Conflict of interests

Scientists who collaborate with industry can find themselves gagged, says David Weatherall

Mounting pressures on scientists by government and funding bodies to develop links with industry and to pursue work likely to be of commercial value are generating a research environment fraught with pitfalls.

Just how badly things can go wrong I discovered myself a few months ago when problems emerged with a study at the Hospital for Sick Children, Toronto. The American cancer researcher David Nathan and I have a long-standing scientific collaboration with the hospital's Nancy Olivieri, who is also a professor at the University of Toronto.

In 1989 Dr Olivieri and her American colleague Gary Brittenham, supported by the Medical Research Council of Canada, began trials to assess the safety of the drug Deferiprone in children suffering from the hereditary blood disease thalassaemia. In 1993 the Canadian pharmaceutical company Apotex provided further funding for the research.

Patients suffering from thalassaemia undergo regular blood transfusions coupled with the use of drugs to prevent death from iron overload. But the most effective drug discovered so far is expensive and difficult to administer. A major research goal is to develop a cheap and effective drug that can be taken in tablet form. Deferiprone is the closest scientists have come to such a drug - and has been beset by problems.

Dr Olivieri's trials indicated that deferiprone failed to prevent iron overloading in a large number of patients, which might also put them at greater risk of heart disease and early death. She also uncovered other potential toxic effects. Much more research needs to be undertaken before the drug is made widely available.

Conflict arose during the drug trial, when Apotex threatened Olivieri that if she published her concerns, the company would take legal action. Bravely, Olivieri, knowing that the drug was already being widely used on "compassionate" grounds, ignored the warning and presented her findings at scientific meetings and in refereed journals.

Unfortunately, the hospital and the University of Toronto failed to support her through this difficult period and relations between the scientists in Toronto who supported Olivieri's actions and the hospital and university administrators broke down completely. Eventually, Olivieri was removed as director of the highly successful Toronto Haemoglobinopathies programme.

The scientific community also made mistakes in this case. One international scientific society accepted research papers written by one of Olivieri's collaborators and Apotex employees. Olivieri was not shown the abstracts before submission even though they were based on her data, which was reinterpreted by the authors.

Olivieri has now been reinstated to her post, following a campaign by many eminent scientists, and a complicated set of negotiations between the president of the university, David Nathan, myself and the hospital. Although peace has been partly restored, and a world-class research programme rescued, many questions remain. It may require a thorough external review of the hospital and university's handling of industrial interactions before the research centre that gave the world insulin can rest easy again.

The entire biomedical community can learn from this unfortunate affair. In any agreement involving research on patients between a clinical scientist and a company, the scientist must have the freedom to tell the patients and the scientific community about any concerns they have regarding the damaging effects of the drug or procedure.

If a company takes legal action or otherwise harasses a scientist in a bid to stifle academic freedom the hospital or associated university that employs the scientist must offer support. Companies that disagree with academic collaborators should do so under the conditions of open scientific debate, or at least via a panel of independent experts.

And learned societies must carefully investigate the origin of research papers offered to them and make sure that there are no conflicts of interest among those involved in collecting and writing up the data. Such conflicts must also be addressed by those universities and teaching hospitals where this kind of work is undertaken.

Sir David Weatherall is regius professor of medicine, Oxford. A longer version of this article appeared in The Lancet.

Join the debate on www.netnexus.org/ext/soapbox or email us at soapbox@thes.co.uk and come to our conference on Making Money from Ideas, see page 12.