They plan to start testing it on humans next year.

Dr Alistair Ramsay, an immunologist at Australian National University, said: "It looks good so far. We just hope that it holds up when we get a chance to test it (on humans)."

Researchers in Canberra and Melbourne separately administered two vaccines, one to prime the immune system and another to boost it into action, before infecting four monkeys with the HIV virus.

The monkeys were initially infected but produced large numbers of T cells that attacked the virus, and cleared it from their system completely within weeks.

Four unvaccinated monkeys contracted the virus, which causes Aids in humans but not in monkeys.

The body's T cells are its first line of defence against infection, targeting and destroying foreign invaders.

According to the Journal of Virology, the researchers had to use two vaccines because neither was effective alone.

First they introduced a modified HIV gene into the monkeys' skin cells to stimulate their immune systems.

Several weeks later, they introduced a second dose of the HIV gene in a modified and harmless form to provoke a surge in the number of T cells.

When the macaque monkeys were then infected with HIV, the T cells multiplied even further and began to attack cells infected by the virus.

The World Health Organisation says up to 40 million people around the world are living with the Aids virus, with Africa the worst-affected continent but Asia also facing a pandemic.

The disease killed an estimated 2.3 million people last year.

Human trials

Macaque monkey

The Australian team plans to use the vaccine in a therapeutic trial on HIV carriers in Sydney and Melbourne next year to see if the vaccine could help ward off AIDS.

Then it hopes to start overseas trials on non-infected humans, probably in southeast Asia.

The team hopes to win US funding, probably from the US-based International Aids Initiative, which helps speed the development of candidate vaccines.

Dr Ramsay said: "We would hope to be in clinical trials within two years - and we'd know within three to five years if it worked."

"The advantage of our system is it is very cheap and does not require refrigeration which can be a major difficulty in tropical countries."

Another researcher involved in the potential vaccine, Commonwealth Scientific and Industrial Research Organisation animal health specialist David Boyle, was more guarded.

"I'm quite cautious about HIV vaccines because there've been a lot of efforts so far and none has been successful," he said.