.One thing that many people don't realize is that many clinical trials for drugs do not test the ultimate outcome for a drug. The clinical trial tests a "surrogate biomarker," which means that the drug changes an easily measured, immediate statistic rather than an endpoint.
.
.
Blood Pressure, Cholesterol, and Your Heart
.
For example, most heart-health drugs on the market test whether or not the drug lowers blood pressure or cholesterol, not whether the drug reduces heart attacks or strokes. High blood pressure and cholesterol levels are associated with increased heart attacks and strokes, but no one knows whether (1) increased blood pressure or cholesterol causes heart attacks and strokes or whether (2) bad cardiovascular health causes increased blood pressure, elevated cholesterol, heart attacks, and strokes.
.
In the latter case, lowering blood pressure or cholesterol would not change the problem condition: poor cardiovascular heath, and would thus not reduce heart attacks and strokes.
It is possible that treating blood pressure and cholesterol problems does not reduce cardiovascular events, but these studies have not been performed for a variety of reasons.
.
First, they're very long-term, expensive, and possibly immoral. You would need to gather 20-30 years worth of data to determine whether there is a difference in heart attack rates between those folks who took blood pressure drugs and those who didn't. In the meantime, are you going to deny those drugs to people outside the study? It is possible that they will help.
.
Second, no drug company wants to know that their billion-dollar drug doesn't really work. Not to mention that no drug company wants to design their clinical trials around 30-year endpoints. Witness what just happened to Schering-Plough's stock.
.
Nature Article
.
There's a very good article in Nature this week about this subject: http://www.nature.com/news/2008/080401/full/452510a.html
.
If one is testing a drug for cancer or a viral disease, obvious surrogate markers like tumor size reduction or viral blood titer made excellent endpoints. It is very likely that these markers will correlate with increased survival rates or prolonged life, but not always. Even some cancer studies have found that reducing tumor size did not increase life expectancy.
.
Diseases with less clear cut causes, like Alzheimer's, or multiple causes, like heart disease, are more problematic. These very diseases also would require very long time horizons to judge ultimate effectiveness.
.
Alzheimer's Disease and Surrogate Endpoints
.
Alzheimer's Disease is a particular problem because we don't know, really, what causes Alzheimer's. (Really, we don't.)
.
The APOE gene has various alleles that increase risk, but even the worse one increases one's odds to around 40%. That's still less than a coin toss.
.
Many people with Alzheimer's dementia have gunky plaques made of amyloid protein on their brains, but not all, and many people with significant dementia and other AD markers (like tau tangles) have few plaques, and some people with lots of amyloid plaques have no dementia. It is not known whether amyloid plaques cause neurons to die, or whether a toxic precursor protein causes neurons to die and amyloid plaques are a garbage dump for this protein.
.
If the first, then reducing plaque buildup should reduce dementia.
.
If the latter, reducing amyloid deposition in plaques will increase dementia.
.
Wyeth Pharmaceuticals is using an unproven cognitive test to study its latest drug for AD, which is at least better than studying amyloid disposition.
.
And again, are you going to deny drugs that might work to a person with Alzheimer's or a high risk of developing it?
No comments:
Post a Comment