Promising Findings on Using Neuroimaging to Predict Treatment Response in Unipolar Depression

Editor, Primary Psychiatry and Psychiatry Weekly, Professor of Psychiatry, New York University School of Medicine

Introduction

With the advent of neuroimaging technology, modern researchers are able to do what earlier generations of psychiatrists could not do: look into functional changes in the human brain. The ultimate objectives are to better understand the pathophysiology of mental disorders and thus devise more effective treatments. Ideally, clinicians could one day use imaging to identify a subtype of disorder and match it to a specific treatment, and then monitor changes in brain activity as a correlate of symptom improvement. Studies have already demonstrated changes in glucose metabolism associated with response to antidepressant treatment modalities, such as cognitive behavior therapy (CBT), antidepressants, and deep brain stimulation. A group of Canadian researchers now report specific changes in positron emission tomography (PET) images among subjects who responded to CBT or venlafaxine.

PET and Treatment Response

They conducted a randomized, controlled trial of subjects in a major depressive episode with existing diagnosis of a major depressive disorder. These subjects were scanned before randomization and after 16 weeks of antidepressant treatment with either CBT (N=12) or venlafaxine (N=12). They found that response rates were comparable between the CBT and venlafaxine groups. According to the authors, “response to either treatment modality was associated with decreased glucose metabolism bilaterally in the orbitofrontal cortex and left medial prefrontal cortex, along with increased metabolism in the right occipital-temporal cortex. Changes in metabolism in the anterior and posterior parts of the subgenual cingulate cortex and the caudate differentiated CBT and venlafaxine responders.”

Responders to either treatment modality demonstrated reduced metabolism in several prefrontal regions. CBT response, for example, was associated with a reciprocal modulation of cortical-limbic connectivity, while venlafaxine engaged additional cortical and striatal regions. Imaging also showed that nonresponders to both types of intervention demonstrated left lateral orbitofrontal and left dorsolateral prefrontal decreases, along with decreases in the rostral anterior cingulate and globus pallidus. Differential changes in metabolism were also observed between the CBT and venlafaxine nonresponder groups. Whereas nonresponse with venlafaxine treatment was associated with increases in the posterior thalamus and the dorsal insula, CBT nonresponders displayed decreased metabolism in these same brain regions.

Conclusion

This study may help to explain why some patients respond better to psychotherapy and others to pharmacotherapy, and why many patients do better when these modalities are combined. This study is also important because it demonstrates differential changes in regional brain activity following response, or nonresponse, to each of the therapeutic interventions. Before concluding that we will soon be able to predetermine which patients should respond to medication or CBT, it is important to note that the investigators were not able to establish any predictive value of early change on subsequent clinical outcome. As with all small studies, this needs to be replicated.

Poster presented at the Society of Biological Psychiatry, Poster #736, Toronto, May 18-20, 2006; American Psychiatry Association, Poster #937, Toronto, May 20-25, 2006.