Institutional Conflicts of Interest

Nearly a decade ago, an 18-year-old student named Jesse Gelsinger died in a gene therapy experiment at the University of Pennsylvania. It was run by a physician-professor who had invented the technology; started a company to develop it with the university's financial support; and used the company's cash to finance the clinical trial, which was conducted at the university. The disclosure forms signed by the patient and his family did not adequately reveal the risks involved.

In the wake of that ethical fiasco, the U.S. Department of Health and Human Services issued guidance to academic medical centers to adopt policies governing the so-called institutional conflicts of interest that exist when universities have a proprietary interest in a technology under development in its labs and clinics. But a new survey published in the Journal of the American Medical Association today found that by 2006, just 38 percent of the nation's academic medical centers had adopted such policies (30 of the 86 institutions that responded; there are about 125 medical schools in the U.S.). Another 29 are working on them.

Even more disturbing was the limited scope of the policies that were put in place. Most of the policies had strict guidelines for top officials at the universities. They had to reveal institutional conflicts of interest. But to whom? The study's authors noted:

The 2004 U.S. Department of Health and Human Services guidance urges institutions to establish policies for conflict of interest committees to provide information, recommendations, and findings to Institutional Review Boards [or IRBs, which are required by law to oversee clinical trials to ensure they adhere to ethical standards]. Substantial gaps exist in informing institutional IRBs of potential institutional conflicts of interest in research projects under review, other than of those of its own members. This is a disturbing finding, may be contrary to federal guidance, and calls for prompt institutional attention.

Note the implicit assumption in the study's finding. It's okay for IRBs to be staffed by physicians or administrators with conflicts of interest.

This raises an important question. Does disclosure of conflicts of interest to members of an IRB -- whether they exist within the institution, the management of the institution, or within the IRB itself -- ensure that the trial under review adheres to the highest ethical standards? Isn't it possible that those conflicts will influence the review? And doesn't the mere presence on the IRB of someone with a stake in the outcome of that trial create the appearance that the conflict of interest may have influenced a decision?

Appearances matter. What if something goes wrong in the trial, like what happened in the Gelsinger case. Won't someone point to the presence of an IRB member with a conflict of interest and suggest that it was due to lax oversight? Would an IRB or its institution want to see that allegation spread across the front page of its local newspaper?

In my view, when it comes to IRBs, anyone with a conflict of interest that relates to a particular trial under review should recuse themselves from overseeing and making any decisions about the ethical validity of that trial. Surely universities can find people without conflicts of interest to sit on these IRBs, which are a crucial firewall for protecting the safety of participants.

Meanwhile, an accompanying editorial by David Rothman of Columbia University's Center on Medicine as a Profession ignored the weaknesses in the policies being adopted by the minority of institutions that have adopted them. Instead, he raised the specter of mandatory federal regulation if the slow-moving universities don't get off the dime and enact policies of their own.

"Will government regulation step in to fill the vacuum? Current federal and state interests in industry-academy relationships provide reason to believe so. Congressional hearings are addressing the implications of industry support for continuing medical education, gifts to clinicians, the sale of physician-prescribing data, and pharmaceutical company efforts to intimidate researchers critical of their products. Currently eight states and the District of Columbia have laws or resolutions affecting marketing of pharmaceuticals. Thus, it may not require another Gelsinger-type incident to bring about significant regulation and legislation."

It's not clear that Congress will pursue the issue. Just a few weeks ago, the HHS inspector general issued a report lambasting the National Institutes of Health for failing to oversee or even collect data on the conflict of interest policies that grant-receiving institutions are supposed to enact for investigators that get the NIH grants. The agency responded by rejecting the IG's suggestion that NIH collect conflict-of-interest disclosures from universities. It would appear you can lead the government agency to water, but you can't make it drink.

There is a solution, of course. Forget about "guidance." Congress could step in and pass a law that prohibits investigators or university officials with conflicts of interest from sitting on IRBs and requires NIH to collect conflict-of-interest disclosure data from grant recipients and post it on its grants database (it's known as the CRISP database). To enforce the law, NIH could cut off all federal grants to that institution if those policies aren't adopted by a date specified in the legislation.

That would get medical schools' attention, and provide institutions, IRBs, investigators, journal editors and the public with a useful tool for managing conflicts of interest.

Posted by gooznews at February 13, 2008 07:46 AM