Non response at week 4 as clinically useful indicator for antidepressant combination in major depressive disorder. A sequential RCT.

We aimed to compare the efficacy and tolerability of mirtazapine versus SSRIs and to assess whether “non-response at week 4” may be a clinical indicator for combining mirtazapine and SSRIs for subsequent treatment. One-hundred fifty-four outpatients with MDD were randomized to receive mirtazapine or SSRIs in step I (4 weeks). Non-responders in step I were randomly assigned to either mirtazapine or SSRIs monotherapy or their combination in step IIa while responders in step I continued the same monotherapy in step IIb for 4 weeks. In step I, mirtazapine showed significantly faster improvement as shown by higher remission rate at week 2 with NNT = 8 compared to SSRIs. Somnolence rate was higher in mirtazapine and nausea rate was higher in SSRIs. In step IIa, combination therapy showed a more favorable time course than SSRIs monotherapy. For subjects taking SSRIs in step I, combination therapy showed significant better improvement in the Hamilton Depression Rating (HAM-D) score both at week 6 (p = 0.006) and 8 (p = 0.013) than SSRIs monotherapy. About 80% of responders at week 4 could reach remission at week 8 and 64% of non-responders could not reach remission at week 8 for patients who continued monotherapy. When mirtazapine was added on for SSRIs non-responders at week 4, the remission rate increased by 5% and HAM-D score improved by 4 points. While for mirtazapine non-responders, SSRIs add-on was not equally effective.