Background:Adjunctive minocycline shows promise in treating affective and psychotic disorders; however, the therapeutic mechanism remains unclear. Identifying relevant biomarkers may enhance the efficacy of novel adjunctive treatment candidates. We thus investigated the peripheral immune-inflammatory profile in a randomized controlled trial (RCT) of minocycline in major depressive disorder (MDD).
Methods:This sub-study investigated serum samples from a RCT evaluating minocycline (200 mg/day, 12 weeks) in addition to treatment as usual for MDD (ACTRN12612000283875). Of the original sample (N = 71), serum assays were conducted in 47 participants (placebo n = 24; minocycline n = 23) targeting an array of 46 immune-inflammatory analytes including cytokines, chemokines, and acute-phase reactants. General estimating equations (GEE) were used to assess whether analyte concentration at baseline (effect modification) and change in analytes (change association) influenced change in Montgomery-Åsberg Depression Rating Scale (MADRS) score over time. The Benjamini-Hochberg approach was applied when adjusting for false discovery rates (FDR).
Results:GEE models revealed several interaction effects. After adjusting for FDR several change association-models survived correction. However, no such models remained significant for effect modification. Three-way group × time × marker interactions were significant for complement C3 (B = -10.46, 95%CI [-16.832, -4.095], q = 0.019) and IL-1Ra (B = -9.008, 95%CI [-15.26, -2.751], q = 0.036). Two-way group × biomarker interactions were significant for ICAM-1/CD54 (B = -0.387, 95%CI [-0.513, -0.26], q < 0.001) and IL-8/CXCL8 (B = -4.586, 95%CI [-7.698, -1.475], q = 0.036) indicating that increases in the serum concentration of these analytes were associated with an improvement in MADRS scores in the minocycline group (compared with placebo).
Conclusions:Change in complement C3, IL-1Ra, IL-8/CXCL8, and ICAM-1 may be associated with greater change in depressive scores following adjunctive minocycline treatment in MDD. Further investigations are needed to assess the utility of these biomarkers.