RSS feed of this blog

Gene tests can guide you to the right medicines


I was talking to a friend about medicine recently when she told me, “Oh, I always take kids’ doses of medicines.” It’s a novel idea because she’s an adult. But a small adult, she pointed out, adding, “I can’t see myself taking the same dose as my husband.”

She was expressing a growing rejection of the idea that “one size fits all” when it comes to medications.

Her thinking ties into the new trend of personalized medicine, which is now becoming ever more feasible with the ability to detect genetic variations that affect the way our bodies metabolize drugs.

This rapidly evolving field is the technology behind Vanderbilt University Medical Center’s PREDICT testing program. Vanderbilt is offering a simple blood test that checks for 200 genetic abnormalities that can affect your body’s metabolism of medicines. Vanderbilt is “the first academic medical center to do so,” says its website.

I wrote about this topic about a year and a half ago. But the rapid advance and significance of personalized medicine warrants an update.


This whole new science encompasses much more than just personalized dosing. The science is the result of the Human Genome Project, allowing us to map the whole human genetic blueprint—and most importantly, to identify “spelling mistakes” in that blueprint.

The mistakes are called single nucleotide polymorphisms (SNPS), and the ability to identify them has been one of those quantum leaps that make for huge advances in the field of medicine.

A great example of the benefit is the drug clopidogrel, brand name Plavix, that hit the market in 1995.

The drug is a kind of anticoagulant. Like aspirin, it keeps platelets from sticking together, thus preventing the first step in the formation of a full-blown blood clot. Blood clots can block off arteries, leading to strokes and heart attacks.

Now, people with severely blocked coronary arteries are sometimes treated by inserting a tube, called a stent, to hold the artery open.

But any foreign object in the body tends to cause blood to clot around it, and this can lead the stent to form a clot that blocks it off. If this happens, it is a disaster, as it cuts off the blood supply to part of the heart muscle—which is what a heart attack is all about.

So patients with stents were tried with clopidogrel in the CAPRIE study, sponsored by Sanofi pharmaceuticals, to see if this clotting-off could be prevented with the medicine.

The drug was not as effective as everyone had hoped. But nonetheless, I remember the Plavix drug reps coming by and pushing it enthusiastically. And by 2010, despite the reservations, it had become the second-largest prescription drug in dollar sales in the world, according to Dr. Eric Topol’s book “The Creative Destruction of Medicine.”

Still, people were troubled by the problem of these non-responders—troubled that there was no way to know which patients weren’t going to respond to the drug. It was a case of trial and disastrous error—until the ability to identify those spelling mistakes, or SNPS, came along.

With the analysis of the human genome, it was discovered that the non-responders had an aberration in the gene coding for the CYP2C19 enzyme, which is the enzyme that is essential for clopidogrel to work.

Some 30 percent of people were found to have this defect—which increased their risk of blocking off their stent with a blood clot by 300 percent.

So, the FDA slapped a new black-box warning on Plavix, telling doctors the drug may not work in people who carry this gene variant, and that there is a test to detect that fact.

The FDA has a list of twenty similar aberrations which can be tested for—information it is requiring to be put on package inserts.

Some of the better known drugs on this list are warfarin (in the blood thinner Coumadin), fluoxetine (in Prozac), metoprolol (in the heart medicine Lopressor and Toprol).

One other medicine on the list is simvastatin (brand name Zocor), which is very effective in reducing cholesterol, but in high doses causes severe muscle inflammation in some people.

Hearing advertisements that Vanderbilt can provide personalized testing to detect this reaction, as I bumbled about the bathroom listening to National Public Radio one morning, I realized pharmacogenomics really has hit prime time.

Detecting aberrations may sometimes mean you cannot take a medicine at all. But other times it warns that you have to personalize the dosage. So, for example, you can give people with the simvastatin metabolic defect only low doses of Zocor.

Another example: the blood thinner warfarin (brand name Coumadin) can make people bleed if their blood gets too thinned. With warfarin, dosing is done now by educated guesswork and doing a lot of blood tests to check the “protime.” But still, bleeding occurs in 2 to 10 percent of patients in the first year of treatment. And 1 percent of these patients die—primarily from strokes.

As is the case with clopidogrel, some people have been found to have an abnormal metabolism when it comes to warfarin. In this case, there are two possible enzyme defects, and a complicated dosing chart has been compiled depending if they have one or both.

Though pharmacogenomics is expanding rapidly, such tests are not universally available yet. Soon, though, it seems likely your doctor will be able to order a pharmacogenomics panel just like any other blood test.

There will, of course, be that issue—as with so many other tests—of deciding when there is sufficient risk to justify doing what will surely be a pretty expensive test. (At present, Vanderbilt offers it for free to those they accept into their experimental program, according to the PREDICT website.)

But with 2.2 million serious adverse drug events and 100,000 deaths a year, there is no question there is a need to personalize our taking of drugs.

Dr. Patrick Neustatter, a longtime family practitioner, is the medical director of the Lloyd F. Moss Free Clinic. He can be reached at healthyliving@freelance