Jupiter Study, C-reactive Protein and Statins

[James L]

A new study of the use of cholesterol-lowering drugs known as statins was presented November 9, 2008, at the American Heart Association convention in New Orleans. The study involved nearly 18,000 people worldwide and was led by Dr. Paul Ridker, director of the Center for Cardiovascular Disease Prevention at Brigham and Women's Hospital in Boston. The study found significant benefits from statin treatment in individuals with high levels of high-sensitivity C-reactive protein, or CRP, which indicates inflammation in the body.

Dr. Elizabeth Nabel, director of the U.S. National Heart, Lung and Blood Institute, stated that, "It's at a minimum an extremely important study and has the potential to be a landmark study."

The study indicated that statin treatment was beneficial in reducing the risk of heart attacks and related problems.

The study was sponsored by the manufacturer of the statin drug, Crestor.

Dr. Mark Hlatky, Stanford University, commented that the study did not answer some questions about risks of giving statins to relatively healthly people.

More details in this IHT article (http://www.iht.com/articles/2008/11/09/healthscience/10heart.php).

[laughingW]

An excellent critical review of the study. Thanks to Peter at Hyperlipid who pointed to it.

Junkfood Science

http://junkfoodscience.blogspot.com/2008/11/when-news-sounds-too-good-statins-new.html

[nrobles]

That is such a powerful article. Thanks for sharing it. Gary Taubes explains the pharmaceutical companies commitment to "studies" that herald their drugs keeping out the downside of potential toxicities related to them. Anybody seen any study on the effect of LC and CRP levels?

[Mitra]

Dr Mike has written about this too:Truth versus hype in the Jupiter study (http://www.proteinpower.com/drmike/cardiovascular-disease/1853/).

[skyhawk]

I went into deep study of the JUPITER trial. I went over it on my own with a fine tooth comb.

In a nutshell here is what I found:

Crestor did not perform any better than any other statin in preventing heart attacks or strokes. The answers can be found in the table 1 endpoints, expressed in absolute risk reduction, and Number needed to treat.

Understanding NNT scores is vital to evaluate the true performance of a drug. if anyone reading this has not done so, I strongly urge you to educate yourself on NNT scores. Simply google NUMER NEEDED TO TREAT.

While relative risk reductions were impressive (47%), 120 people had to take the drug for 1.9 years for one person to benefit. (NNT=120)

In the end, the study changes the NNT to 25 by making a forcast into the future which carries assumptions. This is pure manipulation of the numbers to make a poor outcome look better.

Look at it this way. How would you feel if you paid for super bowl tickets?
Just shy of the half time, the game is halted. The experts gather and decide the final score as if the game had been played to the end, then announce the winner, and what the final score would have been had the game played out to the end. The fans would be outraged. The stadium would go wild. There might even be riots. WHY? because the fans are tuned into the game, and know that it would be wrong.

Well, thats exactly what happened in the JUPITER trial. they stopped the trial early. The drug company just showed the relative risk reductions of 47%, and held their thumbs over the absolute risk reductions of <1% while boasting of a win. Then manipulated the NNT score to their favor.

BTW, the graph lines showing the death rates in the placebo group vs the crestor group were converging. If they crossed each other this would have been a disaster for crestor. I beleive this is the real reason the trial was stoped early.

[Roadstr]

I had a different doc in 2005 that ordered a test from Quest Diagnostics for homocysteine. This is marker that correlates to c-reactive protein... don't know much about, but I had to put in a letter of request to Quest to cover it because they had just classified it as experimental at the time... so, I remember it well!
Here is a study (http://circ.ahajournals.org/cgi/content/abstract/105/22/2595) on it that suggest it could be beneficial to determine risks.

[maxlharris]

Homocystine and C-reactive protein are both way more predictive than total cholesterol levels. Of course, LDL particle size and triglycerides are also way more predictive, but I think C-reactive is even better, if I recall correctly.