Search in the Abstract Database

Search Abstracts 2011

* = Presenting author

P360. Candidate genes for azathioprine's efficacy or toxicity in inflammatory bowel disease patients using a exome-wide genotyping analysis

W. Zabala-Fernández1, M. Barreiro-de Acosta2, A. Echarri3, D. Carpio4, S. Pereira5, J. Castro3, A. Lorenzo2, R. Cruz1, A. Carracedo1, F. Barros1

1Fundación Pública Galega de Medicina Xenómica, Santiago de Compostela, Spain; 2University Hospital, Santiago de Compostela, Spain; 3Arquitecto Marcide, Ferrol, Spain; 4CHOP, Pontevedra, Spain; 5CHUVI, Vigo, Spain

Aim: Pharmacogenetics study of Azathioprine (AZA) in IBD are mainly focused on Tiopurin S-methyltransferase (TPMT) and Inosine triphosphate pyrophosphatase (ITPA) genes. However, mutations in these genes can explain only a small group of the great variability in the response to this drug. The purpose of this study was to identify novel genetic variants underlying the considerable individual differences in efficacy and toxicity to AZA treatment.

Materials and Method: We conducted a genome-wide association study (GWAS) in 232 unrelated individuals treated with AZA and sampled from EIGA (Galician Inflammatory Bowel Disease Group), from four hospitals of Galicia, Spain. All the subjects were clinically evaluated, diagnosed and monitored in order to screen adverse events and clinical effects. In the efficacy study, patients were classified as cases if they had a clinical response to AZA, and as controls if they had not. In the toxicity study, patients were classified as cases if they had side effects related to AZA and as controls if they have not. clinical outcome, age and gender. GWAS Genotyping was done using the Affymetrix GeneChip Human 20K cSNP Kit, which contains approximately 20,000 amino acid changing SNPs representing more than 10,000 genes. To identify SNPs significantly associated with efficacy and toxicity related to AZA treatment, we have developed an association analysis by comparing allele and genotype frequencies between cases and controls under different models adjusting the possible effect of some clinical or phenotypic covariates. All quality control and association analyses were conducted using the software package PLINK.

Results: After data cleaning and quality control (genotyping rate >90%, HWE > 0.001, MAF > 0.010), we chose a set of SNPs with probability values ranking between <10–5 and 10–3 (Fisher's exact test) in the association study in the four main categories of our analysis: efficacy, development of side effects and within side effects, pancreatitis and myelosuppression. A total of 80 SNPs were identified from GWAS analysis, related to diverse biological pathways, such as DNA repair, transcriptional factors, and basal metabolism, among others.

Conclusions: In AZA metabolism multiple gene interactions may be implicated in a complex network of events responsible for the wide heterogeneity of its response. Here we present a catalog of genes related to efficacy and development of side effects to treatment, in the first GWAS study from full exome. These results may represent an important step in elucidating the genetic basis of different patient's response to AZA.