New biomarkers take the guesswork out of finding the right antidepressant

Finding the right treatment for depression involves a lot of trial and error, but research is underway to change that.

Too often, finding the right antidepressant can take a while—only a third of patients see results from the first one they try. A research collaboration spanning five countries has been working to help more people find the best treatment right away. The “International Study to Optimize Treatment for Depression” (iSPOT-D) identifies biomarkers, physical biological indicators, that can tell doctors which antidepressants will work best for which patients, and assist in the development of new ones. We spoke with Stanford University professor Leanne Williams about the project and some early results.

You can keep up with iSPOT-D and its progress by following the project on ResearchGate.

ResearchGate: For patients trying to find the right treatment for depression, what is the experience like currently?

Leanne Williams: Currently, finding the right treatment for depression involves haphazard improvisation. Patients first face the challenge of recognizing that they may have a clinical depression, which can be hard to do due to lack of clear information and the associated stigma. The next challenge is to find the right doctor or therapist. From there, it can be a prolonged process of trying one treatment after another until something works. Only one­ third of patients with clinical depression get better from the first antidepressant they try. The remaining two­ thirds try other drugs—often many and often for years. Clinical depression is now our number-one cause of disability and lost productivity. It takes away our capacity to hope. This disability and lost hope comes in large part from this prolonged trial and error process.

RG: How do doctors typically decide which antidepressant to try first?

Williams: The challenge for the doctor is that they have to choose a particular treatment based on past experience and clinical guidelines, but without access or information from biomedical tests. We could not imagine this situation for other medical conditions. Someone experiencing signs of a heart condition would expect their doctor to use his or her experience and expert knowledge, but to also run tests with the latest technology.

For clinical depression, we haven’t gotten scientifically-based tests connected with doctors at the front line of health care. As a result, while our patients wait, their brains get more and more stuck in depression mode. To me, this is unacceptable. My research is focused on developing biomarker tests that help get the treatment right for the right person at the right time, and connecting these tests into clinical care.

 

“We could not imagine this situation for other medical conditions.”


 

RG: How can identifying biomarkers help?

Williams: Biomarkers help by giving us information about the organ of most relevance to clinical depression: the brain. For clinical depression, the same symptoms can result from different underlying changes in how the brain is functioning. Therefore, knowing what is going on for each person can help determine which treatment is best suited to their style of brain functioning.  As a concrete example, for some people, the anxiety systems of the brain may be overactive and, for others, underactive. Our research has shown that knowing this can improve the accuracy of predicting who will benefit from different types of currently available antidepressants, and improve the accuracy of prediction to as high as 80 percent.

RG: Which medications and biomarkers are you looking at?

Williams: In iSPOT-D, I have been looking at sertraline, escitalopram or venlafaxine. Common brand names for these medications are Zoloft, Lexapro and Effexor, respectively. Because iSPOT-D was the first biomarker study of its kind, we gathered a rich set of information on each participant in the study. This ensured we could identify biomarkers of brain function and genetics, and their interaction with life experience. We gathered information about different aspects of brain function, using brain scans and EEG. We also assessed genetic variants and took account of important life history information, such as experiences of childhood trauma.

RG: What have some of your most significant findings been so far?

Williams: One of our early successes has been matching brain circuit dysfunction to tailored treatment outcomes. We found that brain imaging of the amygdala is a biomarker predictor of response to different types of antidepressants. The amygdala has been implicated in the mood and anxiety related features of depression. With a predictive accuracy of over 80 percent, amygdala hypo-reactivity predicted the general capacity to respond to antidepressants. However, hyper-reactivity to mood-congruent emotions typically signaled that someone would not be responsive to a specific serotonin-norepinephrine reuptake inhibitor (SSRI). These findings were the first proof of concept for a brain imaging treatment predictor in a biomarker clinical trial. Another success was the discovery of a genotype for optimizing antidepressant outcomes and minimizing side effects.

Our research also established the important role of early life trauma in predicting who recovers on antidepressants, and I found a staggeringly high prevalence of early life trauma across this large international sample of outpatients. Trauma was four times higher in this group than in the general population. I went on to show that specific types of trauma, particularly abuse, occurring prior to seven years of age have an enormous impact on the capacity to mount a response to antidepressants. Those experiencing early abuse account for only 19 percent of patients who saw improvement, while those without trauma account for 81 percent. This compelling finding lead me to take account of early life trauma when developing the predictive model reported in PNAS.

 

“Biomarkers help by giving us information about the organ of most relevance to clinical depression: the brain.”


 

RG: In addition to finding the most effective among current treatments, could biomarkers help with the development of new antidepressants?

Williams: Absolutely, yes. Without understanding biomarkers, we are “shooting in the dark” for developing new treatments. Biomarkers help us understand the specific processes that underlie each person’s experience of depression. It is not “one size fits all.”  We have seen that some people have a distinctive disruption to the reward circuits of the brain. These people tend to be most frustrated by the loss of the ability to feel pleasure and feel motivated. Our current treatments do not touch reward circuits. So, it is not surprising that people experiencing this form of depression are over-represented in patients that can’t find an effective treatment out of the current options. Biomarkers will be important in guiding the development of new treatments that target this type of depression and also in identifying the people who may benefit most from them.

RG: What’s next for this research?

Williams: I am now extending iSPOT-D in multiple ways. For example, to anxiety and multiple mood disorders that commonly occur together or have overlapping symptoms and treatments. I am also extending the focus to other forms of intervention, including new non-invasive brain stimulation technologies and evidence-based online apps.

The translational goal of our research is to develop a personalized profile that is a precise read-out of each person’s brain profile and functional context. In addition to that, I have a special translational project underway. It is a proof of concept for translating these biomarkers into real world clinical practice. I call it the development of a precision mental health model for biomarker-guided treatment.

In our research we recognize too that, even if the biomarkers are good predictors, they are by no means deterministic — a person’s lifestyle will impact a biomarker’s forecast, just as the effectiveness of heart drugs is reduced if a person has a poor diet or doesn’t exercise. Still, I believe we need to urgently pursue prospective studies that continue to confirm and refine the biomarkers identified in early successes.

Featured image courtesy of Hayley Mechelle.