Researchers who had financial relationships with GlaxoSmithKline or other manufacturers of antihyperglycaemic drugs were far more likely to take a favourable view on the risk of myocardial infarction with GSK’s Avandia (rosiglitazone) than those who did not, a US review has found.

Researchers who did feel Avandia aggravated the risk of myocardial infarction were far less likely to have any financial conflict of interest with manufacturers of antihyperglycaemic drugs than researchers who saw no increase in risk, the review published in the BMJ showed.

Moreover, in an era of “seemingly ubiquitous” requirements for disclosure of financial interests in medical journals, only 53% of the articles reviewed included a competing interest statement, noted the authors from the Mayo Clinical in Rochester, US.

Their own analysis found that 45% of the articles reviewed were written by authors with financial conflicts of interest. Of these 90 articles, only 69 (77%) included a statement disclosing the conflict of interest in the article itself, while three of the 21 articles that did not disclose the relationship published a statement declaring no conflicts of interest.

Although evidence of a “clear and strong link” between author orientation on the Avandia controversy and the existence of financial conflicts of interest with pharmaceutical companies does not “necessarily” show link is causal, the findings in general “underscore the need for progress in reporting in order for the scientific record to be trusted”, commented the Mayo Clinic team.

GSK itself underlined the company’s commitment to public disclosure both of clinical trial results and of information about payments made to healthcare professionals and researchers. It pointed out that fewer than 5% of the publications reviewed by the Mayo Clinic were original research while many were opinion pieces.

“It is important to note that the authors’ conclusions do not impugn the validity of the scientific data,” the company added.

Avandia debate

The Mayo Clinic team specifically looked at articles citing or commenting on two key publications in the debate over Avandia and cardiovascular risks:

- the meta-analysis by Nissen and Wolski published in the New England Journal of Medicine in June 2007, which first raised the possibility of an association between Avandia and cardiovascular events in patients with diabetes; and

- the subsequent interim report of the RECORD (Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycaemia in Diabetes) trial, which found no statistically significant difference in myocardial infarction, cardiovascular death or death from any cause between rosiglitazone combined with either metformin or sulfonylurea, and the active comparators of metormin plus sulfonylurea.

Among the 202 articles involving 180 unique authors that were reviewed by the Mayo Clinic researchers, 10 reported original research, 91 were letters, editorials or commentaries, and 101 were reviews, meta-analyses or guidelines.

On the risk of myocardial infarction (MI) with Avandia, just 31 (17%) of the 180 unique authors took a favourable view, with 26 recommending continued use of Avandia; 84 (47%) were neutral; and 65 (36%) were negative about the safety of Avandia, with 38 recommending against continued use of the drug.

Among the 31 authors who took a positive view on the Avandia/MI association, 29 or 94% had an identified financial conflict of interest with a pharmaceutical manufacturer, defined as being GlaxoSmithKline, the manufacturers of rival drug pioglitazone (Actos – Takeda and Eli Lilly) or a manufacturer of antihyperglycaemic agents other than rosiglitazone or pioglitazone. Of the favourable authors, 27 (87%) had a financial conflict of interest with GSK and 20 (65%) with Takeda/Eli Lilly (these two categories overlapped).

Financial conflicts of interest included a pharmaceutical company funding the article in question; the author being employed by the pharmaceutical company; a pharmaceutical company funding research other than that covered by the article in question; and the author being a consultant, advisory board member, speaker or lecturer for a pharmaceutical company, or receiving travel or honoraria from a pharmaceutical company, or owning stock in a pharmaceutical company.

Of the 65 authors who took an unfavourable view of the risk of MI with rosiglitazone, only 18 (28%) had a financial conflict of interest with a pharmaceutical manufacturer, with 13 (20%) found to have a relationship with GSK and 14 (22%) with Takeda/Eli Lilly. A total of 47 authors (72%) who were negative about the Avandia/MI association had no relationship with a pharmaceutical company.

Fuelling the fire

As the Mayo Clinic team observed, in the heat of the Avandia controversy patients and clinicians alike were exposed to a number of arguments on both sides of the debate. “How could interpretation of the same evidence result in disparate and impassioned positions?”, they asked. “We aimed to determine whether financial conflicts of interest with pharmaceutical manufacturers could be fuelling the fire. From our findings, it appears that the answer is yes.”

The researchers acknowledged that key opinion leaders who report competing financial interests may not necessarily be biased by those relationships, “but our findings indicate a strong association that users of the scientific literature should recognise”, they commented.

One way of addressing further the issue of MI and Avandia might be to control for investigator bias in research studies “as routinely as we control for other biases”, they suggested.

Checks and balances

GSK pointed to the “checks and balances” in place to ensure the integrity of scientific research, such as independent data monitoring committees in studies like RECORD. Like other pharmaceutical companies, GSK “works with researchers who bring knowledge and perspective from their experience and help advance the science around diabetes and other diseases”, it added.

The company is also committed to transparency both of clinical trial data and of financial relationships, it said. GSK either publishes clinical research involving its medicines as manuscripts in peer-reviewed journals or, failing that, posts the results on the GSK Clinical Study Register.

Starting in the first quarter of 2011, the company will disclose research payments to healthcare professionals and their institutions, with the first posting covering research payments made in 2010 to US healthcare professionals or their affiliated institutions for research studies that began on or after 1 January 2010. This policy will then be extended to healthcare professionals and institutions outside the US, with reports updated annually.