) is thought to be the world’s leading cause of death from a single infectious agent, killing 1.25 million people in 2023. Most TB cases in humans are caused by the bacterium Mycobacterium tuberculosis and can be treated with antibiotics. Nevertheless, a good TB vaccine would be able to prevent infection with TB and the development of disease, supporting the World Health Organisation’s goal of ending the global TB epidemic by 2035.
A good TB vaccine would be able to prevent infection with TB and the development of disease…
Currently, there’s only one approved vaccine against TB: the BCG vaccine. BCG (Bacillus Calmette-Guérin) is a live-attenuated vaccine. This means that it is created by growing live bacteria in culture medium until they have adapted to this artificial environment and lost their capacity to cause disease in humans, whilst retaining the capacity to induce an immune response. BCG is derived from Mycobacterium bovis, a mainly cattle-infecting bacterium closely related to M. tuberculosis.
BCG is widely recognised to be safe, and it is one of the most widely used vaccines. Vaccination with BCG is easily identified by the characteristic blister-like scar left at the site of injection due to the immune response to the vaccine. Yet BCG might be unfamiliar to many of our UK-based readers, who may have only seen this scar on their parents. The NHS removed BCG from its routine vaccination programme in 2005 due to decreased transmission of TB within the UK. Here, TB is often portrayed as a historical threat rather than a current problem, with a 2016 article in The Guardian wondering why “Victorian diseases” such as TB were making a “comeback” in the UK. How can this be reconciled with the massive global burden of TB?
The answer is that BCG fails us where it is needed the most. BCG is believed to have >70% efficacy at preventing TB in schoolchildren in the UK, but the highest burden of TB falls on countries at lower latitudes, close to the equator. At lower latitudes, the protection against TB provided by BCG is often negligible, with no significant reduction in TB incidence following vaccination reported in Southern India, Malawi, and Colombia. Geographically variable efficacy of BCG is thought to be driven by geographical variation in exposure to non-tuberculosis mycobacteria, environmental bacteria that do not cause TB disease but may induce immune responses that either “block” or “mask” the ability of BCG to induce a protective immune response against TB itself.
As a live-attenuated vaccine, BCG cannot be given to immunocompromised people, who have weaker immune systems and may not be able to control the growth of the live bacteria within the vaccine.
Highly variable efficacy is not the only problem with BCG. As a live-attenuated vaccine, BCG cannot be given to immunocompromised people, who have weaker immune systems and may not be able to control the growth of the live bacteria within the vaccine. This would not be a risk if a different type of non-live vaccine were used, for example, a protein-based vaccine. The inability to use BCG in immunocompromised individuals is a big problem because there is a strong geographical overlap between the TB and HIV epidemics: over 60% of TB patients in sub-Saharan Africa are HIV-positive.
Additionally, BCG vaccination interferes with our ability to monitor TB incidence. The tuberculosis skin test (TST) involves injecting tuberculin, a toxin produced by TB, into the skin. If you are infected with TB, your immune system will rapidly respond to this toxin, resulting in redness and swelling at the TST site. BCG vaccination can produce a false positive to the TST for up to 25 years. Though there are alternatives to the TST, such as IGRAs (interferon gamma release assays), the TST is a quick and cheap way of diagnosing TB infection, making it incredibly important in resource-poor settings.
In 2021, the world celebrated BCG’s 100th birthday. Nevertheless, BCG is not effective where it is most needed, cannot be given to immunocompromised individuals, and interferes with the TST. So why, in over 100 years, have we not developed a better alternative?
It’s not for a lack of effort. Though current investment in TB vaccine development falls short of the estimated $1.25 billion required, with just over $200 million invested in 2023, there are core challenges to developing new TB vaccines that underlie this delay. We have a poor understanding of protective immunity against TB, difficulty conducting clinical trials, and difficulty identifying the most promising vaccine candidates, which is crucial for distributing the limited funding.
Effective vaccines often aim to recreate natural immune responses, but immune responses are complex: not all cells or pathways activated following infection are actually protective against the pathogen. Infection with TB does not confer complete immunity—some people remain latently infected, with no symptoms of infection, for years. This makes designing a TB vaccine challenging. It is unclear what immune mechanisms are protective against TB, so it is difficult to identify what immune responses a successful vaccine would stimulate. Cell-mediated responses are thought to be important, but this knowledge does not necessarily translate directly to developing a good vaccine. For example, MVA85A is a candidate recombinant protein vaccine that successfully induced cell-mediated immune responses but was ineffective at preventing TB.
Trialling alternatives to BCG is challenging since it is ethically unconscionable to withhold BCG vaccination. This means trialling alternatives to BCG is more difficult than vaccines intended to support BCG, such as booster vaccines, post-exposure vaccines, or treatment vaccines. Yet, vaccines that support BCG cannot necessarily address all the problems with BCG, such as the inability to vaccinate immunocompromised individuals or its interference with the TST.
A high prevalence of TB is necessary to ensure a large enough proportion of the participants contract TB to assess the efficacy of the vaccine.
There are a limited number of trial sites that are suitable for large-scale efficacy trials, which require a high prevalence of TB, adequate clinical and laboratory infrastructure, and pre-existing data about regional epidemiology. A high prevalence of TB is necessary to ensure a large enough proportion of the participants contract TB to assess the efficacy of the vaccine. If TB is rarer, a much larger study group is required, which is harder to recruit, and so the trial will be more expensive to conduct. Since there are a small number of suitable sites, it is important to down-select vaccine candidates so that only the most promising vaccines progress to efficacy trials. Identifying which candidates are promising is challenging for several reasons. It is unclear which animal models best predict vaccine performance in humans, and we lack an immune correlate of protection and good human challenge models for TB.
Animal models of TB
Vaccine development often relies on trialling new vaccines in animals to assess their safety and efficacy. Humans are the only natural host of M. tuberculosis, but scientists can infect other animals—often mice or guinea pigs—with TB to trial vaccines and study immune responses. However, the way TB disease develops in these animals is not identical to that in humans due to differences in our immune systems. No current animal model accurately represents all elements of disease pathogenesis in humans, so the conclusions we can draw are not complete. If we had an effective vaccine, we would be able to easily identify which animal models gave the best predictions of vaccine performance in humans. Until then, trials conducted in animals may be misleading, slowing down the vaccine development process.
Immune correlates of protection
Immune correlates of protection are immune functions that are associated with vaccine efficacy, for example, antibody levels. These enable us to predict whether a vaccine will be effective before a participant reaches a clinical endpoint, such as TB disease. This means we can conduct smaller and shorter trials. Identifying an immune correlate of protection for TB has been called a ‘game-changer’ for vaccine development, but once again, a correlate of protection can only be identified following efficacy trials of a successful vaccine.
Human challenge studies
Human challenge studies are a type of clinical trial in which participants are vaccinated and then deliberately exposed to a pathogen to assess the safety and efficacy of the vaccine. Over 600 people have undergone malaria challenge studies in Oxford as part of the ongoing effort to develop malaria vaccines. Why is it considered safe to infect people with malaria, but not TB? TB human challenge studies are hindered by several unique ethical concerns. Unlike malaria, TB can be transmitted directly between individuals, meaning there is a high risk of onward transmission to non-consenting (and non-monitored) third parties. There would have to be very strict biosafety restrictions on any TB trial to prevent the initiation of an outbreak. Other concerns lie with the risk to the individual. TB can cause chronic respiratory issues, and although there are effective antibiotic treatments, these must be taken for at least six months and come with several side effects, including the risk of nerve damage.
Unlike malaria, TB can be transmitted directly between individuals…
Finding a human challenge model that can be used to test vaccine performance safely is crucial for successful TB vaccine development. BCG has been suggested as an alternative challenge model. We know that it is safe to vaccinate people with BCG, and there is no risk of onward transmission. Researchers at the University of Oxford have developed an aerosol BCG model, which would mimic the real mode of TB transmission better than injection. The first trial incorporating this challenge model began in 2025, investigating the efficacy of a new vaccine candidate called ID93+GLA-SE.
TB has the unfortunate accolade of being the world’s ‘top infectious killer’. Although the number of deaths associated with TB has decreased, the total number of people ill with TB has risen. With the rise of antibiotic-resistant TB, we desperately need a vaccine that is more effective at preventing disease than BCG, can be given to immunocompromised individuals, and doesn’t interfere with the TST.
Although the number of deaths associated with TB has decreased, the total number of people ill with TB has risen.
TB vaccine development can feel like a catch-22. Developing new vaccines is difficult because we do not know which animal models are the most informative, or what is a good correlate of protection, but until we have a good vaccine, we cannot solve these problems either. Not all is lost—new innovations are bringing hope. A new aerosol BCG challenge model may overcome our inability to use TB in human challenge studies, improving our ability to trial candidate vaccines. Further technological innovations will continue to support vaccine development as scientists design creative solutions to the challenges of TB vaccine development. The future looks promising.
