Multilocus genetic interactions and response to efavirenz-containing regimens: an adult AIDS clinical trials group study.

TitleMultilocus genetic interactions and response to efavirenz-containing regimens: an adult AIDS clinical trials group study.
Publication TypeJournal Article
Year of Publication2006
AuthorsMotsinger AA, Ritchie MD, Shafer RW, Robbins GK, Morse GD, Labbe L, Wilkinson GR, Clifford DB, D'Aquila RT, Johnson VA, Pollard RB, Merigan TC, Hirsch MS, Donahue JP, Kim RB, Haas DW
JournalPharmacogenet Genomics
Volume16
Issue11
Pagination837-45
Date Published2006 Nov
ISSN1744-6872
KeywordsAcquired Immunodeficiency Syndrome, Adult, Anti-HIV Agents, Anti-Retroviral Agents, Benzoxazines, Clinical Trials as Topic, Genetic Predisposition to Disease, Humans, Models, Genetic, Oxazines, Treatment Failure, Treatment Outcome
Abstract

OBJECTIVE: For the HIV-1 reverse transcriptase inhibitor efavirenz, variant drug transporter gene ABCB1 may predict virologic response but not plasma efavirenz exposure. Conversely, variant drug metabolizing enzyme gene CYP2B6 predicts greater plasma efavirenz exposure but not virologic response. We examined whether long-term responses to efavirenz, and/or plasma efavirenz exposure, are better predicted by multilocus genetic interactions than by individual polymorphisms.

MATERIALS AND METHODS: We studied antiretroviral-naïve study participants randomized to receive efavirenz (with or without nelfinavir) plus two nucleoside analogues in study ACTG 384, and who had DNA available for analysis. Participants were followed up for up to 3 years. Nine single nucleotide polymorphisms in ABCB1, CYP2B6, CYP3A4, CYP3A5 and CYP2C19 were identified. Gene-gene interactions were identified using multifactor dimensionality reduction.

RESULTS: Among 340 efavirenz recipients, higher efavirenz AUC24 h values were associated with a single locus model involving CYP2B6 516G>T (73% accuracy; P<0.001). This was also the best model among blacks (69% accuracy; P<0.001), whereas among whites the best model involved a gene-gene interaction between CYP2B6 516G>T and ABCB1 2677G>T (82% accuracy, P<0.001). Among 155 participants who received efavirenz without nelfinavir, virologic failure was associated with a two-locus interaction between ABCB1 2677G>T and CYP2B6 516G>T (65% accuracy, P<0.001). Toxicity failure was best predicted by an interaction between ABCB1 2677G>T and ABCB1 3435C>T (71% accuracy, P<0.001).

CONCLUSIONS: Multilocus genetic interactions between variant drug metabolism and transporter genes may predict efavirenz pharmacokinetics and treatment responses. This finding may have implications for better individualizing antiretroviral therapy.

DOI10.1097/01.fpc.0000230413.97596.fa
Alternate JournalPharmacogenet. Genomics
PubMed ID17047492
Grant ListAI25903 / AI / NIAID NIH HHS / United States
AI27658 / AI / NIAID NIH HHS / United States
AI27659 / AI / NIAID NIH HHS / United States
AI27666 / AI / NIAID NIH HHS / United States
AI29193 / AI / NIAID NIH HHS / United States
AI38855 / AI / NIAID NIH HHS / United States
AI38858 / AI / NIAID NIH HHS / United States
AI46339 / AI / NIAID NIH HHS / United States
AI54999 / AI / NIAID NIH HHS / United States
GM31304 / GM / NIGMS NIH HHS / United States
HL65962 / HL / NHLBI NIH HHS / United States
NS32228 / NS / NINDS NIH HHS / United States
RR000095 / RR / NCRR NIH HHS / United States