The data, published in the New England Journal of Medicine, were from 60 healthy volunteers given the vaccine in Britain between Sept 17 and Nov 18 last year.

The volunteers got one of three doses - low, medium, or high - and data from 28 days after vaccination showed the shot was safe at these doses, with only mild side effects.

"People typically experienced mild symptoms that lasted for one or maybe two days, such as pain or reddening at the injection site, and occasionally people felt feverish," Hill said.

However, the antibody response was weaker than was found in a trial of the same Ebola vaccine in macaque monkeys, in which the animals were also found to be protected.

Hill said the lower antibody levels, together with a lower response detected in the immune system's T-cells, suggested to him that a booster may well be needed.

Jeremy Farrar, director of the Wellcome Trust charity which helped fund the trial, said it provided "good initial evidence that the GSK vaccine will be safe to use in people".

"However, we still don't know whether it will provide protection against Ebola infection in a real-world situation," he said. "That's why trials in West Africa of this, and the other vaccines in development, must begin as soon possible."