Can brain scans show a difference between drug therapy and psychotherapy? A researcher at University of California at Los Angeles uses positron emission tomography to observe the difference in brain changes between these two types of treatment for major depression.
In a groundbreaking effort to attempt quantitative comparison of therapies for depression, Arthur L. Brody, M.D., assistant professor of psychiatry at University of California, Los Angeles' (UCLA) Neuropsychiatric Institute, is using positron emission tomography (PET) scans to observe brain changes in patients receiving talk therapy, as compared to patients receiving psychopharmacological therapy.
Funded by an award from the National Association for Research on Schizophrenia and Depression (NARSAD), Brody, who is also director of UCLA's Interpersonal Psychotherapy Clinic, recently presented his group's work at NARSAD's 11th annual scientific symposium held in New York City. The session was moderated by William E. Bunney Jr., M.D., distinguished professor and Della Martin Chair of Psychiatry at University of California, Irvine, and a member of NARSAD's scientific council.
Brody told the assembly that in functional imaging studies, the brain regions most commonly found to have abnormalities in major depression are the prefrontal cortex, anterior cingulate gyrus, temporal cortex and caudate nucleus (Baxter et al., 1989; Drevets et al., 1997; Mayberg, 1997). Therefore, his group sought to determine metabolic changes in these regions in subjects with major depression from pre- to posttreatment with either paroxetine (Paxil) or interpersonal psychotherapy (IPT). Before discussing that study, however, Brody turned to a published preliminary study in which his group used PET scans to examine changes in 16 depressed subjects treated with paroxetine (Brody et al., 1999).
"What we did there is compare responders to treatment to nonresponders," he said. "And what we found is that responders to treatment had decreases in activity in the ventrolateral prefrontal cortex and the orbital frontal cortex, the part of the cortex right above the eyes. That's a region that has strong input from the serotonin system, so that might explain why that region had a decrease in activity when subjects were treated with a serotonin reuptake inhibitor. Nonresponders did not have these decreases in activity."
Moving on to the current study, Brody described IPT (Klerman et al., 1984; Weissman and Markowitz, 1994), explaining that the therapy's goal is improvement of depressive symptoms through strengthening the patient's interpersonal relationships.
"The focus is on current stressors in a person's life and practical ways to improve relationships to alleviate those stressors and strengthen existing relationships," he said. "It's a very direct therapy where we use behavior change techniques and actually suggest alternatives for patients to do in their current relationships. The therapist in this therapy is a patient advocate, not neutral...and active, not passive. It's a very interactive type of psychotherapy."
The 24 study subjects with major depression were allowed to choose either IPT or medication therapy. This resulted in 14 subjects in the psychotherapy group and 10 subjects in the paroxetine group. Also included were 16 normal controls who received no psychotherapy or medication. All subjects were scanned two times, about 12 weeks apart. Pre- and postmeasures of affect were evaluated by the Hamilton Rating Scale for Depression (HAM-D).
"This was a slightly sicker group overall than in our previous study of depression, in that the people in the IPT group had an average of 3.5 previous treatments for depression, and the paroxetine group had 2.1 previous episodes," Brody noted. "The IPT group also started out with a slightly higher [HAM-D score] than the paroxetine group."
Results for the paroxetine-treated subjects showed a 62% drop in their affective ratings scales scores, whereas for the psychotherapy subjects, there was only about a 40% drop in scores. Brody said that, similar to the previous study, PET scans of the paroxetine patients showed a decrease in activity in the more ventral portion of the frontal lobe.
"Finding this decrease in both of our studies of patients treated with Paxil makes this a much stronger finding," he said. "These were two different groups of subjects and done on two different scanners with slightly different methods, and we found the same thing. Other groups have had similar findings as well [Mayberg et al., 1999]."
In the IPT group, PET scans showed a highly significant increase in activity in the inferior frontal gyrus and anterior insula for all subjects. Normal controls had no significant change on their PET scans other than normal changes seen with repeated scans.
"This is an area that's been associated with language," Brody said of the inferior frontal gyrus and anterior insula region. "It's also been associated with movement and emotion. So one of our hypotheses is that maybe after the talk therapy they received, they were thinking about talking more, or more talkative in general."
Asked about long-term implications of medication versus psychotherapy, Brody said that his group has not done those kinds of analyses, but several other large studies have.
"What actually tends to happen is that people who get treated with IPT...tend to maintain better social functioning later on," he explained. "But in terms of who does better in the long run, medication and psychotherapy both have strengths and weaknesses. Neither therapy has definitively been proven superior to the other, but some studies have shown superior efficacy for medication as perhaps being a little more effective than psychotherapy (in more severely depressed individuals). But the people who get the psychotherapy tend to have better social functioning one, two or three years out of the psychotherapy."
A good next step, added Brody, would be to examine the combination of psychotherapy and medication.
"Medication tends to work a little bit quicker than the psychotherapy, so it would be interesting to see what happens in people's brains as you go along, if treated with the combination."
Regarding the fact that both interventions yielded mood changes but each showed different effects on the brain, Brody said he plans to explore that further.
"Our first hypothesis is that it has to do with different symptoms changing," he said. "The two treatments are actually different in how effective they are for specific symptoms of depression, so, for example, neurovegetative symptoms like sleep trouble and appetite trouble tend to do better with medication, whereas, as I mentioned before, social interactions and anhedonia tend to improve more with psychotherapy. So the first thing we're going to look at is the symptoms that the patients had, and which symptoms changed and didn't change with treatment."
Another issue Brody hopes to address in future work is controlling for possible pretreatment differences in groups.
"In a future study, we would randomize which group the people go in to," he said. "In this preliminary study, we weren't able to do that. We just let people choose whatever treatment they wanted."
Another interesting strategy would be to give controls treatment, but there are ethical questions surrounding the dispensation of active medication to normal subjects, Brody noted.
Asked about the possibility that more weeks of IPT might have shown stronger results, Brody explained that the 12-week time frame was selected because of the Paxil group.
"It would be unfair to subjects with major depression to treat them with Paxil for more than 12 weeks if they weren't responding," he said. "So in order to get a comparable time frame between the two groups, we chose 12 weeks of psychotherapy. But you're right, we used a high dose of Paxil and a relatively low dose of psychotherapy, so that might explain some of the differences."
Commenting on Brody's work, Bunney said there are two issues.
"The first issue is looking at PET changes associated with drugs, and the other is looking at PET changes associated with psychotherapy," he said. "There may be 300 studies looking at changes with drugs, but looking at changes with psychotherapy is absolutely pioneering," Bunney added, "this is...very nice [work]."
In an interview with Psychiatric Times, Brody said his group had uncovered more findings.
"In our statistical parametric mapping analysis, we found some similarities in that both groups had decreases in the ventral lateral prefrontal cortex," he explained. "The finding wasn't as significant in the IPT group as it was in the Paxil group, but that was a big similarity we found."
Another similarity the researchers found was that both groups had decreases in activity in the right caudate. Brody said that both the similarities and the differences found between the groups may be related to specific symptoms each therapy addresses.
"Both groups had their depression get better overall," he said. "The Paxil group had a bigger improvement than the IPT group, but they both got better. My hypothesis would be that the areas of similarity may reflect common symptoms being treated by both therapies, while the areas of difference may reflect different symptoms addressed by each therapy. We haven't proven that hypothesis, but we're looking into proving it now."
What's interesting, Brody added, is that, despite their proven effectiveness, neither IPT nor cognitive-behavioral therapy is widely used.
"There just aren't that many people trained in these kinds of psychotherapies," he said, "and we would certainly like to see more who are."
Baxter LR Jr, Schwartz JM, Phelps ME et al. (1989), Reduction of prefrontal cortex glucose metabolism common to three types of depression. Arch Gen Psychiatry 46(3):243-250.
Brody AL, Saxena S, Silverman DH et al. (1999), Brain metabolic changes in major depressive disorder from pre- to post-treatment with paroxetine. Psychiatry Res 91(3):127-139.
Drevets WC, Price JL, Simpson JR Jr et al. (1997), Subgenual prefrontal cortex abnormalities in mood disorders. Nature 386(6627):824-827 [see comments].
Klerman GL, Weissman MM, Rounsaville BJ, Chevron ES (1984), Interpersonal Psychotherapy of Depression. New York: Basic Books.
Mayberg HS (1997), Limbic-cortical dysregulation: a proposed model of depression. J Neuropsychiatry Clin Neurosci 9(3):471-481.
Mayberg HS, Liotti M, Brannan SK et al. (1999), Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry 156(5):675-682.
Weissman MM, Markowitz JC (1994), Interpersonal Psychotherapy. Current status. Arch Gen Psychiatry 51(8):599-606 [see comments].