Background: Elevated S100 calcium binding protein B (S100B) levels in systemic circulation may induce neuroinflammation and reflect greater blood–brain barrier (BBB) dysfunction. Neuroinflammation in patients with major depressive disorder (MDD), in turn, may reduce likelihood of improvement with serotonergic antidepressants. Methods: Levels of S100B were measured in plasma samples obtained prior to initiation of treatment with bupropion-plus-escitalopram, escitalopram-plus-placebo, or venlafaxine-plus-mirtazapine in participants of Combining Medications to Enhance Depression Outcomes trial (n = 153). Depression severity was measured with 16-item Quick Inventory of Depressive Symptomatology Self-Report and anhedonia was measured with 3 items of 30-item Inventory of Depressive Symptomatology. Differential changes in depression severity and anhedonia over acute-phase (baseline, weeks 1, 2, 4, 6, 8, 10, and 12) in the three treatment arms were tested with logS100B-by-treatment-arm interaction in mixed model analyses after controlling for age, gender, and body mass index. Results: There was a significant logS100B-by-treatment-arm interaction for anhedonia (F = 3.21; df = 2, 142; p = 0.04) but not for overall depression severity (F = 1.99; df = 2, 142; p = 0.14). Higher logS100B levels were associated with smaller reductions in anhedonia (effect size = 0.67, p = 0.047) in escitalopram monotherapy but not in the other two arms. Correlation coefficients of anhedonia severity averaged over acute-phase (including baseline) with baseline S100B levels were 0.57, −0.19, and 0.22 for escitalopram monotherapy, bupropion-plus-escitalopram and venlafaxine-plus-mirtazapine arms respectively. Conclusion: Higher baseline S100B levels in depressed patients resulted in poorer response to escitalopram monotherapy. Addition of bupropion, a dopaminergic antidepressant, partially mitigated this effect.