Results of breast cancer drug trials are routinely 'spun' to make the treatments appear more beneficial than they actually are, doctors claim today.
Researchers frequently downplay negative findings and overemphasise positive results - even when those results are much less important to the patient - claim experts who have reviewed studies.
Academics are also guilty of burying reporting of serious side effects in the smallprint of their articles, according to the Canadian doctors who examined the issue.
How well such scientists report their own research in medical journals matters, because it influences whether doctors opt to use certain drugs, they say.
But Professor Ian Tannock, a medical oncologist at the Princess Margaret Cancer Centre in Toronto, who led the research, said scientists too often painted an overly rosy picture of their drugs for their own ends.
He said: "Better and more accurate reporting is urgently needed."
He and his colleagues, who looked at the way results of 164 large scale (phase 3) trials were reported in journals, found widespread evidence of researchers largely ignoring unfavourable results.
All trials have what is called a 'primary endpoint', the principal test of whether a treatment has worked or not, that the trial is designed to address.
An example would be: 'Does new drug X help patients with advanced breast cancer live longer than existing drug Y?'
Trials also have 'secondary endpoints' that are of interest but not the principal reason for the study. For example, 'Does new drug X extend the time in which tumours are 'stalled' without growing further, compared to drug Y?'
The Toronto team found that, in a third of all trials that failed to achieve the 'primary endpoint', reports focused on these other, less important outcomes.
Some also changed what this primary objective was halfway through, possibly because early results indicated the trial was going to fail.
Writing in the journal Annals of Oncology, Prof Tannock and colleagues concluded: "These reports were biased and used spin in attempts to conceal that bias."
They also claimed to have found bias in the reporting of side-effects in two-thirds of paper, with adverse events relegated to the small print of articles, particularly when a trial met the primary endpoint.
They reasoned: "A possible explanation for this could be that the investigators, sponsors or both, prefer to focus on the efficacy of the experimental treatment and downplay toxicity to make the results look more attractive."
"Richard Francis, research manager at Breakthrough Breast Cancer, said: "It is crucial that all clinical trial results are reported fairly, including all negative results and possibly harmful side effects of any new treatments."
But he added 'secondary endpoints' remained important.
"While it is imperative that survival remains the main aim in these trials, we know from speaking to women who have suffered from breast cancer how important other factors are, including quality of life, side effects and the extended time a patient is able to control the growth of their disease," he said.
"We strongly believe that making clinical trial data freely available would decrease the potential for 'spin' by allowing more researchers to investigate the results.
"This will ensure trials are accurately assessed by regulatory bodies, allow comparisons with other studies and prevent duplication of trials."
Stephen Adams - telegraph.co.uk
