To the Editor: We appreciate Dr. Härtter’s comments on our finding that the CYP2D6 genotype had no effect on the efficacy or tolerability of paroxetine or mirtazapine. He asserts that a difference would not be expected between CYP2D6 genotype groups at a 40 mg/day dose of paroxetine but that an effect would be apparent had we used a lower dose. However, all paroxetine-treated patients received 20 mg/day of this medication for 2 full weeks. There were no significant differences between the CYP2D6 genotype groups in patients receiving 20 mg/day of paroxetine.

Dr. Härtter questions the role of CYP2D6 in the metabolism of mirtazapine. Stormer and colleagues (1) reported a significant role for CYP2D6 in the metabolism of mirtazapine. Furthermore, Kirchheiner et al. (2), who presented dosing guidelines for many antidepressants based on CYP2D6 genotypes, recommended additional research on the impact of CYP2D6 polymorphisms on mirtazapine.

Dr. Härtter considers it "wrong" to study the effects of CYP2D6 genetic variation on paroxetine treatment outcomes. However, as he notes, there is ample support for the role of CYP2D6 in the metabolism of this medication. In addition, Kirchheiner and colleagues (2) recommended major reductions in paroxetine dosing in CYP2D6 poor metabolizers. We found no support for these recommendations.

Dr. Härtter states there is growing evidence that consideration of pharmacokinetics can improve psychotropic efficacy and safety. Actually, the relevance of cytochrome enzymes to clinical psychiatry has been debated for a number of years without resolution. One reason for this continuing controversy is the paucity of prospective pharmacogenetic outcome data from well-designed clinical trials. We believe that our study was the first in psychiatry to actually test prospectively the role of the CYP2D6 genotype on clinical outcomes by using two widely prescribed medications and a statistically meaningful group size.

Finally, we agree with Dr. Härtter that pharmacokinetic considerations can be important in certain clinical settings, but our data indicate that pharmacokinetic variation due to the CYP2D6 genotype should not be a major concern for clinicians during monotherapy with paroxetine or mirtazapine. It is possible that complex interactions may occur between concurrently administered CYP2D6 substrates and inhibitors and that these interactions may be affected by the CYP2D6 genotype. However, we found no such interactions between the CYP2D6 genotype and concurrent medications that affected paroxetine or mirtazapine outcomes. We noted that our results may not apply to other medications, and we hope there will be additional prospective pharmacogenetic trials with other antidepressants.