We examined 6 single nucleotide polymorphisms (SNPs) previously reported to be associated with response to iloperidone therapy for association with response to risperidone therapy. Patients with schizophrenia (DSM-IV) were assessed during 2006 and 2007 for response/nonresponse (defined as ≥ 20%/<20% improvement in Positive and Negative Syndrome Scale [PANSS] total score) after 2 weeks of risperidone treatment (2 to 6 mg/d). Responders continued risperidone treatment; nonresponders were randomly assigned to either risperidone or olanzapine treatment (10 to 20 mg/d) for an additional 10 weeks. Associations between change in PANSS total (primary outcome measure), positive, and negative scores and the 6 SNPs were examined in risperidone-treated patients (N = 145). Genotype frequencies and improvement in PANSS total scores were analyzed for those SNPs significantly associated with change in PANSS total score. The SNPs XKR4 rs9643483 and GRIA4 rs2513265 were significantly associated with change in PANSS total response (adjusted P < .05 for both), with the same direction of effect as reported for iloperidone. For patients with nonresponsive genotypes for these SNPs, mean improvement in PANSS total score for African Americans was two-thirds that seen for whites (XKR4: -13.9 versus -21.4; GRIA4: -12.5 versus -20.9). In this retrospective pharmacogenomic analysis, we found that 2 SNPs previously linked to iloperidone response were also associated with response to risperidone. clinicaltrials.gov Identifier: NCT00337662. © Copyright 2012 Physicians Postgraduate Press, Inc.