www.thelancet.com Vol 381 March 23, 2013
One of the great quests of contemporary medical research
is the search for an improved tuberculosis vaccine—one
that provides greater and more consistent protection
against tuberculosis than the BCG vaccine can achieve.
The stakes are high. The venture is costly and risky, but
has a huge potential payoff . A high-effi cacy vaccine
could revolutionise control of tuberculosis, shifting the
emphasis from treatment to prevention. As the case
numbers slowly fall in high-burden countries, and as new
strains of drug-resistant tuberculosis emerge, a novel
and transformational technology for tuberculosis control
would be cause for great celebration.1
US$70–100 million is spent on vaccine research for
tuberculosis every year and the pipeline of candidate
vaccines is now longer and wider than ever before.2–4 As
each of the candidates moves from preclinical to clinical
stages, passing tests for safety and immunogenicity,
experi ments to assess effi cacy in human beings are
Against this background, Michele Tameris and col-
leagues report in The Lancet results of a phase 2b trial in
infants in South Africa of the vaccine modifi ed Vaccinia
virus Ankara expressing antigen 85A (MVA85A).5
Although the primary objective of the trial was to
assess safety, it also made a preliminary assessment of
effi cacy—and many readers will go, with halted breath,
straight to the conclusions about effi cacy. They will
be confronted with results that are, on the face of it,
disappointing, show ing little evidence of effi cacy in
terms of prevention of tuberculosis or infection with
Although the trial raised no concerns about safety,
the absence of any detectable effi cacy presents the
tuberculosis vaccine community with a serious
challenge. However, the fi ndings reported by Tameris
and colleagues are not a terminal prognosis for
MVA85A, or for any of the other tuberculosis vaccines
in develop ment. To understand why, the results of this
particular trial need to be put in a wider context.
Two main strategies exist for development of
tuberculosis vaccines.6 The fi rst is to replace the widely
used BCG vaccine with an improved whole-organism
vaccine, which is either a recombinant BCG or an
attenuated strain of M tuberculosis. The second is to
develop a subunit boosting vaccine, which is designed
to enhance whatever protection is already provided by
BCG. MVA85A is an example of the latter strategy.
In their elegant randomised, placebo-controlled
trial, Tameris and colleagues followed up 2794 BCG-
vaccinated infants for up to 37 months (median 24·6,
IQR 19·2–28·1) in two nearly equal groups. 39 (2·8%)
of 1395 infants in the placebo group (Candida skin
test antigen) satisfi ed the primary defi nition of active
tuberculosis, of whom 20 were microbiologically
confi rmed. 32 (2·3%) of 1399 infants in the vaccine
group (MVA85A) satisfi ed the primary defi nition of
active tuberculosis, of whom 22 were microbiologically
confi rmed. Thus, vaccine effi cacy was 17·3%, which was
not distinguishable from zero (95% CI –31·9 to 48·2).
Neither was there any evidence for protection against
M tuberculosis infection, as determined by an in-vitro
interferon γ release assay (QuantiFERON-TB Gold In-
tube; Cellestis, Australia). During the trial, 349 (13%)
of 2792 participants became positive on this assay,
171 (12%) in the placebo group and 178 (13%) in the
vaccine group. The ratio of apparent infection to disease
was thus about fi ve to one considering all cases of
tuberculosis, or eight to one for confi rmed cases only.
A major event for new tuberculosis vaccines
3 Hunt J. The four improvements I want to see in the NHS by 2015.
Conservative Home. Nov 26, 2012. http://conservativehome.blogs.com/
of.html (accessed Feb 24, 2013).
Joseph KS, Liu S, Rouleau J, et al. Infl uence of defi nition based versus
pragmatic birth registration on international comparisons of perinatal
and infant mortality: population based retrospective study. BMJ 2012;
Joossens L, Raw M. The tobacco control scale 2010 in Europe. Brussels:
Association of European Cancer Leagues, 2011. http://aerzteinitiative.
bplaced.net/EUrank10.pdf (accessed Feb 24, 2013).
WHO. European health for all database (HFA-DB). http://data.euro.who.
int/hfadb (accessed Feb 27, 2013).
7 Stevens A. What can we learn from the Portuguese decriminalization of
illicit drugs? Br J Criminol 2010; 50: 999–1022.
Moodie C, Stead M, Bauld L, et al. Plain tobacco packaging: a systematic
review. London: Public Health Research Consortium, 2012. http://www.
review.pdf (accessed Feb 24, 2013).
Sheffi eld Alcohol Research Group. Independent review of the eff ects of
alcohol pricing and promotion. London: Department of Health, 2008.
indreview (accessed Feb 24, 2013).
10 Scientifi c Advisory Committee on Nutrition. Update on trans fatty acids
and health: position statement by the Scientifi c Advisory Committee on
Nutrition. London: TSO, 2007. http://www.sacn.gov.uk/pdfs/sacn_trans_
fatty_acids_report.pdf (accessed Feb 24, 2013).
February 4, 2013
See Articles page 1021
Science Photo Library
www.thelancet.com Vol 381 March 23, 2013 973
These results might provide little optimism that
MVA85A will deliver a new tuberculosis vaccine. But
this trial was designed to answer only one of a series of
important questions about new tuberculosis vaccines.
Before drawing any fi rm conclusions, we need to answer
several other questions.
First, could MVA85A be eff ective against infant and
child hood tuberculosis when used independently of BCG?
Substantial evidence shows that BCG protects young
children against tuberculosis; so to seek yet more pro-
tection might be asking too much of MVA85A. This poses
a major problem for tuberculosis vaccine research because
BCG is recommended for infants in all countries with high
burdens of tuberculosis. One of the explanations for the
BCG vaccine’s poor performance in some populations is
that exposure to other mycobacterial antigens can mask
its eff ect—perhaps BCG masked the eff ect of MVA85A?7
Second, in view of the variable performance of BCG
in diff erent populations, can we assume that the same
results will be obtained with MVA85A in other popu-
lations? South Africa has been favoured for vaccine trials
because the transmission rate of M tuberculosis and burden
of disease are comparatively high. The question remains
whether the characteristics that are responsible for this
high burden somehow militate against immunisation.
Third, could MVA85A, working as a booster to BCG,
protect adolescents and adults against pulmonary
tuberculosis in a way that it cannot protect infants?
Immunologically naive infants and young children do
not develop pulmonary tuberculosis in the same clinical
form as adults, and adult pulmonary tuberculosis is the
main target of tuberculosis control.
Fourth, might this vaccine work if administered
to people infected with HIV? MVA85A is also being
tested in HIV-positive adults in Senegal and South
Africa. If these trials are successful, MVA85A might
be a replacement for BCG which, as a live-attenuated
vaccine, is not recommended for people living with HIV.
Fifth, could MVA85A be effi cacious against severe
forms of tuberculosis, including pulmonary tuberculosis,
without preventing infection or mild forms of disease?
A high effi cacy against severe disease could have been
masked in this trial which, by use of invasive diagnostic
methods including gastric lavage, detected relatively
mild forms of tuberculosis infection or disease.
Sixth, how does the effi cacy of MVA85A compare
with other vaccine candidates now in phase 2b trials?
The world eagerly awaits the next set of results on the
effi cacy of two other subunit boosting vaccines, both
from trials in South Africa: AERAS-402/Crucell Ad35 in
infants and GlaxoSmithKline’s GSK M72 in adults.3
Finally, key questions remain about immunogenicity.
The word itself might be misleading, insofar as it is
used to describe any measurable immunological eff ect,
irrespective of the implications for protection. MVA85A
is described as modestly immunogenic because it
generated moderate antigen-specifi c Th1 and Th17
responses (compared with other populations) although
it showed no evidence of protection against infection or
disease. A large bank of samples collected in the recent
trial have yet to be examined and analysed—and might
yet help to identify immunological factors that are
characteristic of individuals who do and do not develop
tuberculosis. The identifi cation of a valid measure of
protective immunity against tuberculosis would be a
discovery of overwhelming importance.
Apart from the spur to solve all these problems,
the search for a new tuberculosis vaccine has other
sources of inspiration. It remains an astonishing fact
that children aged 5–10 years are very resistant to
development of active tuberculosis.8 Is this resistance
suggestive of an immunological mechanism that could
be exploited for vaccine development? In preclinical
research, investigations with animals continue to
generate new and promising results. One example is
H56, a vaccine that combines antigens characteristic of
early infection and latency, and seems to protect mice
against tuberculosis disease before and after exposure to
infection.9 A vaccine that could protect everyone before
and after infection is an epidemiologist’s dream.10
Now is a key moment in tuberculosis vaccine research.
Trials such as that of Tameris and colleagues are at last
generating hard evidence about protection against
tuberculosis in human beings, the most important goal
of immunisation. If the history of tuberculosis vaccine
research teaches us anything, it is to expect surprises. We
need to go on playing the high-stakes game.
*Christopher Dye, Paul E M Fine
HIV/AIDS, Tuberculosis, Malaria and Neglected Tropical Diseases,
World Health Organization, Geneva, Switzerland (CD); and
Department of Infectious Disease Epidemiology, London School
of Hygiene and Tropical Medicine, London, UK (PEMF)
www.thelancet.com Vol 381 March 23, 2013
Folic acid fortifi cation, instituted in the mid to late
1990s in the USA and Canada, and now in more than
50 countries,1 has been highly eff ective for its intended
purpose—to reduce the incidence of neural tube defects
(eg, spina bifi da and anencephaly).2–4 However, lingering
concerns about the safety of excess intake of folic acid
still exist, particularly with respect to cancer.5
Why the concern? Folic acid—or rather the bioactive
forms of folic acid (collectively ”folate”)—is required for
de novo synthesis of thymidine, adenine, and guanine,
three of the four nucleotides needed to assemble
DNA. Because cancer cells, like all cells, synthesise
DNA during mitosis, they require a supply of folate.
Limitation of the supply of folate, or inhibition of folate
metabolism (with anti-cancer drugs), slows or arrests
proliferation of several forms of cancer. Conversely, the
supply of cancer cells with extra folate can promote
their proliferation;5–8 thus the concern that increased
folate intake due to folic acid fortifi cation, on top of
dietary and supplemental folate consumption, could
promote existing neoplasms.
The issue is complicated, however, because in non-
cancerous cells, folate can prevent or limit mutational
events that initiate tumorigenic transformation. When
folate is defi cient, the supply of thymidine to insert
into DNA is scarce, resulting in uracil misincorporation.
When uracil is removed, holes are left behind in
the DNA sequence, which ultimately leads to DNA
strand breaks. Additionally, C→T mutations are more
likely to occur when folate is defi cient. Both events
result in DNA damage that can foster tumorigenic
transformation. Thus, folate has a putative two-
faced relationship with cancer: it can protect against
initiation, but promote proliferation.
Is there epidemiological evidence for these dual eff ects
of folate? In 2007, Joel Mason and colleagues9 pub lished
a controversial ecological study that suggested colorectal
cancer incidence, which had been decreasing in the USA
and Canada, increased temporarily after the institution of
folic acid fortifi cation in the two countries. Subsequently,
the incidence rate began to decrease again.
One interpretation of these data is that folic acid
fortifi cation enhanced the proliferation and growth
of preclinical neoplastic lesions, making them become
clinically evident, and thus temporarily increasing
apparent incidence. Then, the protective eff ect of folate
against neoplastic initiation kicked in, and incidence
rates declined again. This interpretation has been
challenged, however, by the claim that the increased
incidence occurred too quickly after the start of folic
acid fortifi cation to be biologically plausible.10,11 This is a
speculative statement, however, as the rate of neoplastic
growth can be rapidly modulated by manipulation of
micronutrient supply, at least in cell lines and rodents.6–8
However, empirical data that excess micronutrient con-
sumption in man signifi cantly accelerates neoplastic
growth is lacking, and therefore this point of contention
remains open to debate.
In The Lancet, Stein Emil Vollset and colleagues11
present important new data that contribute to the
debate regarding folic acid and cancer. The investigators
conducted a meta-analysis of 13 randomised trials that
Folic acid and cancer—where are we today?
CD is a staff member of WHO. The authors alone are responsible for the views
expressed in this publication, which do not necessarily represent the decisions,
policy, or views of WHO.
1 WHO. G lobal tuberculosis control: WHO report, 2012. Geneva: World Health
Treatm ent Action Group. Tuberculosis research and development: 2011
report on tuberculosis research funding trends, 2005–2010. New York:
Treatment Action Group, 2012.
Aeras. Developing new tuberculosis vaccines for the world. http://www.
aeras.org/portfolio/index.php (accessed Jan 29, 2013).
Brenna n MJ, Stone MR, Evans T. A rational vaccine pipeline for tuberculosis.
Int J Tuberc Lung Dis 2012; 16: 1566–73.
Tameri s MD, Hatherill M, Landry BS, et al, and the MVA85A 020 Trial Study
Team. Safety and effi cacy of MVA85A, a new tuberculosis vaccine, in
infants previously vaccinated with BCG: a randomised, placebo-controlled
phase 2b trial. Lancet 2013; published online Feb 4. http://dx.doi.
6 McShan e H. Tuberculosis vaccines: beyond bacille Calmette-Guérin.
Philos Trans R Soc Lond B Biol Sci 2011; 366: 2782–89.
Fine PE M. Variation in protection by BCG: implications of and for
heterologous immunity. Lancet 1995; 346: 1339–45.
Donald PR, Marais BJ, Barry CE 3rd. Age and the epidemiology and
pathogenesis of tuberculosis. Lancet 2010; 375: 1852–54.
Aagaard C, Hoang T, Dietrich J, et al. A multistage tuberculosis vaccine that
confers effi cient protection before and after exposure. Nat Med 2011;
10 Young DB, Dye C. The development and impact of tuberculosis vaccines.
Cell 2006; 124: 683–87.
© 2013. World Health Organization. Published by Elsevier Ltd/Inc/BV.
All rights reserved.
January 25, 2012
See Articles page 1029
Science Photo Library