Thursday, December 01, 2011

Should Researchers with Significant Conflicts of Interest Be Permitted to Conduct Clinical Trials on Products Made by the Relevant Company?

Yesterday, I discussed a study which reported the results of a clinical trial in which nicotine replacement therapy [NRT] sampling produced no significant increase in smoking cessation at six-month follow-up. Nevertheless, the study touted NRT sampling as a promising strategy for smoking cessation. I also noted that one of the authors had significant conflicts of interest with Big Pharma, having served recently as a paid consultant for a number of companies that manufacture or sell smoking cessation drugs. I suggested that there is at least the appearance that the conflict of interest may have biased the interpretation and reporting of the study findings.

Today I address an even more difficult question: Should this type of research be allowed in the first place? More specifically, should an investigator with a significant conflict of interest with a company be permitted to conduct a clinical trial in which a product of that company is being tested?

I am not questioning the presence of conflicts of interest in early-stage research, especially lab research that does not involve human subjects and in which a drug or product is in the development rather than clinical testing phase. In such research, it seems reasonable to conclude that the conflict can be managed adequately with appropriate disclosures and other measures short of requiring that the conflict be eliminated.

However, in a clinical trial, the product is being tested among human subjects and the results of the trial have a direct effect on the profitability of the product and the financial value of the company. For example, after a clinical trial on a nicotine vaccine demonstrated a lack of effectiveness, the stock value of Nabi Pharmaceuticals plummeted by 70%. In this situation, the consequences of bias introduced by a significant conflict of interest can be substantial.

In addition, because human subjects are involved and there is often more than minimal risk, negative consequences of bias introduced by a significant conflict of interest could potentially be experienced by these human subjects. As I reported earlier this year, a clinical trial of CT scanning for lung cancer that involved a researcher with a significant conflict of interest was cited by the New York Times as not being able to document the receipt of informed consent from 90% of the patients. That study involved substantial risk to subjects, as they received high doses of radiation that could potentially increase their cancer risk. Failure to obtain informed consent from each patient would constitute a huge ethical breach.

Because of the gravity of these concerns, many universities do not allow investigators with significant conflicts of interest to participate as investigators in clinical trials where a product of the company of interest is being tested. For example, Harvard Medical School classifies this situation as being "Not Allowable." At Boston University Medical Center, investigators who consult for pharmaceutical companies and receive compensation of $10,000 or more per year are generally not permitted to engage in clinical trial research on drugs made by those companies.

I agree with the approach that medical centers such as Harvard and Boston University have taken. I do not think that conflicted investigators should be conducting clinical trials where drugs made by the companies who pay them are being tested on human subjects in the later stages of drug development and testing.

It does not appear to me that this approach is being taken in a large number of studies on the effectiveness of NRT. I think this is problematic, as there is abundant evidence that the research literature is heavily biased in the direction of recommending a focus on the use of NRT even though the data show that unassisted smoking cessation is more effective and is more commonly used in successful quit attempts. To the extent that such bias is related to conflicts of interest with pharmaceutical companies, it is quite unfortunate.

The rest of the story is that research such as the study on NRT sampling which I discussed yesterday presents serious potential problems because they involve a financially conflicted investigator taking part in a clinical trial of a drug manufactured by a company/industry in which they have a financial interest. I do not believe that such a situation should be allowable and hope that in the future, more universities will adopt policies similar to those in place at Harvard Medical School and Boston University Medical Center.

No comments: