US researchers have been given the go-ahead to test the technique in patients early in 2008. This follows promising results in laboratory tests and in studies in mice, reported the papers.

If successful, the technique could revolutionise the treatment of cancer and could become available within two years.

Everyone has some ability to fight cancer, via immune cells called natural killer cells, which can identify and kill tumour cells.

But the researchers have now discovered that a much larger population of immune cells called granulocytes can also kill cancer.

The researchers tested the theory that immune cells could fight cancer by mixing granulocytes taken from healthy individuals with cervical cancer cells.

Granulocyte donors are already being used in patients whose immune systems have been depleted by chemotherapy.

The granulocytes were found to be able to kill most cervical cancer cells within 24 hours.

But the effectiveness of granulocytes was found to vary from person to person.

Stress, old age and a change of season appear to have a negative effect on the granulocytes' cancer-killing properties.

What is the research?
The reports are based on preliminary results presented earlier this month at the Strategies for Engineering Negligible Senescence meeting in Cambridge.

For the study, blood samples were taken from more than a 100 individuals. Granulocytes were extracted from the samples and mixed with cervical cancer cells.

A newly developed in vitro assay was then used to measure the ability of the granulocytes to kill the cervical cancer cells.

But the effectiveness of granulocytes varied depending on the donor.

Immune cells from one individual killed 97 per cent of cancer cells in 24 hours, while those from another healthy donor only killed 2 per cent in the same time.

Average cancer-killing ability was found to be lower in granulocytes taken from adults over the age of 50. This capacity was even lower if the donor had cancer.

Granulocyte ability to kill cancer cells fell if taken between the months of November and April or if the donor is stressed.

Forty per cent of the mouse's offspring also survived several injections of cancer cells, suggesting that the cancer resistance was a genetic trait.

This cancer-resilience trait has now been passed on to more than 2,000 of the mouse's offspring, spanning 15 generations and several different mouse strains.

Last year, the researchers successfully treated a range of cancers in mice by injecting them with granulocytes taken from a cancer-resistant mouse.

The team also found it could protect normal mice from what should have been lethal doses of aggressive cancers using granulocyte injections.

What do the researchers say?
Lead researcher Dr Zheng Cui, from the school of medicine at Wake Forest University in North Carolina, said he hoped the technique could be a cure for cancer.

'Our pre-clinical tests have been very successful,' he said.

'If this is half as effective in humans as in mice, it could be that patients could be cured or at least given one or two years extra of high-quality life.'

The treatment method of granulocyte transfusion has been known in medicine for over 30 years, added Dr Cui.

So far, signs of graft-versus-host disease (where the host body attacks the transplanted cells) has not been noted, he said.

'If it works in humans we could save a lot of lives, and we could be doing so within two years.'

But the key is how to identify good immune cell donors correctly, said Dr Cui.

'Our future research will be aimed at improving identification techniques.'

Dr Cui hopes to begin human trials next summer in the US, when the immune systems of donors will be at their peak.

The reduction in the cancer-killing ability of the granulocytes observed during the winter months might be due to a slow metabolism, he added.

What other researchers say
Professor John Gribben, an immunologist based at Cancer Research UK's medical oncology unit at St Bart's and The London School of Medicine, said: 'Treating cancer by using the body's innate cancer-killing ability is an interesting concept.

'But transferring immune system cells from people with an apparent innate immunity into cancer patients could be risky.'

• Granulocytes have been shown to kill cervical cancer cells in laboratory tests.
• Factors such as stress can influence the effectiveness of granulocytes.
• Further tests in humans are expected to begin next year.
• If successful, granulocyte transfusion could become available within two years.