Now the Wall Street Journal has published a major editorial which gives it another whirl. Like some previous efforts, it raises methodological questions about the Nissen and Wolski meta-analysis,
Meta-analysis can be a useful contribution to medical knowledge, but it is in no way conclusive.
Several aspects of the Nissen study are particularly troubling. The assertion that Avandia raises the absolute risk for heart attacks by 43% relied on a methodology that excluded data in which there were no reported adverse events, skewing the results.
Besides, the data Dr. Nissen analyzed did not control for cardiac events, so things like indigestion and heartburn may have counted.
It appealed to authority by citing the Lancet,
Even the Lancet, the British medical journal with its own politicization issues, clucked about the NEJM's 'alarmist headlines.'
After denigrating the Nissen and Wolski meta-analysis, it touted GSK's latest clinical trial, which is still ongoing,
GlaxoSmithKline is conducting its own clinical trials, which are more rigorous, to be concluded next year. The interim results showed that Avandia patients had a 7% decrease in heart attacks, and a 17% decline in deaths from cardiac events, though these aren't yet definitive either.
It also praised GSK's transparency,
GlaxoSmithKline was already engaged in extensive postmarket trials before this scandal broke -- beyond those that would be required by Enzi-Kennedy. And Glaxo makes the data from its completed trials transparent on the Internet, where Dr. Nissen acquired it in the first place. So much for Mr. Waxman's repeated claims that Big Pharma is suppressing negative results.
Finally, the WSJ editorial concluded with assertions that the issue is political, not methodological, epidemiological, or clinical,
The larger political context here is the evidence of collusion to gin up one more drug-company "scandal." Dr. Nissen has a reputation as an adversary of the pharmaceutical industry, and was involved in the controversy that led Merck to withdraw Vioxx from the market in late 2004.
At bottom, the Avandia fuss is political, not medical, and it turns on risk itself: how to strike the best balance between patient safety and lifesaving therapies.
An interesting counterpart of the WSJ's arguments appeared in the British Medical Journal two days earlier, but apparently went unnoticed by the WSJ. (See Kazi D. Rosiglitazone and implications for pharmacovigilance: postsurveillance data should be systematically collected and publicly available. Brit Med J 2007; 334: 1233-4. Link is here.)
The WSJ raised doubts about meta-analyses in general, and the Nissen and Wolski article in particular. But the BMJ, while acknowledging meta-analysis is far from perfect, was far more measured,
Meta-analyses have unique strengths and weaknesses and this one is no exception. Its singular strength is the statistical power generated by data on 15 560 patients from published and unpublished trials. However, it includes clinically heterogeneous trials and criteria used by individual trials to classify adverse events are somewhat unclear. Only summary data are available in the public domain—for example, whether or not a person had a myocardial infarction, not when it occurred—which makes time to event analyses impossible. Also, the total number of adverse events was small, so that misclassification of a few events could alter the conclusions.
Although the WSJ seemed to put a lot more faith in the ongoing trial being run by GSK, the BMJ editorial noted that the results available from this trial so far do not undercut concerns about risks of cardiac adverse events due to rosiglitazone,
In response to the concerns raised by this meta-analysis, an unplanned interim analysis of a large, manufacture sponsored, randomised, open label, non-inferiority trial specifically designed to investigate the cardiovascular safety of rosiglitazone was recently released. Compared with patients taking metformin and a sulphonylurea, people taking a regimen that included rosiglitazone had no significant increase in the risk of myocardial infarction (hazard ratio 1.16, 0.75 to 1.81), although they had a significantly increased risk of heart failure (2.24, 1.27 to 3.97). When these new data are added to the trials in the previous meta-analysis, rosiglitazone is associated with an increased risk of myocardial infarction (odds ratio, 1.33; 1.02 to 1.72)
Moreover, even if one wants to entirely discount the Nissen and Wolski meta-analysis, there was another meta-analyis of clinical trials of rosiglitazone, done by GSK itself, which produced similar results, although its results were not quickly made public,
Of note, a similar meta-analysis submitted by the manufacturer to the EMEA and the FDA in August 2006 showed an increased risk in ischaemic events (hazard ratio, 1.31, 1.01 to 1.70)
The WSJ editorial simply ignored GSK's own meta-analysis.
Finally, although the WSJ claimed that GSK made its clinical trial data transparent on the Internet, it failed to mention that it did so only as a consequence of a legal settlement of a case which alleged the company had been less transparent about data about the side-effects of its antidepresant paroxetine (Paxil). (See Steinbrook R. Registration of clinical trials - voluntary of mandatory. N Engl J Med 2004; 351: 1820-1822, link here and our post here).
So the Wall Street Journal's new whirl has kept the Avandia spin cycle merrily revolving. This spin cycle continues to distract from what I think, in my humble opinion, are really the major questions.
The first question is clinical: what are the benefits and harms of rosiglitazone as a treatment of Type 2 diabetes, and therefore for which patients under what circumstances should this drug be used?
The second question is about policy: what barriers, if any, have prevented physicians and patients from getting the best possible answer to the first question, and what can be done about them?
I understand that WSJ editorials may want to protect an important industry that provides the drugs physicians need to help patients from excess political interference, and excess regulation that might decrease its innovation and efficiency. The pharmaceutical industry is very important, and does make major contributions to society.
But that should not blind us to poor industry practices. Continuing defense of poorly done clinical studies, failure to do needed post-marketing studies, and suppression of clinical research only makes the whole industry seem more "shifty," (see post here), and would ultimately only give more ammunition to those who advocate the most radical regulation or even nationalization.