Tobacco-attributable disease remains the largest potentially modifiable cause of mortality. Strategies to reduce smoking prevalence include developing more effective smoking cessation treatments. Nicotine metabolism is a predictor of smoking behaviors, including response to treatment. The goal of this Phase I project is to establish the technical and scientific feasibility of developing the Smokescreen(r) Translational Analysis Platform (TL), a service that will predict an individual's nicotine metabolism from Smokescreen(r) Genotyping Array (GTA) data (with content focused for pharmacogenetic analysis) in individuals from multiple world populations. Germline variants in the CYP2A6 gene play a major role in nicotine metabolism activity and are difficult to genotype due to the region's complex alleles and sequence similarity with other genes. We will assess the genotyping and imputation accuracy of the Smokescreen(r) GTA for CYP2A6 variants, using samples from public genomic projects and from three laboratory studies of nicotine metabolism. This assessment will support the technical validity of using Smokescreen(r) GTA CYP2A6 genotypes in predictive models. We will then use existing Smokescreen(r) GTA genotype data from three laboratory studies of nicotine metabolism to compute CYP2A6 haplotypes and predict metabolism in individuals of African, Asian, and European ancestry; the results will be evaluated in relation to observed nicotine metabolism. In addition, we will define new nicotine metabolism pathway prediction models using the latest Bayesian variable selection methodologies that incorporate external functional, clinical, and pathway knowledge. These results will be packaged into prototype reports for the Smokescreen(r) TL service and presented to translational and clinical researchers for feedback. Accomplishments, opportunities and challenges will be summarized for the Smokescreen(r) TL service. Computable models of nicotine metabolism integrated into a unified genetic platform will provide opportunities for future validation in laboratory studies of diverse populations and can be used retrospectively or prospectively in clinical trials of smoking behaviors, including response to smoking cessation therapies. With additional research using clinical trials of smoking cessation therapies, this platform will ultimately provide model estimates and prognostic information for translational research and for use in smoking cessation therapy assignment.
Quitting tobacco-smoking remains a challenge for a significant portion of the U.S. population. This project aims to create a service to help scientist identify factors related to how nicotine is processed in the body and use these factors to improve rates of quitting, reduce side effects, and assess risk of related diseases.
| Baurley, James W; McMahan, Christopher S; Ervin, Carolyn M et al. (2018) Biosignature Discovery for Substance Use Disorders Using Statistical Learning. Trends Mol Med 24:221-235 |
| McMahan, Christopher; Baurley, James; Bridges, William et al. (2017) A Bayesian hierarchical model for identifying significant polygenic effects while controlling for confounding and repeated measures. Stat Appl Genet Mol Biol 16:407-419 |
| Schuit, Ewoud; Panagiotou, Orestis A; Munafò, Marcus R et al. (2017) Pharmacotherapy for smoking cessation: effects by subgroup defined by genetically informed biomarkers. Cochrane Database Syst Rev 9:CD011823 |
