GENOMIC AND IMMUNOLOGICAL BIOMARKERS FOR PREDICTING TREATMENT OUTCOME IN DEPRESSION

Abstract The aim of this project is the validation of genomic and non-genomic biomarkers for predicting antidepressant treatment outcome to identify biologically more homogenous subgroups of patients that might benefit from specific treatments targeting the identified biomarker system. Based on previous findings, we focused on the stress regulation gene FKBP5 for the selection of the genomic biomarkers and on a set of 41 cytokines/chemokines for defining immunological biomarkers. 307 inpatients hospitalized for treating a major depressive episode were included, who participated in the MARS/OptiMD study. MARS/OptiMD is an open-label multicentric treatment study conducted at four clinical centers in Germany. Biomarkers were extracted from DNA, RNA, and serum samples collected after study inclusion. Antidepressant treatment outcome was evaluated weekly by trained personal with the 21-items Hamilton Depression Rating scale (HAMD-21). Primary endpoint in this analysis was response after six weeks of treatment defined as a 50% HAMD-21 reduction compared with baseline. Replicating our previous findings with suggestive evidence, we observed that reduced FKBP5 expression was associated with response with approaching significance, and this trend was moderated by the FKBP5 genotype (rs1360780). Using Illumina EPIC Methylation BeadChips, we identified 67 DNA methylation markers located within the boundaries of the FKBP5 gene. A principal component analysis revealed five vectors with two (vector 1 and vector 2) showing a robust positive association between higher methylation levels and antidepressant treatment response. Again, these effects were moderated by the rs1360780 genotype. Among 41 cytokines/chemokines, six markers showed an association with treatment response with the strongest effect observed for interleukin 13. For interleukin 3, a trend was found towards an association with treatment response that was moderated by the individual CRP level as marker of the inflammation status. A combined analysis of both types of biomarkers, genomic and immunological, suggests that the observed effects on treatment outcome are independently from each other. Our results confirm and extend previous findings on the prominent role of genomic FKBP5 biomarkers, and partly also of cytokines/chemokine biomarkers, for predicting antidepressant treatment outcome in depression. Importantly, these effects are moderated by person factors like the FKBP5 genotype and the individual inflammation status, pointing to a suitability of these markers in identifying subgroups of patients that might benefit from personalized treatments targeting the associated biomarker systems.