ABSTRACT
BACKGROUND: A better understanding of factors underlying antidepressant non-response may improve the prediction of which patients will respond to what treatment. Major depressive disorder (MDD) is associated with alterations in fatty acid metabolism, (neuro)inflammation and amygdala-reactivity. However, their mutual relations, and the extent to which they are associated with prospective antidepressant-response, remain unknown. PURPOSE: To test (I) alterations in (neuro)inflammation and its associations with fatty acid metabolism and amygdala-reactivity in MDD-patients compared to controls, and (II) whether these alterations are associated with prospective paroxetine response. METHODS: We compared 70 unmedicated MDD-patients with 51 matched healthy controls at baseline, regarding erythrocyte membrane omega-6 arachidonic acid (AA), inflammation [serum (high-sensitivity) C-reactive protein (CRP)], and in a subgroup amygdala-reactivity to emotional faces using functional magnetic resonance imaging (fMRI) (N=42). Subsequently, we treated patients with 12 weeks paroxetine, and repeated baseline measures after 6 and 12 weeks to compare non-responders, early-responders (response at 6 weeks), and late-responders (response at 12 weeks). RESULTS: Compared to controls, MDD-patients showed higher CRP (p=0.016) and AA (p=0.019) after adjustment for confounders at baseline. AA and CRP were mutually correlated (p=0.043). In addition, patients showed a more negative relation between AA and left amygdala-reactivity (p=0.014). Moreover, AA and CRP were associated with antidepressant-response: early responders showed lower AA (p=0.018) and higher CRP-concentrations (p=0.008) than non-responders throughout the study. CONCLUSION: Higher observed CRP and AA, their mutual association, and relation with amygdala-reactivity, are corroborative with a role for (neuro)inflammation in MDD. In addition, observed associations of these factors with prospective antidepressant-response suggest a potential role as biomarkers. Future studies in independent samples are needed to replicate and test the clinical applicability of these biological predictors for treatment response to result in a precision/personalized medicine approach for treatment.
