PGx7666: Exploratory pharmacogenetic investigation of efficacy response to advair in COPD studies 116135, 116136, 111117, 114930, 114951, 116134, 112352, 113107, 113109, 116974

Trial description: ADVAIR (fluticasone propionate/salmeterol) has been approved for both asthma and COPD in the United States (US), Europe, and many other countries. Fluticasone propionate is a synthetic trifluorinated corticosteroid with anti-inflammatory activity, while salmeterol, a long acting beta2- agonist, binds to the beta2 adrenergic receptor which results in relaxation of the bronchial smooth muscles and increase in airway diameter. The current evaluation is to assess if genetic variation is associated with efficacy response in COPD subjects treated with ADVAIR from 116135, 116136, 111117, 114930, 114951, 116134, 112352, 113107, 113109 and 116974 studies. Approximately 2000 subjects with COPD treated with ADVAIR provided a PGx sample and will be included in this analysis.This exploratory PGx study includes Intent-To-Treat (ITT) subjects enrolled in aforementioned ten GSK clinical studies, who received at least one dose of drug ADVAIR, provided optional written consent for genetics research, a deoxyribonucleic acid (DNA) sample, and were successfully genotyped. The primary endpoint to be analyzed is change from baseline trough forced expiratory volume in one second (FEV1) at the end of the study treatment period (Day 84). Association between genotype and primary endpoint will be tested using a linear regression model, with inverse normal transform on residuals, including the top 10 genetic ancestry principle components, study, FEV1 baseline, country of study, smoking status, and treatment arm as covariates, under an additive genetic model. Two secondary endpoints: 1) change from baseline St. George’s Respiratory Questionnaire (SGRQ) score at the end of study treatment period and 2) response by SGRQ score at the end of study treatment period, will be also be tested for association with genome-wide genetic variation. Similarly, linear and logistic regression models will be used for genetic association with the secondary endpoints, again adjusting for top 10 genetic ancestry principle components, study, SGRQ baseline, country of study, smoking status, and treatment arm. Additional exploratory analyses, e.g. analysis by subgroup or inclusion of additional covariates, may be conducted to further investigate the influence of genetic variants on clinical study outcomes in response to drug treatment. Analyses will be conducted by pooling subjects across studies. A dummy variable representing the source clinical study will be included in the regression model to account for study differences. For genotype association analyses, all genome-wide variants, genotyped or imputed (MAF >=1% and imputation Rsq >= 0.3), will be included in the linear or logistic regression, under an additive genetic model. Variants will the considered statistically significant at the genome-wide level (p<=5x10-8).Study samples will be genotyped by BioStorage Technologies/Bioprocessing Solutions Alliance (Piscataway NJ, USA) using the Affymetrix Axiom Biobank Genotyping Array with GSK custom content v2. Haplotype reference data from the 1000 Genomes phase_3 project will be used to impute variants for analysis across the genome including the HLA region.