Clinical trial chiefs more likely tied to industry

NEW YORK (Reuters Health) - Cancer researchers who have the greatest ability to influence research are also the researchers with the greatest financial ties to the pharmaceutical and biotechnology industries, according to a report released today.

In a study, researchers found that cancer researchers who design clinical trials, analyze or interpret the data, or play other key scientific roles are four times more likely to have financial ties to industry than their counterparts who have lesser roles in these studies, such as recruiting study subjects or collecting the data.

Taking the lead role in conception and design, analysis and interpretation, or drafting the manuscript, affords substantial influence over the outcome of the study or the way the results are presented, Dr. Steven Joffe of Dana-Farber Cancer Institute, Boston and colleagues note in the Journal of Clinical Oncology.

Their findings stem from a look back at 235 drug trials published between January 2006 and June 2007 in the journal, which is the official journal of the American Society of Clinical Oncology. Sixty-four percent of the authors -- 1,881 of 2,927 -- said they performed at least one key role in the study and 842 (29 percent) reported at least one financial tie.

Authors who performed central intellectual roles in clinical trials were 4.3-times more likely than other authors to report financial relationships to industry. This relationship was present among both industry-sponsored and non-industry-sponsored studies, although it was stronger for industry-sponsored trials.

"What our study shows," Joffe told Reuters Health, "is that these financial relationships are not evenly distributed among investigators. Rather, they are concentrated among those investigators who are most centrally involved in the research, and who therefore have the greatest potential to influence the design, conduct or reporting of the research."

All of that means "the potential for bias appears to be even greater than was previously thought," Joffe warned.

Last year, the Institute of Medicine, one of the National Academies of Sciences that advises U.S. policymakers, issued a report stating, in essence, that doctors need to distance themselves from industry by voluntarily swear off lunches, drug samples and strictly disclose research funding to strengthen protections against financial conflicts of interest. (See Reuters Health report, April 29, 2009).

Despite efforts at reform, financial partnerships between influential physician scientists and industry remain commonplace, with up to 70 percent of authors of scientific articles or medical meeting abstracts disclosing financial relationships with industry.

The push in recent years for investigators to disclose their ties to industry is not enough, according to Dr. Harlan Krumholz of Yale University in New Haven, Connecticut.

"Disclosure of ties unfortunately cannot tell us whether the relationships are appropriate and whether they have led to trial decisions that have favored industry over the public good," Krumholz, who was not involved in the Joffe study, noted in an email to Reuters Health.

"In some cases, they may represent high integrity interactions between industry and academic scientists seeking knowledge about the effect of a treatment on patients. In other cases, they could represent low integrity interactions in which science and marketing commingle and academic scientists are unduly influenced through financial and non-financial benefits provided by industry," Krumholz said.

In a telephone interview with Reuters Health, Dr. Bernard Lo, director of the program in medical ethics at the University of California at San Francisco said not all types of relationships between researchers and companies are of equal concern.

"One type of relationship that is now being singled out is being on a speakers bureau for a company, where you give a talk for a company and they really control the content of the talk; they often give you slides, train you to give the company message and its not really (the speaker's) independent work. Knowing this might lead you to review their article a little more carefully."

"Also, the dollar value of the relationship is worth knowing. If someone is making a couple thousand dollars a year, are you really suspicious they are going to be unduly influenced by the company? But if it is $50,000 a year or $100,000 a year, then I think you might be a little more skeptical. But this type of information often isn't disclosed."

"Using the scientific literature to advance an agenda is always a concern when the author is tied to industry though a financial relationship," David H. Johnson and Leora Horn from the Vanderbilt-Ingram Cancer Center, Nashville, Tennessee wrote in a commentary published with the Joffe study.

Even if a uniform disclosure system is eventually developed, Johnson and Horn worry that "lengthy disclosures may become the publication industry's equivalent to the Transportation Security Administration Orange Alert sign now permanently placed at one's local airport -- there, but not really seen."

In essence, they believe it is "up to the reader to assess and judge" the quality of a published study.

SOURCE: Journal of Clinical Oncology, online January 11, 2010.