SU87 - GENETIC PREDICTORS OF ANTIPSYCHOTIC RESPONSE TO LURASIDONE IDENTIFIED IN A GENOME-WIDE ASSOCIATION STUDY AND BY SCHIZOPHRENIA RISK GENES

Abstract

Background Schizophrenia (SCZ) is a heterogeneous disorder with a complex genetic contribution to its etiology and response to antipsychotic drugs (APDs). Biomarkers which predict response to APDs increase their benefit/risk ratio. We sought to identify common variants in genes which predict response to lurasidone, an atypical APD, by associating Genome-Wide Association Study (GWAS) data and change in Positive And Negative Syndrome Scale (PANSS) scores from two 6-week randomized, placebo-controlled trials of lurasidone in schizophrenia (SCZ) patients. Methods Population stratification was conducted based on the genotyping data from Affymetrix 6.0 SNP array using 1KG as reference genome. Only genetic-validated patients of European (EUR) or African (AA) ancestries treated with lurasidone (n = 171/131) or placebo (n = 63/54) were included. Treatment response was quantitatively evaluated by change in PANSS-Total (ΔPANSS-T) between baseline and Last Observation Carried Forward (LOCF). Linear regression with an additive model for minor alleles, adjusted for dosage and genetic architecture (3 major PCs), was utilized to test the association between the common genetic variants (MAF > 0.05) and ΔPANSS-T. Both GCTA-GREML and PLINK Polygenic Risk Scoring (PRS) implemented polygenic risk modeling in the EUR group to determine if SCZ risk SNPs identified by PGC GWAS also contributed to the determination of treatment response. Results No Genome-Wide Significant (GWS) findings were found. However, the top genomic loci, with uncorrected p Discussion This is the first evidence from clinical trials that SCZ risk SNPs (or genes) are related to clinical response to an atypical APD. A non-hypothesis driven GWAS study identified genetic biomarkers which predicted treatment response to lurasidone. These results are currently replicated by an independent sample.