669 - Abstract (11th CROI)

Session 94 Poster Abstracts
Resistance Predictors of Virologic Response and Clinical Outcome
Tuesday, 1:30 - 3:30 pm
Poster Hall

669
A Genotypic Score for Efavirenz Hypersusceptibility Is Associated with Virologic Response to EFV + Indinavir +/- Abacavir in Nucleoside-experienced Patients
L Demeter*¹, V DeGruttola², S Lustgarten², S Eshleman³, S Hammer⁴, M Fischl⁵, and K Squires⁶
¹Univ. of Rochester, NY, USA; ²Harvard Sch. of Publ. Hlth., Boston, MA, USA; ³Johns Hopkins Univ., Baltimore, MD, USA; ⁴Columbia Univ., New York, NY, USA; ⁵Univ. of Miami, FL, USA; and ⁶Keck Sch. of Med., Univ. of Southern California, Los Angeles, USA

Background: ACTG 368 compared Abacavir (ABC) + elfavirenz (EFV) + indinavir (IDV) vs EFV+IDV in nucleoside-experienced (ZDV [or d4T] + 3TC), PI- and NNRTI-naīve subjects. Subjects (n = 26) who were NNRTI-experienced were assigned to receive open-label ABC+EFV+IDV. Significant virologic benefit of ABC was not seen. We asked whether phenotypic resistance at baseline or the presence of mutations associated with EFV hypersusceptibility predicted virologic response at week 16 and 32.

Methods: Analyses of baseline phenotype (Virco Antivirogram) and genotype (Celera ViroSeq) were performed in 198 and 186 subjects, respectively, who had appropriate baseline testing and RNA results at week 16 and 32, using logistic regression. Samples were classified according to number of of EFV hypersensitivity mutations (T215F/Y, D67N, H208Y, L210W, K103R, V179I) described in a recent report. Virologic failure was defined as a plasma HIV RNA >500 copies/mL.

Results: At baseline, phenotypic resistance was present in 44% to ZDV (fold change in IC₅₀ ≥4); 91% to 3TC (fold change >4.5); 37% to ABC (fold change >3); 5% to EFV (fold change >6); and 1% to IDV (fold change >3). There was no association of phenotypic sensitivity score or phenotypic ABC resistance (defined as either fold change >3 vs >3, <4 vs 4 to <8 vs fold change >8, or log₁₀ fold change) with virologic failure in either treatment group at weeks 16 or 32. Greater numbers of EFV-hypersensitivity mutations decreased the likelihood of virologic failure at week 16 (see the table). The correlation between EFV-hypersensitivity mutations and response rate was not significantly different for those who received ABC compared to those who did not.

# EFV-hypersensitivity mutations	% with VF @ w16 (#/total)	OR (95%CI)
0	44% (26/59)	1.00
1	31% (11/36)	0.56 (0.23,1.32)
2	19% (9/47)	0.30 (0.12,0.71)
3	13% (4/30)	0.19 (0.05,0.56)
4 or 5	7% (1/14)	0.10 (0.01,0.54)

Conclusions: The presence of phenotypic ABC resistance at baseline does not fully explain our inability to detect a significant treatment benefit of ABC in this study. A genotypic score for EFV-hypersensitivity mutations, based on recently reported mutations, correlated well with outcome in this study. We postulate that high response rates due to EFV-hypersensitivity mutations, coupled with baseline ABC resistance, contributed to our inability to identify a benefit of ABC in this patient population.

Keywords: efavirenz hypersusceptibility; abacavir; drug resistance