Identifying Neurological Biomarkers to Aid in Major Depressive Disorder Diagnosis and Treatment

Recently, a leading pharmaceutical company used our BNA™ platform, powered by our age-matched, normative, comparative database to assess the effects of an SNRI antidepressant on depression and cognition. How will you use BNA™ with your patients to assess, track, and improve their depression?

Background

Identifying neurological biomarkers to aid with diagnoses, treatment effect evaluation of mood and cognitive symptoms, and treatment outcome prediction in major depressive disorder (MDD) is an unmet need. In a recent study, the clinical utility of BNA™ to assess the effects of SNRI antidepressant on depression and cognition, and as an MDD management support tool was evaluated.

Methods

Twenty-five MDD patients (20-65 years old) with cognitive dysfunction complaints, underwent an 8-week open-label treatment using an SNRI medication. Event-related potential (ERP) data, generated during cognitive tasks, were collected at pre-treatment, two weeks, and eight weeks into treatment. Correlations with MDD symptoms and cognitive functioning were investigated. Exploratory analyses assessed the differences in baseline and endpoint ERP scores between MDD patients and matched controls (N=29).

Results

Baseline ERP scores were correlated with the Hamilton Depression Rating Scale (HDRS-17) total scores;
the perceived deficits questionnaire (PDQ) scores; and the digit symbol substitution test (DSST), associating higher ERP scores with higher depression severity and greater cognitive dysfunction.
Baseline-to-endpoint changes in ERP scores were correlated with the respective changes in PDQ scores, associating a larger decrease in ERP scores with a larger improvement in subjective cognitive function.

**Figure 1.** Significant correlations of depression severity and cognitive performance with BNA scores. BNA scores significantly correlated with depression severity assessments and cognitive performance at baseline (upper panel). Baseline-to-endpoint changes in BNA scores correlated with baseline-to-endpoint changes in cognitive assessments (lower panel). Pearson’s r, p-value, and sample size are presented for each correlation. Shaded areas indicate 95% CI.

Treatment Outcome Prediction

Finding neuro markers to predict treatment response provides one of the greatest challenges in depression management. To date, decisions regarding specific treatment protocols for depression are based on clinical experience and risk factors with limited data on outcome prediction. As an adequate treatment trial takes 8 weeks, the long and often unsuccessful search for an effective antidepressant is accompanied by a significant decrease in patients’ quality of life, an increased risk of suicidal action, and decreased chance of response and remission with each attempt. Thus, there is a great clinical need for reliable markers of MDD treatment outcomes.
Early prediction of response may also be useful for predictive subject enrichment in clinical trials – patients who are more likely to respond to treatment than other patients can be selected by using EEG biomarkers that predict treatment response in the same or similar drug class. This strategy can enhance benefit–risk relationships, lead to a larger effect size, and permit smaller study populations.

**Figure 2.** Treatment responders show more anterior ERP components at baseline, while non-responders show more posterior components. The bottom figures show an example of the topographic location of a responding patient (left) and a non-responding patient (right). Patients’ peak activations are in blue, age-matched normative group’s peak activation is in black. These topographic location figures are available in the BNA™ ERP reports.

Conclusion

BNA provides an objective measure of depression, tracking treatment-related cognitive changes.
A BNA-based biomarker provides an early prediction of response to SNRI treatment.
These results provide preliminary evidence for the potential use of the BNA technology in the diagnosis, treatment optimization, and management of MDD.