Dummy treatments used in clinical trials of cancer treatments can produce positive effects - but are unlikely to have a direct impact on tumours, research has found.

Placebo treatments such as sugar pills which should have no effect are used by scientists to provide a comparison with the effects of new drugs.

Substantial improvements in symptoms and quality of life are unlikely to be due to placebo effects Gisele Chvetzoff & Ian Tannock

In double-blind trials, some patients are given a new drug, and others are given a placebo - but none know which they are getting.

Even though patients who are given a placebo should show no signs of improvement, they sometimes do.

This is known as the placebo effect, and is possibly linked to the psychological boost a patient experiences if they believe - albeit mistakenly - that they are being treated with effective drugs.

Dr Gisele Chvetzoff, of the Centre Leon Berard in Lyon, France, and Dr Ian Tannock, of the Princess Margaret Hospital in Toronto, Canada, set out to discover whether the placebo effect could actually impact on the development of cancerous tumours.

However, they found that while placebos do seem to reduce symptoms such as pain and loss of appetite in some patients, they only rarely had any direct impact on the tumour itself.

Many studies

The researchers examined nearly 50 clinical trials that involved the use of placebos.

In some trials as many as 21% of patients given a placebo reported a reduction in pain, or decreased use of pain medications.

Up to 27% showed an improvement in appetite, and up to 17% actually gained weight.

However, only two trials examined by the researchers reported a direct impact on the development of the tumour in more than 2% of patients.

And it is possible that this apparent effect may in some cases be explained away by inaccurate measurement.

The researchers, writing in the Journal of the National Cancer Institute, say: "Placebos are sometimes associated with improved control of symptoms such as pain and appetite but rarely with positive tumour response.

"Our review suggests an upper limit on effects that might be expected from placebos.

"Substantial improvements in symptoms and quality of life are unlikely to be due to placebo effects."