Leveraging the Microbiome, Local Admixture, and Machine Learning to Optimize Anticoagulant Pharmacogenomics in Medically Underserved Patients

点击翻译按钮获取中文摘要

ABSTRACT Currently available pharmacogenomic (PGx) algorithms have critical limitations, including a lack of generalizability to non-white populations. Under-representation in clinical studies, the propensity to cause adverse events, and a lack of consideration of admixed populations in clinical PGx guidelines are all factors that contribute to limited utility of PGx algorithms in diverse populations. Thus, our originally awarded proposal focused on improving warfarin stable dose prediction, as it continues to remain one of the most prescribed drugs in the United States and a leading cause of adverse drug events particularly in underserved patients such as African Americans (AAs) and Latinos. Preliminary results from this proposal demonstrate that generation of local ancestry (LA) estimates enables inclusion of admixed populations and improves power in genetic association studies on diverse and admixed populations. Thus, we seek to expand upon our original proposal to perform more inclusive pharmacogenetic studies by generating LA estimates in the large, racially/ethnically diverse AllofUs cohort. We will investigate the relationships between LA and PGx variants and showcase the utility of LA estimates and the AllofUs cohort by identifying novel PGx variants associated with warfarin stable dose. Our overarching hypothesis is that LA can be used to enable genomic association analyses that are more inclusive of admixed and diverse cohorts and to uncover novel findings that were previously overlooked in ancestrally European populations. We will pursue two Specific Aims (SAs) to test this hypothesis: (SA1) Characterize LA for major pharmacogenes and its correlation with global ancestry and PGx variants in diverse populations from AllofUs and; (SA2) Leverage LA to identify novel PGx variants related to warfarin stable dose in admixed AllofUs participants. In SA1, We will estimate LA using RFMix from genome array and sequencing data in the AllofUs Controlled Tier. We will test if LA at clinically relevant pharmacogenes correlates with patient-level global ancestry and presence of clinically relevant pharmacogenomic variants. In SA2, we will incorporate LA estimates from SA1 into genome-wide association analyses for warfarin stable dose using Tractor while controlling for clinical characteristics and clinically relevant PGx variants in admixed individuals from AllofUs, including Hispanic, AA, and multi-race individuals. The outcomes of this work will provide a framework for LA investigation with other PGx drug-gene pairs and enable the identification of novel PGx variants that affect drug response in medically underserved, diverse populations. This research has the potential to identify new sources of variability in warfarin dose, improve the safety and efficacy of warfarin treatment, and reduce disparities in PGx research for medically underserved patients.