A systematic review of the antidepressant effects of glucagon-like peptide 1 (GLP-1) functional agonists: Further link between metabolism and psychopathology

Introduction An increasing amount of preclinical and clinical evidence links together metabolic regulations and psychopathological mechanisms, in particular linking mood disorders with changes in Glycogen Synthase Kinase 3 beta and 5′Adenosine Monophosphate-activated Protein Kinase expression and activity. New hypoglycemic drugs, including thiazolidinediones and glucagon-like peptide 1 (GLP-1) functional agonists, which work by these mechanisms, have also been described as potential antidepressants. The putative role of thiazolidinediones in depression has been already supported, but no clear evidence exists yet for GLP-1 functional agonists. We conducted a systematic review and meta-analysis of the literature to describe the effect of GLP-1 functional agonists on depression rating scales and either support or confute a potential antidepressant role. Methods We searched the PubMed and Scopus databases for terms related to DPP-4 inhibitors and GLP-1 receptor agonists, and depression, including symptoms and rating scales with acronyms and full names. We included longitudinal interventional and observational studies on GLP-1 functional agonists used for depression symptoms. We applied a random effects meta-analysis on standardized mean differences before-after treatment, comparing GLP-1 functional agonists versus control treatments. Results Literature searches found 815 papers, 8 of which were eligible for meta-analysis. Both control treatments (-0.67, 95%C.I. -0.99 – -0.36, Z = 4.24, p < 0.0001) and GLP-1 functional agonists (-1.28, 95%C.I. -2.34 – -0.21, Z = 2.35, p = 0.02) resulted in a significant reduction of depression rating scores, although GLP-1 functional agonists tended to be superior. When a selection was made, including only studies conducted on diabetic patients that did not exclude depressed patients, the effect of GLP-1 functional agonists (-2.09, 95%C.I. -2.28 – -1.91, Z = 22.5, p < 0.00001) was significantly superior to that of control treatments (-0.57, 95%C.I. -0.66 – -0.49, Z = 13.6, p < 0.00001). Discussion Results of this meta-analysis must be carefully considered, since the amount of studies available was low and heterogeneity was high. If further trials will confirm this hypothesis, GLP-1 functional agonists may be considered as antidepressants, either as adjuncts or in mono-therapy, with a peculiar value for preventing the adverse metabolic effects of long-term antipsychotic therapies used in rehabilitation.