1. ¹Depression Clinical and Research Program, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA
2. ²Psychiatric and Neurodevelopmental Genetics Unit, Center for Human Genetic Research, Massachusetts General Hospital, Boston, MA, USA
3. ³Department of Psychiatry, Psychiatric Genetics Program in Mood and Anxiety Disorders, Massachusetts General Hospital, Boston, MA, USA
4. ⁴Department of Psychiatry, University of Texas Southwestern Medical Center at Dallas, Dallas, TX, USA
5. ⁵Department of Clinical Sciences, University of Texas Southwestern Medical Center at Dallas, Dallas, TX, USA

Correspondence: Dr RH Perlis, Massachusetts General Hospital, 15 Parkman Street, WACC 812, Boston, MA 02114, USA. Tel: +1 617 726 7426; Fax: +1 617 724 3028; E-mail: rperlis@partners.org

Received 17 October 2007; Revised 30 November 2007; Accepted 8 December 2007; Published online 20 February 2008.

Abstract

Recent rodent models of antidepressant response implicate a novel set of genes in mechanisms of antidepressant action. The authors examined variants in four such genes (KCNK2 (TREK1), SLC18A2 (VMAT2), S100A10, and HDAC5) for association with remission in a large effectiveness trial of antidepressant treatments. Subjects were drawn from the Sequenced Treatment Alternatives to Relieve Depression (STAR^*D) study, a multicenter, prospective, effectiveness trial in major depressive disorder (MDD). Outpatients with nonpsychotic MDD were initially treated with citalopram for up to 14 weeks; those who did not remit with citalopram were sequentially randomized to a series of next-step treatments, each for up to 12 weeks. Single-nucleotide polymorphisms in four genes were examined for association with remission, defined as a clinician-rated Quick Inventory of Depressive Symptomatology (QIDS-C₁₆) score 5. Of 1554 participants for whom DNA was available, 565 (36%) reached remission with citalopram treatment. No association with any of the four genes was identified. However, among the 751 who entered next-step treatment, variants in KCNK2 were associated with treatment response (Bonferroni-corrected, gene-based empirical p<0.001). In follow-up analyses, KCNK2 was also associated with effects of similar magnitude for third-step treatment among those with unsatisfactory benefit to both citalopram and one next-step pharmacotherapy (n=225). These findings indicate that genetic variation in KCNK2 may identify individuals at risk for treatment resistance. More broadly, they indicate the utility of animal models in identifying genes for pharmacogenetic studies of antidepressant response.