Home Search Abstracts View Session E-mail Abstract Author


Session 98 Poster Abstracts
Pharmacogenomics
Session Day and Time: Wednesday, 1 - 4 pm
Poster Hall


552
The Assessment of Multi-locus Gene Effects upon Efavirenz Pharmacokinetics
Tabitha Mahungu*1, S Khoo2, C Smith3, M Youle1, M Johnson1, D Back2, and A Owen2
1Royal Free NHS Trust, London, UK; 2Univ of Liverpool, UK; and 3Univ Coll London, UK

Background:  Previous studies have identified gender, ethnicity, and a G to T change at codon 516 as significant predictors of efavirenz (EFV) Ctrough. We have applied a cumulative scoring algorithm for multi-locus gene effects to further explore the relationship between genotype and phenotype.

Methods:  We included 79 patients who were stable on an EFV-containing (600 mg once daily) regimen, of whom 15 were female and 27 black African or Caribbean. Plasma concentrations were determined by validated high-performance liquid chromatography (HPLC) and the projected Ctrough (C24h) calculated using population pharmacokinetic data. Single nucleotide polymorphisms (SNP) at 6 different loci­—MDR1 (C3435T, G2677T, C1236T), CYP3A4 (A392G), CYP3A5 (A6986G), CYP2B6 (C1459T, G516T), BRCP (C421A) and MRP1 PROM (G260C)—were identified using real-time polymerace chain reaction. Median Ctrough was compared between different genotypes. Inclusion criteria for single nucleotide polymorphisms (SNP) as predictors of high EFV concentrations were:  p <0.05, by t-test; and 0.05 < p < 0.10, but with median EFV Ctrough outside the bioequivalence range (80 to 125%). The cumulative effect of selected SNP on predicting high EFV Ctrough was then assessed.

Results:  Of the total, 78 patients had undetectable HIV viremia; 12 had EFV Ctrough <1000 ng/mL, and 5 had EFV Ctrough >4000 ng/mL. The homozygous TT genotype at position G516T was significantly associated with a higher mean EFV Ctrough (p <0.0001), as was black ethnicity (p = 0.02). SNP at MDR1 (C3435T), CYP3A4 (A392G), and MRP1 PROM (G260C) were also found to be significant predictors of high EFV Ctrough. The association between number of “susceptibility gene” loci and EFV Ctrough was significant when examined using a composite of these 4 SNP:  0 (n = 27; median [EFV] 1287 ng/mL), 1(n = 24; 1507 ng/mL), 2 (n = 11; 1514 ng/mL), >3 (n = 17; 2088 ng/mL); p = 0.02; Cuzick trend test.

Conclusions:  This analysis shows a significant cumulative multi-locus gene effect in relation to EFV Ctrough. However, many other genes might also influence drug disposition. Inclusion of other candidate genes is likely to strengthen these associations but the true contribution of genetic versus environmental factors remains to be elucidated.