Pharmacogenetics of Efavirenz Discontinuation for Reported CNS Symptoms

This is an evaluation of genetic associations with efavirenz (EFV) discontinuation for central nervous system (CNS) symptoms within 12 months of treatment. Patients were treated at an HIV primary care clinic in Nashville TN from 1998 to 2012. Previously known SNPs in CYP2B6 and CYP2A6 were used to define metabolizer genotypes (extensive, intermediate, slow metabolizer). Over 500,000 SNPs from genome-wide genotyping were used to define MDS (Multidimensional Scaling) coordinates to account for population stratification. Patients were defined as cases if they discontinued EFV for CNS symptoms within 12 months, otherwise they were defined as controls if they did not stop treatment. Among 563 evaluable participants, the hazard ratio for EFV discontinuation for CNS symptoms was 4.9 (95% C.I. 1.9 TO 12.4, p=0.001) in slow metabolizers compared to extensive metabolizers. This association was very significant in Whites 6.5 (95% CI: 2.3 to 18.8; p = 0.001), but not in Blacks 2.6 (95% C.I. 0.5 to 14.1; p = 0.27). The reason for this difference by race is not clear and warrants further investigation. ]]>
Inclusion Criteria HIV-infected patients ART naÏve patients who Initiated EFV-containing ART regimens during clinical care at Vanderbilt Comprehensive Care Center in Nashville, TN from 1998 to 2012. Patients who had at least one follow-up visit after ART initiation visit. Patients who provided written informed consent for genetic testing through the CFAR Data and Specimen Repository. Patients who had genotype data in CYP2B6, CYP2A6 and whole genome exome data. Exclusion Criteria Initiated the EFV-containing regimen during participation in an ACTG clinical trial. ]]>
This study started in 2012 and was completed in 2016 with the publication in "Pharmacogenetics and Genomics" of the paper "Pharmacogenetics of efavirenz discontinuation for reported central Nervous system symptoms appears to differ by race". Phenotype data was extracted from de-identified electronic medical record from the Vanderbilt Comprehensive Care Center in Nashville, TN. Treatment discontinuation events were reviewed separately by study investigators (Paul Leger and David Haas) and then were adjudicated to decide on causes of efavirenz discontinuation without knowledge of metabolizer status. Genome-wide exome genotype was obtained on stored DNA samples of patients who provided informed consent for genetic studies. Over 500,000 SNPs were generated using Illumina HumanCore Exome at Vanderbilt and were used to generate Multidimensional Scaling (MDS) coordinates to account for population stratification. Furthermore, genetic variations in CYP2B6 and CYP2A6 were genotyped using custom-designed MassARRAY iPlex Gold assay by Sequenom Inc. and TaqMan assay at the Vanderbilt Technologies for Advanced Genomics (VANTAGE). Three SNPs in CYP2B6 (rs28399499, rs3745274, rs4803419) and 1 SNP in CYP2A6 (rs28399433) were used to define metabolizer status according to previously published data from our group.]]>