Study Confirms Vioxx Causes Heart Risks
18 CommentsBy Ed Silverman // October 13th, 2008 // 7:08 pm
An analysis of an infamous Vioxx study found the notorious painkiller does, indeed, double the risk of heart attacks and strokes, although the likelihood of a serious cardiovascular event lessened one year after people no longer took the pill. What’s more, other drugs in the same class, known as Cox-2 inhibitors, may pose the same risk, according to the analysis published in The Lancet.
“The good news is the data suggests that the risk doesn’t persist forever. The risk goes back toward normal after a year of follow up,” Robert Bresalier of the MD Anderson Cancer Center at the University of Texas, one of the study authors, tells Reuters.
The original study, known as Approve, was funded by Merck and designed to determine whether Vioxx could prevent polyps that increase the risk of colon cancer. Instead, the Approve study caused concern that Vioxx increased cardiovascular risks and prompted Merck to withdraw the pill in September 2004.
Interestingly, a 2005 analysis by Bresalier and other researchers that was published in The New England Journal of Medicine suggested it took 18 months for Vioxx to increase the risk of heart attacks and strokes. As Reuters notes, that time frame played a big role in Merck’s legal defense, although the journal later posted a correction on its Web site, saying the difference was not statistically significant.
The new analysis, done with independent statisticians, suggests the risk occurs early and persists, according to Bresalier. “This data shows you can’t precisely determine the timing of the risk. It does appear to start relatively early,” he tells Reuters.
The Approve study looked at the effects of three years of treatment with Vioxx in 2,587 patients, who were checked for side effects while on the drug and two weeks after they stopped, and included one-year follow-up data on patients who stopped taking the drug because of cardiovascular side effects. “In essence, the relative risk remained about the same,” Bresalier tells Reuters.
People who took Vioxx in the study had about the double the risk of having a heart attack or stroke than those who took a placeob.
In a statement, Merck says the overall findings were consistent with the Approve study, but not all patients were able to be followed after they stopped taking the drug and that risk factors for cardiovascular events may have changed after patients stopped taking the drug. Research “using limited data from a prematurely terminated study needs to be interpreted very cautiously and in the context of the rest of the data from the extensive clinical development program for Vioxx.”
Meanwhile, Bresalier tells Reuters that a number of studies since the initial safety warning on Vioxx suggest other Cox-2 inhibitors, including Pfizer’s Celebrex and non-steroidal anti-inflammatory drugs (NSAIDS) such as ibuprofen, also carry a higher risk of heart trouble.
“I think the preponderance of data now does suggest this is a class effect,” he tells Reuters. “There seems to be an increased risk for most if not all non-steroidal anti-inflammatory drugs. That doesn’t mean these aren’t good drugs. It’s just there is some increased risk we need to be aware of.”
Justice in MI
““I think the preponderance of data now does suggest this is a class effect,” he tells Reuters. “There seems to be an increased risk for most if not all non-steroidal anti-inflammatory drugs. That doesn’t mean these aren’t good drugs. It’s just there is some increased risk we need to be aware of.”
We have been hearing this from Merck and its defenders for some time. Are there any actual studies comparing the relative CV risks of Vioxx versus other NSAIDS (other than VIGOR with naproxen, and Graham’s Celebrex/Vioxx study)?
coiwatch
Baigent discloses “a grant for investigator-initiated research from Merck.” Is the delicate term “a grant” commensurate with the conflict of interest that might be created by 105 million pounds of funding from Merck to the Clinical Trial Service Unit which employs Baigent?
Ed Silverman
Hi COI Watch,
One advantage of linking to the published study is that we get to see the conflicts that are disclosed. And there are several among the study authors involved here. Does that affect his interpretation? I don’t know, but at least we have some additional insights to understand some of the circumstances that may inform his views.
Regards,
ed
Andrew
Straight Cox-2s, yes! All NSAIDs, very doubtful. Aspirin, the original, obviously has good effects. naproxen has some good data. As for Vioxx and celebrex, you couldn’t pay me enough to take them!
Justice in MI
Again, in the interest of clarity, is there any data anywhere that reasonably suggests that non-cox2 selective NSAIDS have a CV risk that is comparable to any of the cox-2s?
Justice in MI
Not having received an answer to my question above, I take it there is no such data. (I understand that ibu and other NSAIDs have been associated with HTN).
Thus the suggestion from Merck that “the preponderance of data now does suggest this is a class effect” - in which class means most NSAIDS (selective and non-selective) - appears to be yet more self-serving spin.
If data exists which genuinely suggests otherwise, still intersted to hear of it.
Doug Bremner MD
You are right JiM. That statement was simply false.
Nathan
Doug, are you sure? JiM did you read the Lancet article? Here’s a quote from the introduction: “In September, 2004, the trial was stopped early on recommendation of its data safety and monitoring board because of concerns about cardiovascular toxicity,5 and rofecoxib was withdrawn from the worldwide market. A few months later, other selective COX-2 inhibitors, celecoxib6 and parecoxib with valdecoxib,7,8 were shown to have similar toxic effects.”
Here are references 6,7,and 8. The Lancet article implies that these references show cardiac tox for other selective COX-2 inhibitors. I haven’t read these yet, so I’m taking the authors at thier word.
—————–
6 Solomon SD, Pfeffer MA, McMurray JJ, et al. Effect of celecoxib on cardiovascular events and blood pressure in two trials for the prevention of colorectal adenomas. Circulation 2006; 114: 1028–35.
7 Nussmeier NA, Whelton AA, Brown MT, et al. Complications of the COX-2 inhibitors parecoxib and valdecoxib after cardiac surgery. N Engl J Med 2005; 352: 1081–91.
8 Ott E, Nussmeier NA, Duke PC, et al. Efficacy and safety of the cyclooxygenase 2 inhibitors parecoxib and valdecoxib in patients undergoing coronary artery bypass surgery. J Thorac Cardiovasc Surg 2003; 125: 1481–92.
atlex
JiM,
Remember, not everyone can spend most of the day online. There is evidence that some of the NSAIDs carry risks equal to some of the COX-2s, depending on the dose. Most retrospective analyses show, for instance, that celecoxib used at normal doses (200mg or less/day) carries a risk no greater than ibuprofen, etc. Of course, if I remember correctly, celecoxib is less COX-2 specific than Vioxx; that might have something to do with it.
By the way, I’ll look for the sources of this information and pass them along when I have a chance.
Atlex
atlex
JiM,
Here’s one reference courteous of Ed.
http://www.pharmalot.com/wp-content/uploads/2008/02/hennekens-cox-2.pdf
This book is certainly not closed on this controversy, but it would be an overreach to attribute Vioxx’s troubles to the whole class.
Atlex
Justice in MI
Nathan - yes, the lancet piece is a comparison among cox-2s. My question was a comparison between cox-2s and non-selective NSAIDS.
As Graham’s study suggests, there may, indeed, be a difference among the cox2s. As far as comparison with non-selective NSAIDS, I am ready to keep the book open. I would appreciate it if others did the same, which I think the Merck spokesperson clearly was steering us away from.
Re: the study Atlex cites, the comparison with ibu was 800 mg, tid, certainly an extremely high does in the scheme of things. So, in my view, this study contributes little to the discussion as it would impact the average person.
Nathan
JiM,
Now I understand your question. However, think about it a little. NSAIDS inhibit COX-1 and COX-2. Since the selective COX-2 inhibitors are showing cardiac tox, one would fully expect that nonselective inhibitors would do the same. The only caveat might be if you somehow believe that hitting COX-1 is cardioPROTECTIVE and counterbalances the ill-effects of COX-2 inhibition.
Nathan
Here’s a primary literature reference. I didn’t read the whole thing, but it seems to address your question.
http://archinte.ama-assn.org/cgi/content/full/160/6/777
Justice in MI
Interesting study, Nathan. CHF is a somewhat different issue, but certainly relevant. Perhaps ironically, this 1999 piece ends with this:
“It is possible that drugs that are selective inhibitors of the inducible cyclo-oxygenase 2 will have a lower rate of adverse effects on the kidney and cardiovascular system, but this remains to be established in well-designed pharmacoepidemiological studies.”
Re: the overall logic, my lay understanding is that the relevant balance of thromboxane, prostacyclin, etc. and their relationship with amount of cox 2 is complex. Thus, the study that Atlex cites suggests that naproxen, even at high doses, has a significantly lower rate of CV AEs than either other high dose NSAIDS or coxibs. Since naproxen is non-selective - and thus I assume inhibiting of both cox’s - there would seem to be more going on than the amount of cox2 alone.
Nathan
JiM,
Actually, there appears to be quite a bit of meta-analysis to show that NSAIDS increase cardiac risk. You are asserting that the Merck spokesman was just “spinning” decent PR out of a bad result. From what I could read, NSAIDS aren’t as bad as COX-2 inhibitors, but there is pretty clearly an increased risk. Moreover, remember that the selective COX-2 inhibitors are under patent - and therefore there are ongoing clinical studies. I don’t think that there is a lot of money being sunk into clinical trials of generic NSAIDS. Therefore, most analyses are retrospective in nature and therefore a little less credible. Here’s just a few links I found: (I’m at home, so I don’t have full-text for the journals — so news releases will have to suffice)
http://www.theheart.org/viewArticle.do?primaryKey=501247
http://www.arthritis.org/cvd-nsaid.php
http://www.theheart.org/viewArticle.do?primaryKey=671815
Justice in MI
Thanks for taking the time to post these, Nathan. They are useful and informative.
I do think there is a certain amount of spin going on in the way the implications data we have are being conveyed, but perhaps in the eye of the beholder.
atlex
JiM and Nathan,
I want to remind both of you that this may not be a black and white issue that a drug is cardio toxic or not. It could well be that in high doses, most or all of these might be cardio toxic. The key is whether they are safe at therapeutic doses. JiM, you correctly indicated that in one study, 800mg TID of ibuprofen, has demonstrated some cardio toxicity. This is not an uncommon dose; thus, there should be concern. It could well be that at therapeutic doses, refecoxib is toxic, while celecoxib is not.
Atlex
Justice in MI
Thanks, guys. I have learned from the exchange. Agree that having the right comparators is key and that there is, indeed, much to learn about “class effect” and what is, and isn’t, in “class,” however defined.
I’ll leave the dead horse beyond saying, when spokesperson suggests “class effect” in general way, average person would interpret that to mean that the two or three Advil they are taking per day (600 mg) might have the same risks/benefits as the 400 mg. of Celebrex. That may turn out to be the case, but I’m not sure there is a “preponderance of evidence,” as spokesperson said, that points in that direction.