720    

Background:  ESS30009 is a randomized trial comparing tenofovir/abacavir/lamivudine (TDF/ABC/3TC) with efavirenz/ABC/3TC in antiretroviral-naive adults. Early termination of the TDF/ABC/3TC arm resulted from an unexpectedly high rate (50/102; 49%) of early virologic non-response. At week 12, population genotypic analysis of plasma HIV-1 RNA identified the M184V/I and K65R mutations in 40 of 41 (98%) and 22 of 41 (54%) of these virologic non-responders, respectively. Clonal genotypic analysis was conducted on week 12 samples identified by population genotypic analysis as having M184V/I only to determine whether underlying K65R was present at a level below the population genotypic analysis limit of detection.

Methods:  Plasma-derived HIV-1 population genotypic analysis was obtained from ViroLogic for 41 virologic non-responders at week 12. Of the 40 non-responders with M184V/I ± K65R mutations by population genotypic analysis, the median HIV-1 RNA measured at week 12 was 4.2 log₁₀  and 4.8 log₁₀ at baseline. Clonal genotypic analysis was attempted by GlaxoSmithKline on week-12 samples identified by population genotypic analysis as having M184V/I but with no detectable K65R. The non-responders’ samples analyzed were selected to represent a range of week-12 plasma HIV-1 RNA levels (6500 to 320,000 copies/mL).

Results:  Clonal genotypic analysis was performed on 4 virologic non-responders identified by population genotypic analysis as having M184V/I alone at week 12. Of 4 virologic non-responders, 4 (100%) analyzed had evidence for underlying K65R. K65R was observed in 5 of 22 (23%, HIV RNA = 320,000 copies/mL), 1 of 27 (3.7%, HIV RNA = 100,000 copies/mL), 1 of 8 (13%, HIV RNA = 15,000 copies/mL), and 1 of 119 (0.8%, HIV RNA = 6500 copies/mL) clones, respectively. Consistent with a previously reported longitudinal study of the selection for the K65R/M184V double mutant, 3 of 4 (75%) of these week-12 isolates had evidence of K65R linked with M184V. Of these, 1 non-responder had evidence of additional linkage with the Y115F mutation. Finally, the L74V mutation was not observed in this study.

Conclusions:  In this study, 4 of 4 Virologic non-responders originally identified with M184V/I alone, were found to have evidence for underlying K65R. These data suggest that the incidence of K65R at week 12 was higher than the 54% originally reported and illustrate that standard genotypic techniques may miss underlying resistance. The lack of evidence for L74V at week 12, by both population genotypic analysis and clonal genotypic analysis, is consistent with the reported increase in viral susceptibility that this mutation confers to TDF.

Keywords: nucleosides; K65R; M184V