Published online 28 September 2006 | Nature | doi:10.1038/news060925-12

News

Immune response vital in cancer fight

Activity of body's defences predicts outcome better than tumour spread.

When the immune system attacks a tumour it bodes well for the patient.When the immune system attacks a tumour it bodes well for the patient.Punchstock

How can you tell how bad a cancer is, and how likely the patient is to survive? New evidence suggests that the best way may sometimes be to look at how well the immune system has attacked tumours, rather than focussing on how far the tumours have spread.

It has long been known that the immune system can home in on cancerous cells and that immune cells can take up residence within tumours. But it hasn't been clear whether such cells have much of an impact on the progress of the disease.

A study by Jérôme Galon, Franck Pagés, and a team of researchers at INSERM in Paris, France, has brought the significance of these immune-system invaders to light.

The team studied tumours taken from over 400 patients with colon or rectal cancer and measured the type, location, number and gene expression of immune cells found in the tumours, among other things. They then compared how well this information matched up with what happened to the patients.

Most tumours are classified into stages based on their size and the extent to which they have spread throughout the body. A person with a stage 0 tumour — the lowest stage — has an early cancer that has yet to spread, whereas a person with a stage 4 tumour —the highest stage — has advanced cancer that has spread to another organ. It seems only logical that the higher the stage, the worse the prognosis for the patient.

In general, that is the case. But Galon and his co-workers found that a better indicator of patient outcome was the degree to which immune cells had infiltrated the colon or rectal tumour.

"We found that there is an importance of natural anti-tumour immunity against human cancer," says Galon. Whether a cancer recurs or not after treatment, he adds, may have more to do with the immune system than with the tumour itself. Even patients with small tumours that have not spread, says Galon, will have a bad prognosis if they have a weak immune response to the cancer. The results of his study will be presented in this week's Science.1

Surveillance system

ADVERTISEMENT

Although the assays in Galon's studies may be too arduous for daily prognostic use, the results suggest that the immune system plays a role in keeping colon and rectal tumours in check. That lends credence to the notion that the immune system maintains a surveillance system for cancer, says Robert Schneider, a professor of pathology at Washington University in St Louis, Missouri. The concept of such a system has been around since the 1950s, but its popularity has waxed and waned over the years as conflicting data has trickled in.

Galon's study is the most rigorous analysis thus far in humans. "Before, the clinicians said, 'We don't see good evidence for cancer immunosurveillance,'" says Schneider. "This paper really goes a long way toward saying that these things are occurring and they're clinically relevant." A previous study with ovarian cancer also showed hints that the immune system affects cancer, though in that case the presence of a particular type of immune cell in the tumours actually suppressed immunity and led to poorer patient outcomes2.

Galon and Pagés have only evaluated colon and rectal tumours so far, but Galon says they will be testing the same phenomenon in other tumour types, such as prostate and breast cancers.

Hyam Levitsky, an oncology professor at Johns Hopkins University in Baltimore, Maryland, and a self-proclaimed immunosurveillance supporter, is confident that the immune system will be shown to have an effect on those tumours as well. "I think it's a certainty," he says.

The results bode well for the development of therapies that modulate the immune system, such as vaccines against cancer.

Visit our responsevitalincancer.html">newsblog to read and post comments about this story.  

  • References

    1. GalonJ., et al. Science, 313. 1960 - 1964 ( 2006).
    2. CurielT., et al. Nature, 10. 942 - 949 ( 2004).