A vaccine against tuberculosis (TB) reduced the risk of
developing active TB in HIV-negative people with latent TB infection by 54% in
a large phase 2b study published last month in The New England Journal of Medicine.

TB is the leading infectious cause of death
worldwide and up to a quarter of the world’s population is estimated to be
latently infected with Mycobacterium tuberculosis. Latent infection, without
symptoms, can last for a lifetime and will only develop into active TB if the immune system ceases to control it. Malnutrition, diabetes,
advanced HIV infection and old age are the most common reasons for the development
of active TB.

The BCG vaccine provides good protection against TB in
children but is less effective in protecting against pulmonary TB in adults.

A vaccine that could prevent the development of TB in adolescents
and young adults would have the greatest impact on the global burden of TB,
according to a systematic review of TB vaccine modelling studies, published in

Studies of previous TB vaccines have been disappointing,
with two phase 2 studies failing to show a protective effect.

However, a vaccine candidate developed by GlaxoSmithKline
(M72/ASO1E) and the non-profit TB vaccine developer Aeras was shown
to produce strong immune responses in a range of populations, including people
living with HIV, in phase 2a studies.

A phase 2b proof-of-concept study of the vaccine’s efficacy in
adults aged 50 and under with latent TB was carried out in Kenya, South Africa
and Zambia. People were eligible to join the study if they tested positive
using the Quantiferon-TB Gold Assay, indicating latent TB, and tested
negative for active TB on the Xpert MTB/RIF assay. The study was restricted to
HIV-negative people. The median age of study participants was 29 years and 43%
were women.

The primary outcome of the study was the prevention of
active TB not associated with HIV infection, confirmed with either culture or
Xpert MTB/RIF.

The study randomised 3575 participants to receive two doses
of the vaccine one month apart, of which 3330 received both doses and 3283 were
considered to qualify for the according-to-protocol analysis (follow-up
commencing 30 days after first dose).

During a mean follow-up period of 2.3 years, ten cases of
active TB were diagnosed in the vaccine group (0.3 per 100 person-years) and 22
in the placebo group (0.6 per 100 person-years) giving an overall vaccine
efficacy of 54% (95% CI 2.9-78.2, p = 0.04).

When the study investigators controlled for country, sex,
diabetes, BCG vaccination and smoking, the result was almost identical. When
they used a more sensitive definition of active TB – positive by both culture
and Xpert MTB/RIF test – the vaccine efficacy was 70.3% (95% CI 19.4-89%).

Redness and swelling at the injection site was uncommon but
fatigue, headache, malaise and muscle aches were reported more frequently in
those who received the vaccine (58-69% of the vaccine group compared to 26-47%
of the placebo group). Fever was also more common in the vaccine group (19% vs

The study investigators say that the difference in vaccine
efficacy between analyses requiring one positive test and two positive tests
may tell us something about the development of active TB from latent TB. Approximately
one-third of the active TB cases were confirmed by a single positive test and in ten ‘single positive’ cases it took a long time to either produce a positive
result by culture or by Xpert MTB/RIF amplification, suggesting a lower
bacterial load in these cases.

Progression from latent infection to active TB may represent
“a transition through a spectrum of inflammatory and infected states”, they

A sub-group analysis found that the vaccine was highly
effective in people aged 25 and under, but the investigators are cautious about
this finding. They say that it might be explained by better immune control of
latent infection in older people, meaning that the vaccine provided little
additional benefit, or that the vaccine was more effective when the interval
between BCG vaccination and the trial vaccination was shorter.

The vaccine will need to undergo further tests in larger studies
in various populations to show whether it is more effective in younger people,
and to refine estimates of the vaccine’s protective effect. The very wide
confidence intervals reported in this study are likely to be reduced by larger

The vaccine will also need to be tested in people living
with HIV and in people without latent TB infection, as the protective effect of
the vaccine against infection in settings where TB is endemic is unknown.

Last week Aeras announced that its portfolio of vaccines and
vaccine development programme had been acquired by the International AIDS
Vaccine Initiative (IAVI). Aeras’s extensive network of trial sites in southern Africa and the overlap between HIV and TB were cited as reasons for the incorporation of TB vaccines into IAVI’s portfolio.