It's been 5 years since the then CEO of giant drug-maker Merck, Raymond Gilmartin, testified to the US Senate Finance Committee regarding evidence showing that Vioxx increases the risk of death due to heart attacks, strokes and other cardiovascular events.
In a Summary of Prepared Testimony, Gilmartin stated that prior to the APPROVe trial (a study looking at the protective effects of Vioxx on the risk for colon cancer), the research done by his company 'showed no difference in confirmed cardiovascular event rates between Vioxx and placebo'.
But a new and comprehensive data analysis published in the November 23, 2009 issue of Archives of Internal Medicine, looked at all deaths that should have been reported by Merck from 1999 (when Vioxx entered the market) to September 2004 (when it was pulled from the market, after APPROVe was halted due to health risks to participants). The analysis pooled patient data from every available published and unpublished study and looked at the information in revised ways. Specifically, measurements for death from any cause and the risk for cardiovascular events were calculated. Over 20,000 patients were included. The researchers found that an association between taking Vioxx and increased risk for cardiovascular events was evident 3 ½ years before the drug was pulled from the market.
Some of the data used in this research became available only when litigation against Merck made it public. This means that it was highly likely Merck withheld some or all of its less favorable study results during the time the drug was on the market. (Yes, they could do that legally...) But the study's authors disclose their ties to industry, as well. In their report, they reveal that every member of the team has, either currently in the the past worked as a consultant for the people who took Merck to court (plantiffs) over the Vioxx issue.
|
| (c) Anne Asher 2007 Licensed to About.com |
The Archives analysis evaluated the risk of death or the occurrence of a cardiovascular event in those who took Vioxx as it compared to those who took a placebo. Again, it used all studies, whether the results were good or bad for Merck. The analysis found that by the time the drug was off the market in September 2004, the risk was 43% greater for the Vioxx-takers. To be fair to Merck, the way this analysis was done was not standard industry practice.
One reason the study was done was to determine if Merck should have known (and made known the fact that) their drug posed serious risks to health when they first put it on the market. Another reason for the data analysis, according to its authors, was to help improve public health surveillance efforts for drug safety. The researchers believe the information from the analysis can be used to develop more targeted, comprehensive systems in the future.
I was interested in Merck's response to this new study, so I went to their website. Sure enough, a statement had been posted. Merck believes the researchers used unreliable methods which led them to reach incorrect conclusions. The company points out it had hired an outside firm to determine if the events reported by studies' investigators did indeed qualify as "thrombotic". They claim their method was "more rigorous than that used by the authors of the Archives paper". From my point of view, the possibility that they "hired their own people" to recommend ways to exclude some or all of the cardiovascular events that had occurred in Vioxx takers is decidedly there. And in that case, the risk for death when taking Vioxx would seem to be lower than it really is. But to be fair to Merck, unlike the statistical methods used in the study, hiring independent reviewers to look at the inclusion criteria applied to the cases is a standard practice for drug companies as they conduct their research.
Merck next complained that the researchers looked at "all cause" deaths. Originally, the researchers say, Merck's studies had not been designed around finding data on the risk of death or cardiovascular events due to taking COX-2 inhibitors. But the analysis is.
Because the researchers' mission was to help improve future public health surveillance efforts, perhaps another reason to calculate and report an "all cause" measurement was to provide an overall way to alert the FDA and the public when a the use of a drug shows a "disturbing trend", such as an increase in the death rate.
And finally, Merck complained that the researchers kept looking for cardiovascular events even after the patients had stopped taking Vioxx. They cite Alzheimers patients, who were followed for over a year later in some cases. In their statement, they say, "it does not make sense from a medical perspective to include events that occurred long after patients had stopped taking any medicine". But the company itself had specified that such information should be collected in the original trials. More significantly, ¾ of deaths reported in the analysis were Alzheimer's patients; these are patients who may well have had a problem communicating their symptoms to their doctors. Signs of heart attacks and strokes are not always obvious, and this justified following patients even after they had stopped taking Vioxx, according to the researchers.
Years ago, while this controversy was going on, I created a Vioxx Timeline. My research into FDA and other documents does indicate the evidence of an increase in heart attacks in patients who took Vioxx.
What's your opinion? Do you think the new study is too biased in favor of litigants against Merck? Or do you think the study has enough truth to counteract any potential bias? What do you think of Merck's response to the study? Has the drug manufacturer made valid points and that the new data analysis has reliability problems? Leave your comments below.
| A Vioxx Timeline | Will All Cox-2 Inhibitors Cause Heart Problems? | What are NSAIDs? |
Sources:
Merck. Merck Statement on Article in The Archives of Internal Medicine Involving VIOXX. Merck website. November 23, 2009. Accessed November 23, 2009.
Gilmartin, R.V., Summary of Prepared Testimony President, CEO, Merck and Co., Inc, before the US Senate Committee on Finance. Nov 18, 2004. Accessed Nov 23 2009.
http://finance.senate.gov/hearings/testimony/2004test/111804RGtest.pdf
Roxx, J.S., MD, MHS, et al. Pooled Analysis of Rofecoxib Placebo-Controlled Clinical Trial Data Lessons for Postmarket Pharmaceutical Safety Surveillance. Arch. Intern. Med. Nov 2009.
