Will The Crestor Study Sell More Cholesterol Pills?
25 CommentsBy Ed Silverman // November 9th, 2008 // 9:47 am
The results of AstraZeneca’s hotly anticipated Jupiter study - which measured levels of a protein called CRP that can indicate arteries are inflamed and point toward heart disease - were just released at the American Heart Association meeting and they are likely to stir as much debate as promised.
The cholesterol pill reduced the risk of cardiovascular death and heart attacks by 44 percent compared with patients on a placebo. The 17,802 participants were men over 50 years old and women over 60 years old - ripe ages for heart attacks, but who were healthy. They did not display signs of heart disease and had cholesterol levels not requiring meds under current guidelines. The study is available in The New England Journal of Medicine.
Other findings: the combined risk of heart attack, stroke or CV death was reduced by 47 percent; the risk of heart attack was cut by 54 percent; the risk of stroke was cut by 48 percent, and total mortality was reduced by 20 percent. In its statement, the drugmaker says 25 patients would need to be treated to prevent one major CV event, if the results are projected over a period of 5 years.
The results prompted some docs to say Jupiter ushers in a new era of care. “This takes prevention to a new level,” Doug Weaver, head of cardiology at Detroit’s Henry Ford Hospital, who wasn’t involved with the study, tells The Wall Street Journal. “It defines a new population” of patients at risk. Of course, if many docs act accordingly, this could ring registers for others selling statins (with the likely exception of Merck and Schering-Plough, which sell the controversial Vytorin).
But there are caveats. For one, there is disagreement about the utility and reliability of the CRP test. And the study yielded a higher rate of diabetes, which Sanford Bernstein analyst Tim Anderson wrote in an investor note may “temper enthusiasm, at least slightly, for more widespread adoption of cholesterol reduction therapy in patients with normal LDL (cholesterol) levels, but high CRP levels.”
And in an editorial in The New England Journal of Medicine, Mark Hlatky of Stanford University’s School of Medicine writes that “the results…raise two important questions about the primary prevention of coronary disease. Should indications for statin treatment be expanded? And how should measurements of (CRP) be used?”
“The relative risk reductions achieved with the use of statin therapy…were clearly significant. However, absolute differences in risk are more clinically important than relative reductions in risk in deciding whether to recommend drug therapy, since the absolute benefits of treatment must be large enough to justify the associated risks and costs. The proportion of participants with hard cardiac events in Jupiter was reduced from 1.8 precent (157 of 8,901 subjects) in the placebo group to 0.9 percent (83 of the 8,901 subjects) in the (Crestor); thus, 120 participants were treated for 1.9 years to prevent one event…
“There are also no data on the long-term safety of lowering LDL cholesterol to the level that was attained with rosuvastatin in JUPITER, which is lower than in previously reported trials…Finally, the cost of Crestor (roughly $3.45 per day) is much higher than that of generic statins…
“Jupiter provides yet more evidence about the effectiveness of statin therapy in reducing cardiovascular risk, even among persons who would not currently be considered for pharmacotherapy. Guidelines for primary prevention will surely be reassessed on the basis of the JUPITER results, but the appropriate size of the orbit of statin therapy depends on the balance between the benefits of treatment and its long-term safety and cost.”
For more background, you can read the initial reports of the study results and reactions in The Wall Street Journal, Bloomberg News, Forbes, The New York Times, Reuters and the BBC.
UPDATE: We should note that Paul Ridker, the lead investigator who works at Brigham and Women’s Hospital, is listed as a co-inventor on patents for the $20 test that measures CRP, and that the trial was funded by AstraZeneca.
UnitedStates1776
I would like to see a comparison study done between patients taking a statin, patients performing 30 minutes of aerobic exercise daily, patients who did both and then a control group of patients doing neither.
PaulGGG
These studies have been done. Patients on aggressive diet and exercise were able to lower their cholesterol 10-15%, which is a good thing.
Several problems:
- People at high risk have cholesterols of 200 and 300, so a 10% reduction to 170 to 270 (10%) is not very impactful based on epidemiology studies.
- Most people don;t stick to diet and exercise
Now, to be fair, of course diet and exercise are good and we should all do it for lots of reasons, and of course, drugs come with cost and risk.
However, here we have a class of drugs that has been studied at nauseum and with almost every single study being impressive. Hard to dismiss this, even for the harshest critics.
Justice in MI
Can a better math person than I am help with the following:
The article paraphrases the study, “In its statement, the drugmaker says 25 patients would need to be treated to prevent one major CV event, if the results are projected over a period of 5 years.”
So would you call that an NNT of 25 or of 125 or something else?
If anyone has the wherewithal to translate the apparent diabetic risks into NNT that would also be much appreciated. And also NNT on total mortality.
Finally, I’m a bit confused about whether we’re talking about CV deaths and/or MIs and strokes survived? Weren’t there results in both categories?
Linda
The truth is being ignored.
Levels of cholesterol in the blood stream are far less important than the condition and integrity of the blood vessel walls. The arteries/veins in many people are becoming weak and porous due to poor diet and nutrition, ie: studies are coming to light showing high levels of insulin at a regular rate are damaging to arterial walls (one piece of the puzzel). If the arterial walls are healthy, strong, and elastic, there is little likely hood for atherosclerosis to occur. Cholesterol is simply seeping into unhealthy arteries and sticking, not because of the levels of cholesterol, but because of the poor conditions of the arterial walls.
We need more studies of this part of vascular health and less that pander to drug companies so that they can sell more drugs. I am sure they are in economic crisis also, as people have less ability to buy drugs on a regular basis.
Lets look at our poor food chain, indrustrial farming practices, too many subsidies for corn for instance. Many of our foods are over processed and unhealthy, duh. Much of the ‘mainstream’ food sources cause our bodies to overproduce insulin, not just sugar, though sugars are a big part of it. We simply need to begin staying away from the bad stuff so that they will stop producing it. If we demand purer natural foods, and that is all we buy, they will meet that need, albeit slowly, but eventually. Got to start somewhere.
Cobalt
Merck would still make money from Zocor, the presumed ’safe and effective’ half of Vytorin.
Long term reduction of mortality data would be helpful, though cost/benefit analysis taking the diabetes risk into account, as JIM suggests, should be done before treatment recommendations are made.
Condor
More advice — from experts — suggesting that we not OVER-apply the admittedly-impressive findings from Crestor’s “JUPITER” — @ the AHA, and NEJM, this morning:
http://shearlingsplowed.blogspot.com/2008/11/but-before-you-go-long-on-az-pfe-or.html
Even so, this does shift the shore-line — on higher-risk patients. . . away (again!) from Vytorin/Zetia.
Namaste
ryan
According to the New York Times (link is above): “In the Jupiter study, in which people either got rosuvastatin or a placebo, there was no increase in muscle or kidney problems for those taking the statin.” How could there be no increase in muscle problems? I know several people who have had bad muscle side effects from statins.
Anonymous
Justice
http://content.nejm.org/cgi/content/full/NEJMoa0807646
Copy and paste into your browser, not sure why reliance on articles for data, when you can go to the source.
NNT 25, 4 yrs projected due to study being completed early. 1.9 yrs. Consider it a running rate. NNT in 2 years was 95 meaning treating 95 patients on Crestor would result in 1 fewer events. The significant difference in the 2 numbers is that the arms were projected based on rate of events in a study stopped prematurely due to unequivical benefit in the treatment group. (Simply unethical to withhold therapy in the placebo arm.) Generally speaking ATP guidelines are based on the risk of an event in 10 years, and an algorythmn of therapeutic lifestyle changes and pharmachologic therapy are recommended to reduce the risk of said events. So NNT being 25 for four years is the projection based on the anticipated length of the trial. I may also note that the trial was not time bound but rather event driven. If x amount of events occured within the projected time frame, a correlated risk reduction would be considered significant as per the prespecified criterior of the trial.
This is a very significant trial, Not necessirly strictly correlated to CRP. One of the greater aspects is the # of women treated and the subanalysis of various treatment groups. You may note that even in the subanalysis of patients that had relatively low CRP, the results indicated no significant difference in the benefits for that group than the overall population of the trial. The demographics are quite representative of the patient population in any physicians practice. Inclusion criterior dictated a CRP of greater than 2, Normal range for reference is 1-3 Mean baseline 4.2 Crestor/4.3 Placebo.
Look at the chart on page 6, note that in the placebo group the mean LDL did not change over the 2 year period, and consider in this study the treatment group had roughly a 50% reduction in risk of events. Staggering based on pre=Jupiter knowledge.
Anonymous
Sorry for any confusion, Table 2 in the abstract. page 6 in the pdf file.
Ed Silverman
Hi Anonymous,
I actually provide the link directly to the NEJM study in the second paragraph. Just click on the words ‘the study,’ which are highlighted in green. I try to provide such links whenever possible, regardless of the underlying documentation involved.
Regards
ed
stata
Ed:
Your a hack. You ommited key context that even npr got right. Ridker and Harvard have a crp lab test related patent.
http://www.npr.org/templates/story/story.php?storyId=96784439
PaulGGG
You people are mixing this up here. Zocor, the simvastatin part of Vytorin was shown to have pretty much the same effect. Also, Vytorin is not unsafe, just that it didn’t show a positive effect in that badly designed study on carotid intima thickness. That makes it not effective by one measure, but not unsafe.
Crestor is a good product, but has shown big problems at high doses.
Bottom line, all these products are fine when used appropriately and not pushed to high doses.
Justice in MI
Thanks, Anonymous. I did look at the study that Ed linked, but I was having a hard time getting NNT from that - perhaps I just missed it or it was a question of ‘translation.’
Anonymous
Sorry Justice and Ed, I had reviewed the study prior to visiting this site, I must have overlooked the link.
Doc
It will sell more generic simvastatin.
Ed Silverman
Hi Stata,
I did not deliberately omit the info concerning Ridker. If you note, the disclosure is mentioned in the study, and I provided a link to the study itself in the second graph of the post.
If I erred, it was assuming everyone would notice that info, since it was provided in the study, as opposed to highlighting it myself.
Regards
ed
Dan A.
CRP is primarily elevated upon an inflammatory reaction that may be caused by many medical conditions. So CRP being a CV risk factor is understandable, yet could be indicative of another disease state, such as a microbe invasion. Makeres of HTN meds have and are doing studies to prove a correlation between elevated CRP and the development of … Read Morehypertension as well. Yet CRP overall is a non-specific prognostic indicator due to the numerous etiologies that could be attributed to the CRP elevation, and as such, many etiologies should be ruled out before it is presumed to be solely attributed to CV dysfunction.
Condor
Hey PaulGGG — this has been discussed to death here and elsewhere — but the official jury (the FDA) is still out on the elevated risks of cancer (found in SEAS) from Vytorin/Zetia.
Cheers!
Nathan
Interesting that even Steve Nissen, ever the pharma critic, seems to view this trial favorably. Here’s a quote from the NPR article:
“That is a larger reduction in the things that we care the most about — morbidity and mortality — than we’ve ever seen in a statin trial,” says Steven Nissen, who chairs the department of cardiovascular medicine at the Cleveland Clinic. “And it occurred very quickly, in only 1.9 years of average duration of treatment.”
Insider
NNT = the reciprocal of the absolute risk reduction.
Gooze has a nice piece:
http://www.gooznews.com/archives/001243.html
Lets cut thru the CRP!
Outside the Box
These data seem to be very positive for statin use - especially in providing physicians with a strong argument to get patients to continue on therapy once their cholesterol is down to normal. As for approval for use of statins in the normal cholesterol / high CRP group, I wonder if anyone has actually explained the mode of action here? Is it possible that statins are actually vascular anti-inflammatory agents and that their effect on cholesterol is actually a side-effect? I may be wrong here, but doesn’t the body produce cholesterol as part of its anti-inflammatory process? Whatever the MOA is, isn’t it going to be necessary to demonstrate it before the FDA will approve the CRP indication? Are there other diagnostic methods of identifying the presence of vascular inflammation that would support or replace CRP as the marker?
Either way, it is hard to see these data being anything other than beneficial for the entire class of statins.
Justice in MI
Nathan - on McNeil/Lehrer, Nissen even called the results a ‘blockbuster.’
Perhaps he is, indeed, running for Commish.
Justice in MI
I’d second Insider’s suggestion to read Gooz on this. It is the kind of analysis that led to my questions about NNT in the first place, not only about CV risk reduction but about diabetes risk increase. In terms of relative risk, they are fairly close.
Jim
I read the post on Gooz and it answered some of my questions and made some valid points.
I also went to the NEJM site and looked over the various reports.
The reports show that among other things, individuals were identified as smokers; having metabolic disorder, or having a family history of CVE.
I must admit I am a statistical illiterate and am curious as to how many of the individuals who suffered a CVE during the span of the study came from the above groups?
Jack Friday
My bottom line:
Good result for a “not needed” statin. James Stein and Jon Keevil of the University of Wisconsin-Madison calculated it would cost about $408,000 to save a life using Crestor, compared with $24,000 using a generic.