Updated

A GlaxoSmithKline experimental malaria vaccine touted as a new weapon in the fight to eradicate the disease proved only 30 percent effective when given to babies as part Africa's largest ever clinical trial.

The surprisingly poor result for the world's first potential vaccine against malaria leaves uncertain whether it can have a useful role in fighting the mosquito-borne disease that kills hundreds of thousands of children a year.

[pullquote]

Philanthropist Bill Gates, who has helped fund its development, said further data was needed to determine whether and how the vaccine might be used.

"The efficacy came back lower than we had hoped, but developing a vaccine against a parasite is a very hard thing to do," he said in a statement.

Results from the final-stage trial with 6,537 babies aged six to 12 weeks showed the vaccine provided "modest protection", reducing episodes of the disease by 30 percent compared to immunization with a control vaccine, researchers said on Friday.

That efficacy rate one year after vaccination is less than half the 65 percent reported in a smaller mid-stage trial in 2008 that followed babies of a similar age for six months.

It's also a lot less than the 50 percent seen last year in Phase III trial data from five to 17 month-olds.

Vaccinating babies, rather than toddlers, is the preferred option, since the new vaccine could then be added to other routine infant immunizations. A separate program for older children would involve a lot of extra costs.

Despite the limited success, Britain's top drugmaker said it would push ahead with developing the vaccine, called RTS,S or Mosquirix, and GSK Chief Executive Andrew Witty said he still believed it would be an important tool in fighting malaria.

"We've been at this for 30 years, and we're certainly not going to give up now," Witty told a conference call.

GSK does not expect to make any profit from the vaccine, which would only be sold in poor countries. Witty reiterated a promise that if RTS,S is ultimately approved for market, it would be priced at cost of manufacture plus a 5 percent margin, and the margin would be reinvested by GSK in malaria research.

Given the target market, it is governments and international groups that will fund the vaccine's roll-out, and they now need more positive data before deciding whether it is worth buying.

"We will have to have more information to give us a clearer idea as to how useful this vaccine will be," said Seth Berkley, CEO of the GAVI Alliance, which funds bulk-buy vaccination programs for poorer nations.

In particular, Berkley told Reuters he wanted to see longer-term data, including the effect of booster shots, and an analysis of how the vaccine performed in different settings.

Details of the malaria trial, which is Africa's largest ever clinical trial involving almost 15,500 children in seven countries, were presented at a medical meeting in Cape Town and published online by the New England Journal of Medicine.

Witty said he would have liked to have seen efficacy rates of around 50 percent in infants, but stressed that more data would become available before the trial ends in 2014 which may throw more light on why rates of success are so variable.

"It's not as positive as we all might have hoped ... but it may take us down the road where we start to think about the vaccine for specific types of patients in specific types of areas," he said. "It may open up a more customized approach to how this potential vaccine gets used."

Malaria is caused by a parasite carried in the saliva of mosquitoes. It is endemic in more than 100 countries worldwide and infected around 216 million people in 2010, killing around 655,000 of them, according to the World Health Organization.

Control measures such as insecticide-treated bednets, indoor spraying and anti-malaria drugs have helped cut malaria cases and deaths significantly in recent years, but experts say an effective vaccine is vital to completely overcome the disease.

Scientists around the world are working on other potential malaria vaccines but RTS,S is by far the furthest ahead in development.