A new systematic review of animal studies testing a vaccine for tuberculosis raises questions about whether the studies provided sufficient evidence to move into trials of children.
The new vaccine was a virus-expressing antigen 85A (MVA85A) designed to boost the immunity offered by the existing Bacillus Calmette-Guérin (BCG) vaccine which has little protective effect in practice. The review, published today in the International Journal Epidemiology, evaluates the animal evidence that contributed to the decision to conduct human studies.
LSTM's Professor Paul Garner is the senior author on the review, and coordinating editor at the Cochrane Infectious Diseases Group. Working along with colleagues at the University of Edinburgh and South Africa, the team included data from studies where animals were given the MVA85A booster with BCG and then exposed to a TB challenge, and compared their outcome with that in animals receiving BCG alone. The team identified eight studies published up to September 2014, involving a total of 192 animals.
Different kinds of animals were used in the different studies, but overall the authors found that studies were too small, the quality was low, and details of the experimental methods used were poorly reported. Combining the findings from all studies gave no evidence to support the effectiveness of MVA85A as a BCG booster.
What surprised the review authors was the delay with publishing the largest monkey trial with the longest follow up. This trial had 5 deaths out of 6 monkeys in the new vaccine group compared to two deaths in the 6 monkeys in the control group who only got BCG. This trial only appeared in the public domain after the researchers had received funding for an experimental human trial and had recruited over half of the 2797 infants in South Africa needed.
"This is very disappointing", said Professor Garner. "Tuberculosis remains a major health challenge in many parts of the world and - as with every field of research - we all have a responsibility to work to the highest standards of scientific integrity. We need to make our findings accessible to as wide an audience as quickly as possible. You don't bury results you don't like. It can mislead other scientists, and misinforms the public".