Talk Therapy or Antidepressant? A Brain Scan Predicts Which Works Best for Your Depression
2013-06-24
There hasn’t been much in the way of hard science to help doctors or patients decide on the best treatments for depression — until now. For the first time, brain imaging may be able to help determine who will get better in therapy and who improves more on medication.
Depression affects an estimated 1 in 5 people over a lifetime, and talk therapies and antidepressant medications can help a significant proportion of those patients. But figuring out who will benefit most from which treatments remains a major challenge; while nearly 22 million Americans take antidepressants, 40% of people are not helped by the first treatment — drug or talk therapy — they try. And since it often takes weeks to relieve symptoms, choosing the wrong first treatment can lead to extra months of suffering.
A new study involving brain scans, however, provides hope that such hit-or-miss strategies and lost treatment time may soon be a thing of the past. Published in JAMA Psychiatry, the study included 65 people with depression who had been randomly assigned to receive either 12 weeks of treatment with the antidepressant escitalopram (Lexapro), or cognitive behavioral therapy (CBT), the most studied talk therapy for depression. CBT helps patients to reframe negative thoughts and perceptions and change their behavior in order to address their mood.
Before any of the patients received their respective treatments, researchers took PET scans, which capture the function and activity of certain cells, of their brains. The scientists then compared the brain activity among the people who responded strongly enough to resolve their depression to that of patients who did not respond at all to whatever treatment they received (they discarded the results of those who had only a partial response).
The people who got well differed from the nonresponders in the activity of the insula, a region that assesses signals related to pain, heart rate, temperature, blood sugar and other internal states. Changes in these levels could suggest a need for action, such as seeking food or shelter, or the need to defend against a threat. The insula also tends to be especially active during emotions like disgust, fear and sadness, but it also plays a role in pleasure.
Insula activity predicted who responded best to either of the treatments. Those who found relief from talk therapy tended to have reduced activity in this area before treatment, compared with activity in other parts of their brain. “Low activity in the insula at baseline may reflect impaired sensitivity to signals [of] one’s internal state,” says Dr. Helen Mayberg, professor of psychiatry, neurology and radiology at Emory University and the lead author of the study, suggesting that this may make people unduly focused on negative experiences like rejection.
CBT, however, teaches patients that their initial perceptions are just thoughts and that a more positive interpretation of their first, and gloomy perceptions, is possible. So those with lower insula activity may benefit from this active reframing of their environment, since it helps to divert their attention away from the negative thoughts and emotions that have become their primary focus.
The people who responded to Lexapro, however, had the opposite finding: their insula activity was elevated before treatment, possibly making them prone to rumination or more focused on their internal experiences of anxiety, sadness and disgust. Therefore, medications could potentially alleviate their depressive symptoms by turning down the volume of this predominantly negative information flow. The result? Greater connection with others, which in turn can reduce anxiety and sadness. Indeed, previous studies show both that medications like Lexapro and Prozac can mitigate emotional oversensitivity and reduce insula activity.
“This provides proof of principle of a potential brain-based biomarker and future strategy for precision medicine for a psychiatric disorder, a strategy well established in the treatment of cancer and infectious disease,” says Mayberg.
Aimee Hunter, assistant director of the Laboratory of Brain, Behavior and Pharmacology at the University of California, Los Angeles, who was not associated with the research, also sees the potential that the scans have for making treatments more precise. “This is an exciting first-stage effort to identify a brain marker with the potential to help guide initial treatment selection for people with depression,” she says.
The results imply, however, that the combination of both treatments — the current gold standard for care — may not be best for everyone because they might work at cross-purposes. “Given that the two treatments have different effects on the brain and in some cases [may change the insula] in opposite directions, [that] is a reasonable hypothesis,” she says.
But the data still doesn’t rule out the possibility that some patients may benefit from both. In some cases, medication or other treatments that stimulate the brain may make it more receptive to therapy, when it initially could not be reached that way. If that’s the case, it’s possible that using medications and talk treatments sequentially might be more effective.
Hunter also notes that Mayberg and her colleagues made some important assumptions about how patients respond to therapy that might require further research to understand. They assumed, for instance, that there were differences in the type of person who would respond to talk or medication. “The strategy assumes that an individual who would respond well to medication would not respond well to CBT, and vice versa. This simply is not known,” says Hunter. “When a person responds well to an initial course of treatment, it is difficult to know how they would have responded to a different treatment.”
Mayberg is already planning a larger and more detailed study to validate PET scans as a tool in choosing depression treatment. “The next step for my team is to treat by brain type and see if using the insula biomarker does better than the [typical] 40% to 50% remission rate,” she says. “We have a larger study using other imaging tests and additional measures as well as more than one medication to address some of these issues.”
For people with depression, those results can’t come soon enough. Experts and policymakers have long debated whether talk or drugs are best, but this study suggests that, as with virtually all other diseases, different treatments work for different people. “Despite the limitations of this study and the complexities inherent in treatment response, this investigation moves the ball forward toward biologically based treatment selection, and away from reliance upon sheer trial and error,” says Hunter. And that can only benefit patients.