Skip to content

Highlight: Toward Signposts for Precision Medicine

PET data superimposed on anatomic MRI scan data showing brain activity in front part of insula.

Low pretreatment resting brain activity in the front part of the insula (right side of the brain image—the red area where green lines converge) signaled a significantly higher likelihood of remission with CBT and a poor response to escitalopram. Conversely, hyperactivity in the insula predicted remission with escitalopram and a poor response to CBT. Image shows PET data superimposed on anatomic MRI scan data.33

Source: Helen Mayberg, M.D., Emory University

Treatment selection in areas of medicine outside of mental health, such as cancer and heart disease, is increasingly based on an understanding of the multiple possible causes of these diseases in different individuals and the ability to use biomarkers32 (e.g., indicators from blood and genetic tests) to guide and precisely tailor treatment. However, treatment of a condition like depression remains based largely on trial and error. A health care provider will try a treatment—a medication or psychotherapy—for a month or two to see if it works. As a result, fewer than 40 percent of patients achieve remission with their first treatment.33 The time lost exacerbates an already lengthy delay before relief from the symptoms of depression can begin. And the financial impact of care is immediate, even though relief is not.

Because depression can emerge from many different underlying causes, it is unlikely that there will ever be a single treatment that works for everyone. Rather, what we can aim for is the ability to predict which particular treatment works for a particular individual. Early studies using positron emission tomography (PET) scans have provided information on the areas of the brain affected by depression and the effects of treatment. Recent studies showed that pretreatment scans of patients diagnosed with depression could predict which patients would respond to treatment with cognitive behavioral therapy (CBT) versus a standard antidepressant medication, escitalopram. Activity in one specific brain area—the insula—proved to be a reliable predictor of treatment outcome. Low resting brain activity in the front part of the insula indicated a higher likelihood of remission with CBT and a poor response to the antidepressant escitalopram. Conversely, high activity in the insula predicted remission with escitalopram and a poor response to CBT. This area of the insula is known to be important in regulating emotional states, self-awareness, decision making, and other cognitive tasks.

While PET scans are expensive and not likely to be used broadly, this research constitutes a proof of concept: the identification of biomarkers, such as distinctive patterns in a PET scan, can provide an evidence base for choosing the optimal treatment for each individual. The challenge is to find biomarkers (including cognitive performance) that are simple, inexpensive, and reliable predictors of treatment response.

Highlight from Strategic Objective 3