Brain activity may help forecast most effective antidepressant medication


by Kayt Sukel

October 23, 2009

When Nathan Bailie, 36, sought help for his depression four years ago, he didn’t think that finding the correct medication to regulate his condition would be such a long and complicated road.

“I told the doctor that I didn’t know what it was like to be happy and we needed to figure this thing out before it got worse,” he said. He was prescribed a selective serotonin reuptake inhibitor (SSRI), but two months later, struggling with uncomfortable side effects, he still suffered his depressive symptoms. Today, Bailie is successfully treating his condition with medication but, all told, it required one year, two doctors and four different drugs to find the right one.

“It was very frustrating,” he said. “Especially when you factor in some of those drug side effects.”

Bailie is not alone. More than half the people diagnosed with major depressive disorders will not find relief with the first antidepressant they try, and for many finding the right one is a lengthy and painful process. But researchers at the University of California Los Angeles (UCLA) may have found a clinical biomarker—a pattern of electronic brain signals—that could help psychiatrists prescribe the right antidepressant medication the first time around.

Testing patients, patience

“Finding the right drug for a patient is currently more art than science, unfortunately,” says Vatsal Thakkar, a practicing psychiatrist and professor at the New York University School of Medicine. He says that family history, depression phenotype, side effects and even insurance plans play a role in how he decides to prescribe one or another antidepressant drug. But even with all that information, he estimates that most of his patients end up trying more than one medication. And, regrettably, that means patients will spend more time bearing the debilitating symptoms of depression.

“There are more than 20 Food and Drug Administration approved antidepressants on the market. A doctor has a lot of choice,” says Andrew Leuchter, a researcher at UCLA’s Semel Institute for Neuroscience and Human Behavior. “But what doctors don’t have are strong clinical indicators to suggest that a particular patient should get a particular medication in terms of effectiveness.” And with so many patients not receiving the right drug on the first pass, the results can be incredibly discouraging.

 “We know that more than 85 percent of patients who stick with treatment will get well. The problem with these drugs is that the side effects are immediate but therapeutic effects take time,” says Leuchter. “The current process is really an empiric medication trial and hard on patients. But right now it’s the best that we’ve got.”

Using biomarkers to point doctors in the right direction

Leuchter believes that psychiatry could benefit from the development of clinical biomarkers, physical readings with prognostic significance, to help guide doctors when making drug decisions. In a study published in the September issue of Psychiatry Research, Leuchter and colleagues argue that one such biomarker, brain-wave patterns as measured by quantitative electroencephalography (QEEG), has high predictive value in choosing between two different antidepressant options.

QEEG is a method of recording the brain’s electrical activity (caused by the firing of neurons) through electrodes placed on the scalp. It is used clinically to diagnose disorders like epilepsy, coma and brain death and also is used in research. Leuchter argues that its sensitivity to changes in brain wave patterns makes it a prime candidate to help in discovering biomarkers.

“It is extremely sensitive to drug effects,” says Monte Buchsbaum, director of the NeuroPET Center at University of California San Diego and editor of Psychiatry Research, and a member of the Dana Alliance for Brain Initiatives. “Even a small effect in brain activity can be easily detected.”

Leuchter and colleagues used QEEG to evaluate the brain wave patterns of people  who had been diagnosed with major depression disorder both before they took any antidepressant medication and one week after they started taking the drug escitalopram (also called Lexapro), one of the most commonly prescribed antidepressant medications in the United States. Study participants were then randomly selected to continue with the escitalopram or instead switched to buproprion (Wellbutrin XL). The researchers found that a previously defined biomarker they named the antidepressant treatment response (ATR) index could predict response to the drugs at 74 percent accuracy.

“The ATR is a composite of several EEG variables scaled into a single index measure that goes from 0 to 100,” says Leuchter. “We found that when an individual gets a score close to 0, it indicates a low likelihood of response to Lexapro. If the patient gets a number closer to 100, there’s a high likelihood of response.”

But most importantly, Leuchter argues, the group was able to show that people with a low ATR score were much more likely to respond if switched to the Wellbutrin. “This is the first time that anyone has shown differential response to two distinct antidepressant medications—not just with an EEG but with any test,” he says.

Buchsbaum agrees. “The two drugs have different functions. Lexapro is an SSRI while Wellbutrin increases dopamine and norepinephrine output. And they may treat two different subtypes of depression,” says Buchsbaum. “This EEG biomarker did not merely show whether a person would respond to a drug but also offers a tool to help physicians choose between two different drugs. It makes this biomarker very clinically useful.”

From the bench to the bedside

Leuchter and colleagues plan to use the ATR index to look at people’s response to other drugs; preliminary data suggests that a similar signal can be seen with most antidepressant medications. Leuchter also believes that the EEG might provide clinically relevant information about other facets of antidepressant drugs—perhaps even side effects.

“It’s something that we’re actively looking at, and there’s some data that suggests we can say something meaningful about side effects,” he says. “The flip side of efficacy is safety, and we want to address the safety concerns of these drugs as well.”

In terms of using QEEG to help select the right antidepressant medication for patients with depression, Bailie says he wishes that such a system was available when he first sought treatment. “It took a long time to get me on the right meds,” he says.

Thakkar agrees that the idea of using QEEG as a clinical biomarker is promising. “Anything that helps us make quicker and more accurate decisions that will get people better I welcome,” he says. “It’ll be interesting to see where this goes and whether it passes the bar to become a useful clinical tool.”