Margot, Nicolas A MA; Enejosa, Jeff MD; Cheng, Andrew K MD, PhD; Miller, Michael D PhD; McColl, Damian J PhD; and the Study 934 Team
Combination therapies composed of two classes of antiretroviral (ARV) drugs are the standard of care for the first-line treatment of HIV-1 infection. Significant progress has been made in both the efficacy and simplicity of dosing of the ARV treatment regimens available to HIV-1-infected subjects who are either ARV therapy-naïve or experienced. Fixed-dose combination (FDC) regimens, including dual nucleoside/tide (NRTI) combinations such as emtricitabine + tenofovir disoproxil fumarate (FTC + TDF), lamivudine + zidovudine (3TC + ZDV), and lamivudine + abacavir (3TC + ABC) have greatly simplified first-line ARV therapy. The approval of a FDC composed of two NRTIs (FTC + TDF) and the nonnucleoside reverse transcriptase inhibitor (NNRTI) efavirenz (EFV), dosed once-daily as a single pill, has further simplified initial therapy options. Despite these advances, the emergence of ARV drug resistance while on therapy and transmission of drug-resistant viruses can significantly impact the successful long-term treatment of HIV-1 infection1,2
Clinical trial GS-01-934 (Study 934) was an open-label, 3-year comparative study in ARV treatment-naïve subjects that investigated the once-daily combination of the NRTIs FTC and TDF compared with the twice-daily FDC of the NRTIs 3TC and ZDV.3-5 The once-daily NNRTI EFV was used as the third agent in both treatment groups. Through 144 weeks, significantly more subjects in the FTC + TDF + EFV group compared with the 3TC + ZDV + EFV group reached and maintained plasma HIV-1 RNA less than 400 copies/mL (71% on FTC + TDF + EFV versus 58% on 3TC + ZDV + EFV; P = 0.004). Furthermore, there was a trend toward greater CD4+ T-cell increases in the FTC + TDF + EFV group compared with the 3TC + ZDV + EFV group (312 versus 271 cells/mm3; P = 0.09).5
The mutations in HIV-1 reverse transcriptase (RT) involved in resistance to the ARV drugs used in Study 934 have been well characterized. Both FTC and 3TC select the M184V/I mutation in HIV-1 RT, resulting in high-level (greater than 100-fold) reduced susceptibility to both drugs.6-9 Tenofovir and its prodrug, TDF, select the K65R mutation in RT that confers three- to four-fold reduced susceptibility to tenofovir in antiviral assays and can reduce phenotypic susceptibility to other NRTIs, including ABC, FTC, 3TC, didanosine (ddI) and zalcitabine.10-17 The K70E RT mutation in RT may also be associated with TDF therapy, but the level of resistance to TDF caused by K70E is lower than that of K65R.18 ZDV (and stavudine [d4T]) can select several mutations in HIV-1 RT, including M41L, D67N, K70R, L210W, T215Y/F, and K219Q/E/N/R mutations, collectively known as thymidine-analog associated mutations (TAMs).19-21 TAMs confer varying levels of phenotypic resistance to ZDV and can mediate cross resistance to all other approved NRTIs.22,23
Primary resistance to the NNRTI EFV (EFV-R) is mediated by the K103N, Y188L, and G190S/A mutations in RT, which confer high-level phenotypic resistance to EFV.24 Additional NNRTI resistance (NNRTI-R) mutations such as L100I, V106M, V108I, Y181C/I, and P225H enhance EFV-R or can mediate cross resistance to EFV.24-26 NNRTI-R is among the most prevalent form of genotypic resistance detected in ARV treatment-experienced subjects and is among the most common type of transmitted drug resistance observed in ARV-naïve subjects.2,27
The purpose of this report is to describe the genotypic and phenotypic resistance patterns that developed in the HIV-1 RT gene from the plasma HIV-1 of subjects with virologic failure in both treatment groups of Study 934 through Week 144. Analyses of baseline resistance were also conducted to determine the effect of transmission of viruses with pre-existing NRTI and NNRTI-associated drug resistance mutations on treatment response to both of the EFV-containing regimens used in Study 934.
Overall Clinical Study Design
Study 934 was a 144-week randomized, open-label, multicenter, active-controlled study comparing two simplified dosing regimens containing emtricitabine (FTC), tenofovir disoproxil fumarate (tenofovir DF or TDF), and EFV, all dosed once-daily, with an active control arm composed of the twice-daily fixed-dose combination of lamivudine/zidovudine (3TC + ZDV; Combivir) plus once-daily EFV. ARV-naïve, HIV-1 infected subjects with HIV-1 RNA levels greater than 10,000 copies/mL at study entry were randomized 1:1 to either treatment regimen and subjects were stratified on the basis of screening CD4+ cell count (less than or greater than 200 cells/mm3). For subjects randomized to receive FTC + TDF + EFV, FTC and TDF were administered individually for the first 96 weeks and then, after regulatory approval, as the once-daily FTC + TDF FDC (Truvada) through Week 144. 3TC + ZDV were administered as the twice-daily FDC Combivir throughout the entire study. The details of the clinical protocols, informed consent, Institutional Review Board approval, and efficacy results for this study at Weeks 48, 96, and 144 have been described.3-5
A total of 517 ARV therapy-naïve subjects were randomized into the study; eight subjects were subsequently excluded from the efficacy analyses either because they did not receive study medications (n = 6) or as a result of protocol violations (n = 2), specifically ARV treatment experience. The intent-to-treat (ITT) population consisted of the remaining 509 subjects (FTC + TDF + EFV group, n = 255; 3TC + ZDV +EFV group, n = 254). Baseline genotypic analysis of all enrolled study subjects further identified 22 subjects harboring one or more primary NNRTI-R mutations, which placed them at risk of suboptimal therapy on an EFV-containing regimen. Study investigators were informed of the baseline resistance status of these specific 22 subjects; however, nine of 22 of these subjects had experienced virologic failure before being discontinued from study medications. Twenty of 22 subjects with baseline NNRTI resistance were discontinued from their study medications; however, all 22 subjects were excluded from the primary efficacy analyses of Study 934. The remaining 487 subjects, with no detectable primary NNRTI-R mutations at baseline, formed the modified-ITT (mITT) population, the primary efficacy analysis population for Study 934 (FTC + TDF + EFV group, n = 244; 3TC + ZDV + EFV group, n = 243).
Resistance Analysis Population
Determination of plasma HIV-1 RNA levels in Study 934 subjects used the Roche Amplicor HIV-1 Monitor Test (Version 1.5 Ultrasensitive) with a lower limit of detection of 50 copies/mL of plasma HIV-1 RNA. Analysis of protease/reverse transcriptase (PR/RT) resistance development (genotypic and phenotypic resistance) in Study 934 was performed for all subjects in the ITT population meeting resistance analysis criteria. The criteria for inclusion in the resistance analyses at Weeks 48, 96, or 144 were: 1) subjects who had been maintained on study drug and had confirmed plasma HIV-1 RNA concentrations 400 copies/mL or greater by Weeks 48, 96 or 144; 2) subjects who had been maintained on study drug and achieved less than 400 copies/mL of plasma HIV-1 RNA on at least one occasion but then subsequently had confirmed rebound in plasma HIV-1 RNA concentrations 400 copies/mL or greater on at least two consecutive visits; or 3) subjects who discontinued study drugs or the study on or after Week 8 (the first time point with available plasma storage for resistance analysis) with plasma HIV-1 RNA concentrations 400 copies/mL or greater on the last study visit. Subjects who discontinued the study with 400 copies/mL or greater of plasma HIV-1 RNA before Week 8 were not analyzed for resistance development as a result of the lack of available plasma samples.
Plasma samples chosen for analysis of HIV-1 resistance development represented the confirmed failure time point that included interim time points in the study, not just at the predefined study end points, Weeks 48, 96, and 144. For example, if confirmed rebound in plasma HIV-1 RNA 400 copies/mL or greater occurred at Week 24, the Week 24 plasma sample was analyzed and counted toward the Week 48 resistance analysis; likewise, resistance data generated on a Week 72 plasma sample counted toward the Week 96 analysis. Furthermore, subjects in the study could be analyzed for resistance development more than once if they continued on study drug and were confirmed virologic failures in more than 1 year. For the purposes of cumulative analyses through Week 144, all resistance mutations developed by any subject in multiple years were included in the final tally. Subjects on study drug who experienced a single time point or “blip” with HIV-1 RNA 400 copies/mL or greater at Week 48 or Week 96 were classified as “unconfirmed rebound” and were subsequently analyzed only if they then experienced confirmed rebound in HIV-1 RNA 400 copies/mL or greater in the sample immediately after the unconfirmed rebound. These subjects counted toward the tally for the year of the study in which they had confirmed HIV-1 RNA 400 copies/mL or greater; for example, unconfirmed rebound at Week 48 with confirmation and analysis at Week 60 counted toward the Week 96 resistance analysis. Subjects with unconfirmed rebound in HIV-1 RNA 400 copies/mL or greater at Week 144 were analyzed for resistance development because Week 144 represented the last time point of the primary study. Subjects with nonsequential blips in HIV-1 RNA 400 copies/mL or greater, that did not subsequently confirm, were not analyzed unless such a blip represented the last time point on the study.
The TLOVR definitions of virologic failure used for calculation of efficacy versus those used for analysis of resistance (last time point 400 copies/mL or greater) do allow for some “TLOVR exceptions.” Specifically, two subjects (one from each treatment group) were classified as TLOVR responders but were included in the resistance analysis because their respective Week 144 time points were 400 copies/mL or greater of plasma HIV-1 RNA. Conversely, one subject from the 3TC + ZDV + EFV arm was classified as a TLOVR failure as a result of suboptimal virologic response but was not analyzed for resistance development because this subject never had confirmed rebound in HIV-1 RNA and the viral load at the last time point on study was less than 400 copies/mL. Besides these TLOVR exceptions, all other subjects defined as virologic failures by TLOVR were included in the resistance analysis population along with subjects who were TLOVR failures for other reasons but had 400 copies/mL or greater at the last time point.
Genotypic and Phenotypic Analyses
Monogram Biosciences, Inc. (South San Francisco, CA) was the designated reference laboratory for the protocol-specified genotypic and phenotypic resistance analyses for Study 934, including both baseline and postbaseline virologic failure analyses. Plasma HIV-1 genotyping of baseline samples was generated using standard population-based sequencing methods in the GeneSeq Assay, which included amino acid residues 1 to 305 of HIV-1 RT and all 99 amino acid residues of HIV-1 protease (PR).28 Assignment of PR/RT mutations in the GeneSeq Assay was by comparison of the test sequence to a wild-type reference sequence (NL4-3 strain of HIV-1). Baseline HIV-1 subtype information was also provided with the GeneSeq Assay. Postbaseline plasma HIV-1 genotypic and phenotypic data from subjects with virologic failure were obtained using the PhenoSense GT Assay, which provided both genotypic and phenotypic data for all US Food and Drug Administration-approved NRTIs, NNRTIs, and PIs available at the time.28 HIV-1 phenotypic data derived from Study 934 subjects were expressed as fold change in susceptibility compared with the wild-type reference (NL4-3 strain of HIV-1). The NNRTI etravirine and the ritonavir-boosted PIs darunavir and tipranavir were not included on the GeneSeq and PhenoSense GT Assays during the period in which Study 934 was conducted. HIV-1 derived from baseline plasma samples from subjects with baseline NNRTI-R and baseline TAMs were also retrospectively analyzed for PR/RT phenotypes using the PhenoSense Assay.
Definitions of Resistance
When Study 934 was initiated, primary EFV-R mutations were defined according to the 2004 International AIDS Society definitions and included the HIV-1 RT mutations K103N, V106A/M, Y181C/I, Y188L/H/C, G190A/S/E, and P225H.29 Additional secondary NNRTI-R mutations included L100I, V108I, M230L, and P236L. Definitions of NRTI-associated resistance mutations (NAMs) also used the International AIDS Society 2004 guidelines and included HIV-1 RT mutations A62V, K65R, T69D, T69SS insertions, L74V/I, V75I, F77L, Y115F, F116Y, Q151M, and M184V/I. Thymidine analog mutations in RT were defined as M41L, D67N, K70R, L210W, T215Y/F, and K219Q/E/N/R. The development of E44D, K70E, and V118I RT mutations was also studied. NRTI or NNRTI mutations at these codons were defined as being associated with resistance if the phenotype of the resulting virus was above the PhenoSense GT assay cutoffs for the respective NRTIs and NNRTIs tested (FTC, TFV, 3TC, ZDV, ddI, d4T, ABC, EFV, NVP, and DLV). Primary PI-associated resistance mutations were defined as D30N, L33I/F, M46I/L, G48V, I50L/V, V82A/F/L/S/T, I84V, and L90M mutations in HIV-1 protease.29
Protease/Reverse Transcriptase Replication Capacity
Viral replication capacity (RC) was measured in the PhenoSense GT Assay using an infectious molecular clone of HIV-1 derived from NL4-3. PR and RT genes were amplified from plasma HIV-1 RNA and cloned into the NL4-3 vector as previously described.28 Recombinant viruses were prepared by transfection of the vector DNA into 293 cells along with an expression vector that encodes the Env protein of murine leukemia virus. Virus was harvested 2 days after transfection and was used to inoculate new 293 cell cultures, which were then incubated for another 2 to 3 days. Input virus at infection was normalized based on measurement of luciferase signal after transfection. Luciferase expression was driven from a cassette that replaces the HIV-1 Env gene in the NL4-3 molecular clone. Luciferase activity (after normalization) was used as a measure of PR/RT RC, which was expressed as percent RC of a recombinant NL4-3 reference (defined as 100%) as described.28
Analysis of Virologic Failure According to Baseline HIV-1 Genotype
Baseline RT genotypes were compared with the wild-type reference sequence (NL4-3) and assigned as having “any change” or “no change” for each of the RT codons 1 to 305 (the region of RT sequenced in the GeneSeq Assay). Subjects were then grouped according to whether they had confirmed virologic failure as defined by the algorithm of confirmed 400 copies/mL or greater of HIV-1 RNA by Week 144. Additionally, subjects with unconfirmed rebound 400 copies/mL or greater of HIV-1 RNA at Week 144 were also considered virologic failures in this analysis (as in the postbaseline resistance analysis). P values for the comparison of virologic failure with “no change” versus “any change” at each of RT codons 1 to 305 were calculated using a two-sided Fisher's exact test. No correction was made for multiple comparisons in this analysis.
HIV-1 genotypic and phenotypic data were maintained in an Access database (Microsoft, Redmond, WA). Mutation development analyses were conducted in Excel (Microsoft). Statistical analyses of the impact of baseline resistance mutations on treatment response were conducted using SAS (Cary, NC). P values from two-sided tests that were <0.05 were considered statistically significant. Fisher's exact test was used unless otherwise specified. Statistical analyses of RC between genotypic groups were performed with Mann-Whitney tests using GraphPad InStat (GraphPad Software, San Diego, CA).
Baseline Genotypic Analyses
HIV-1 PR/RT genotypes at baseline were obtained on 501 of 509 (98.4%) subjects in the ITT population; eight subjects (1.6%) lacked genotypic data as a result of assay failure. The majority of subjects had genotypes that were considered wild-type at baseline relative to the NL4-3 reference; however, 24 of 501 subjects with baseline genotypic data (4.8%) had evidence of baseline NNRTI-R mutations. Of these, 22 of 24 had primary NNRTI-R mutations; the K103N EFV-R mutation in RT was the most common mutation detected (n = 17) followed by G190A/E mutations (n = 4) and the Y181C mutation (n = 1). Two of 24 subjects had the secondary NNRTI-R mutation V108I only (Table 1). The 22 subjects with primary NNRTI-R at baseline were distributed evenly between the two treatment groups (n = 11 in both treatment groups) and were from 16 sites in both the United States and the European Union. Subjects with primary NNRTI-R at baseline were excluded from the primary efficacy analyses and were analyzed separately from subjects having no primary NNRTI-R mutations at baseline (mITT population, n = 487 subjects).
NRTI-associated resistance mutations were observed at baseline in 31 subjects (6.2%) and were similarly distributed between the two arms (FTC + TDF + EFV, n = 18; 3TC + ZDV + EFV, n = 13) (Table 1). The most common NAMs observed were substitutions at codon T69 of RT (n = 14 patients; Table 1), including a novel single amino acid insertion T69TT in one subject (along with D67N and K219Q). TAMs were found in 13 of 501 subjects with genotypic data at baseline, including two of 22 subjects with primary NNRTI-R mutations. TAMs detected at baseline included M41L (n = 7), K219N/Q (n = 4), D67N (n = 3), L210W (n = 2), and K70R (n = 1). The majority of subjects with baseline TAMs (without NNRTI-R) had only one TAM as a full mutant or as a mixture with wild-type; two of 13 subjects had more than one TAMs at baseline, including D67N + K219Q mutations or M41L + L210W + T215C mutations (the latter a T215 reversion mutation indicative of possible transmission of a ZDV-resistant virus), respectively. T215 reversion mutations were found in nine subjects at baseline (T215S, n = 3; T215D, n = 2; T215L, n = 2; T215C, n = 1; and T215E, n = 1); six of nine also had TAMs at baseline. One subject also had a K219K/T mixture at baseline. The M184V mutation was detected in only one subject at baseline along with TAMs and primary NNRTI-R mutations. Other potential NAMs detected at baseline were V75M/L (n = 2). The K65R or L74V/I mutations and the multi-NRTI-R patterns (Q151M or T69 insertion patterns) were not detected in any subject at baseline. Twelve subjects (2.4%) entered the study with one or more PI-R mutations, including L33F/I (n = 2 each), M46I/L (n = 4 and n = 2, respectively), V82A (n = 3), and L90M (n = 3).
Analyses of Baseline Phenotypic Resistance
The plasma HIV-1 from subjects with primary NNRTI-R at baseline (n = 22) were analyzed phenotypically using the PhenoSense Assay (Table 2). Mean phenotypic susceptibilities to EFV, NVP, and DLV (not shown) at baseline were 28.6-, greater than 68.2-fold, and greater than 56.8-fold reduced relative to the wild-type reference (NL4-3), respectively, and were similar across both treatment arms. Phenotypic susceptibility to EFV in these subjects ranged from fully susceptible to greater than 242-fold reduced, ie, above the upper limit of quantification of the PhenoSense Assay. Three subjects whose HIV-1 carried the K103K/N, Y181Y/C, or G190G/E RT mutations at baseline had wild-type susceptibility to EFV at baseline. These genotype/phenotype discordances were consistent with the detection of genotypic mixtures of mutant and wild-type viruses in these subjects.
The HIV-1 from the majority of these 22 subjects harbored NNRTI-R only, and the mean phenotypic susceptibilities to NRTIs in these 22 subjects were similar to wild-type (Table 2). For TDF, ZDV, d4T, ddI, and ABC, mean phenotypic susceptibility was 0.9- to 1.0-fold of the NL4-3 reference. For FTC and 3TC, mean phenotypic susceptibility was reduced greater than four-fold, reflecting the presence of the M184V mutation in one subject at baseline. Mean fold change in susceptibility to FTC and 3TC in the 21 subjects with baseline NNRTI-R but without the M184V mutation was 0.86- or 0.87-fold, respectively. Two of 22 subjects also had phenotypic resistance (above the lower cutoff) to one or more boosted PIs, consistent with the presence of primary PI-R mutations (data not shown).
Susceptibilities to all NRTIs in subjects with baseline TAMs only (11 of 487) or T215 reversion mutations only (two of 487) were similar to wild-type with a mean fold change of 0.9- to 1.1-fold to all NRTIs (Table 3). One subject with M41L, L210W, and T215C RT mutations at baseline had a 2.68-fold reduction in ZDV susceptibility (above the ZDV cutoff, 1.9-fold). This subject was randomized to FTC + TDF + EFV treatment and achieved confirmed less than 400 copies/mL of HIV-1 RNA through Week 144. For the 13 subjects with baseline TAMs, mean phenotypic susceptibilities to the NNRTIs EFV, NVP, and DLV (the latter not shown) were 1.1-fold, 1.2-fold, and 1.3-fold reduced, respectively. One subject had a V108I RT mutation at baseline and showed reduced susceptibility to NVP (4.94-fold), but achieved less than 400 copies/mL on 3TC + ZDV + EFV through Week 144. All other subjects with baseline TAMs had phenotypic susceptibilities to NNRTIs that were below the respective phenotypic cutoffs. One subject in this group also had genotypic and phenotypic resistance to some PIs but achieved less than 400 copies/mL of HIV-1 plasma RNA on 3TC + ZDV + EFV (data not shown).
Postbaseline Resistance Analyses (modified intent-to-treat population)
Within the mITT population, significantly more subjects in the FTC + TDF + EFV group compared with the 3TC + ZDV + EFV group achieved confirmed plasma HIV-1 RNA levels less than 400 copies/mL (71% and 58%, respectively; P = 0.004) through 144 weeks.5 Through 144 weeks, 50 of 487 (10.3%) subjects in the mITT population met criteria for analysis of resistance development. The difference between the proportions of patients classified as TLOVR failures in each arm versus those analyzed for resistance development reflected the fact that TLOVR failures were not always virologic failures. Subjects defined as TLOVR failures discontinued the study for several reasons, including adverse events, loss to follow up or withdrawal of consent, and may have been less than 400 copies/mL of plasma HIV-1 RNA at study discontinuation; if so, they did not qualify for analysis of resistance development. Furthermore, subjects who discontinued the study before Week 8 with greater than 400 copies/mL were not analyzed because no plasma samples for genotypic and phenotypic resistance analyses were available before Week 8.
Reflective of the difference in efficacy between the two treatment groups, numerically fewer subjects from the FTC + TDF + EFV group compared with the 3TC + ZDV + EFV group were in the resistance analysis population (RAP, 19 of 244 [7.8%] versus 31 of 243 [12.8%], respectively; Table 4). Postbaseline genotypic and phenotypic data were obtained for all 19 FTC + TDF + EFV subjects and for 29 of 31 3TC + ZDV + EFV subjects; postbaseline PhenoSense GT data were not obtained for two subjects from the 3TC + ZDV + EFV group as a result of assay failure on multiple attempts.
Similar proportions of subjects from both treatment groups developed any form of resistance mutation in RT by Week 144 (FTC + TDF + EFV, n = 13 of 19 [68.4%]; 3TC + ZDV + EFV, n = 22 of 29 [75.9%]). EFV-R was the most common form of resistance that developed in both groups (FTC + TDF + EFV, n = 13 of 19 [68.4%]; 3TC + ZDV + EFV, n = 21 of 29 [72.4%]; Table 4). The K103N RT mutation was the most common EFV-R mutation detected and it developed in 26 subjects (FTC + TDF + EFV, n = 8 of 19 [42.1%]; 3TC + ZDV + EFV, n = 18 of 29 [62.1%]). Although the difference in the frequency of development of the K103N mutation specifically was statistically significant (P = 0.046), this difference is unlikely to be clinically significant. Overall, the frequency of development of EFV-R mutations between the two treatment groups was not significantly different (P = 0.16). Other EFV-R mutations that developed included K101E, K103E, V179D, Y188H, G190A/S, and M230L mutations in RT.
The most common NAM that developed in both groups was the M184V/I mutation and it developed significantly less frequently in the FTC + TDF + EFV group by Week 144 (FTC + TDF + EFV, two of 19 [10.5%]; 3TC + ZDV + EFV, 10 of 29 [34.5%]; P = 0.021). EFV-R was also detected in 11 of 12 subjects who developed the M184V/I mutation; only one subject in the 3TC + ZDV + EFV group developed M184V/I in the absence of any other NAMs or NNRTI-R.
TAMs developed in two subjects in the 3TC + ZDV + EFV group only and in conjunction with the development of EFV-R. One subject developed a D67D/N mixture, and the other subject added a K70K/R mixture plus M184V. The latter subject had D67N and T215D mutations at baseline suggestive of transmission of an NRTI-R virus. Two subjects (one in each group) developed noncanonic mutations at RT codon D67 (D67D/G mixtures) that were not associated with phenotypic resistance to any NRTIs; one of these subjects (3TC + ZDV + EFV group) also developed the V118I RT mutation. One subject (3TC + ZDV + EFV group) developed the A62V RT mutation along with M184V and K103N mutations. By Week 144, no subject in either group had developed either the K65R or the K70E RT mutations. Development of resistance to more than one class of antiretroviral drugs was also more prevalent in the 3TC + ZDV + EFV group compared with the FTC + TDF + EFV group (10 of 29 [34.5%] versus two of 19 [10.5%], respectively; P = 0.021). Nine of 10 subjects in the 3TC + ZDV + EFV group and both subjects in the FTC + TDF + EFV group who developed M184V also developed EFV-R. Additionally one 3TC + ZDV + EFV subject developed a TAM plus EFV-R (without M184V/I). Development of EFV-R and NRTI-R in both groups occurred predominantly by Week 48 (Fig. 1). Nine of 13 (69.2%) FTC + TDF + EFV subjects and 16 of 21 (76.2%) 3TC + ZDV + EFV subjects who developed EFV-R did so by Week 48. Both subjects in the FTC + TDF + EFV group and seven of 10 (70%) subjects in the 3TC + ZDV + EFV group whose HIV-1 developed M184V/I also did so by Week 48. One of the two 3TC + ZDV + EFV subjects whose HIV-1 developed TAMs did so by Week 48.
Postbaseline phenotypic data were obtained for 48 of 50 subjects in the mITT resistance analysis population (Table 5). Development of phenotypic resistance to the NNRTIs EFV, NVP, and DLV (the latter not shown) was observed (mean fold changes of greater than 36.8-, greater than 51.5-, and greater than 31-fold, respectively) and were similar between the two groups. For subjects in both groups whose HIV-1 developed M184V/I, there was a mean fold change of greater than 90-fold to both FTC and 3TC. Subjects who did not develop either EFV-R or the M184V/I mutation in their HIV-1 retained wild-type sensitivity to NNRTIs and to both 3TC and FTC, respectively. No subject in the resistance analysis population from either group developed reduced phenotypic susceptibility to TDF.
Both the K65R and M184V/I mutations have been shown to reduce viral RC.13,30-32 The mean RC of the 12 viruses with the M184V/I mutation (from postbaseline samples) was significantly reduced versus that of the 36 viruses without M184V (mean RC of 50% versus 88%, respectively; P < 0.02, Mann-Whitney test). The majority of viruses with M184V/I (11 of 12) also developed EFV-R mutations; thus, an effect of EFV-R mutations on RC cannot be excluded.
Development of Resistance in Patients With Baseline Nonnucleoside Reverse Transcriptase Inhibitor Resistance
Twenty of 22 subjects with primary NNRTI-R mutations at baseline were discontinued from their EFV-containing study regimens. Nine subjects (FTC + TDF + EFV group, n = 4; 3TC + ZDV + EFV group, n = 5) had plasma HIV-1 RNA concentrations 400 copies/mL or greater at the time of discontinuation and met resistance analysis criteria and 11 subjects had HIV-1 RNA less than 400 copies/mL at the time of discontinuation and were not analyzed. Two subjects (one in each group) continued on their study medications and achieved and maintained plasma HIV-1 RNA levels less than 400 copies/mL and less than 50 copies/mL through 144 weeks. Both subjects had the K103N mutation at baseline with phenotypic EFV-R (8.4-fold and 29.0-fold reduced susceptibility) but both achieved virologic suppression, apparently on dual NRTI therapy.
Of the nine subjects with virologic failure, seven of nine developed additional RT resistance mutations (FTC + TDF + EFV group, n = 3; 3TC + ZDV + EFV group, n = 4) and two of nine subjects had postbaseline genotypes that were unchanged from baseline (K103N at baseline and virologic failure). All seven subjects whose HIV-1 developed additional mutations developed M184V and three subjects developed additional EFV-R mutations (FTC + TDF + EFV group, n = 2 developed V108I + P225H or L100I; 3TC + ZDV + EFV group, n = 1 also developed V108I + P225H). One subject in the FTC + TDF + EFV group with T215L at baseline developed a T215C/F/L/R mixture. No subject in the FTC + TDF + EFV group with baseline EFV-R mutations subsequently developed K65R (data not shown).
Postbaseline phenotypic data were obtained for all nine subjects with virologic failure; mean fold changes in EFV, NVP, and DLV susceptibility postbaseline were greater than 328-, greater than 245-, and greater than 181-fold reduced, respectively (data not shown). Reduced susceptibility to FTC and 3TC was observed in the seven subjects who developed the M184V/I mutations (mean fold change in susceptibility of greater than 100-fold for both FTC and 3TC). Postbaseline susceptibility to all other NRTIs in these 9 subjects remained equivalent to wild-type (below the lower clinical cutoffs).
Analysis of Baseline Reverse Transcriptase Mutations and Treatment Response
The presence of baseline NNRTI-R mutations has been implicated with reduced clinical response to EFV-containing regimens.33,34 A statistical analysis was undertaken to examine the impact of baseline resistance mutations on treatment response in Study 934. For the ITT population (for each group separately and for both groups combined), the percent virologic failure (based on the definition of confirmed 400 copies/mL or greater at Week 144 or study discontinuation) was calculated for each mutation at codons 1 to 305 of RT. The Fisher's exact test P value was calculated comparing percent virologic failure between “any change” and “no change” in the amino acid at each RT codon. Additionally, repeat comparisons were made for subjects whose HIV-1 contained any NAM, any TAM, M184V/I, L210W, T215Y/F, any NNRTI-R mutation and specifically K103N at baseline. The key findings of this analysis are shown in Table 6.
The presence of “any NNRTI-R mutations” at baseline and the specific presence of K103N at baseline was strongly associated with virologic failure in the study, both for each group individually (P < 0.05) and for both groups combined (P < 0.001) (Table 6). Among the 10 subjects who experienced virologic failure with “any NNRTI-R” at baseline, eight of 10 had the K103N RT mutation at baseline either alone or with other RT mutations (Table 6). The other two subjects in this group had NNRTI-R mutations at baseline other than K103N, including K101E, K103R, and G190A RT mutations and the V108I RT mutation, respectively. The latter subject also had D67N and T215D RT mutations at baseline, consistent with transmission of an NRTI-resistant virus, and experienced virologic failure on 3TC + ZDV + EFV. Among the remaining subjects with baseline primary NNRTI-R (n = 13), 11 of 13 were considered TLOVR failures but all had plasma HIV-1 RNA less than 400 copies/mL at the time they were actively discontinued from the study regimen; whether these additional subjects would have subsequently experienced virologic failure had they remained on their study regimen is unknown. Primary NNRTI-R mutations observed in these subjects at baseline included K103N/S (n = 7), G190A (n = 2), G190E (n = 1), and Y181C (n = 1); one subject with K103N at baseline also had multiple TAMs (D67N, K70R, and K219Q) and the M184V RT mutation. The remaining two subjects, one each per treatment group and both with K103N at baseline, continued on the study regimen and achieved and maintained less than 400 copies/mL of plasma HIV-1 RNA through Week 144.
Other substitutions observed at RT codon K103 at baseline were not significantly associated with virologic failure; K103R was observed in 12 subjects at baseline (FTC + TDF + EFV, n = 8; 3TC + ZDV + EFV, n = 4, data not shown). Two of 12 subjects with K103R at baseline experienced virologic failure; both subjects were in the FTC + TDF + EFV treatment group; however, one of these was among the previously mentioned 22 subjects with primary NNRTI-R at baseline and also had the G190A NNRTI-R mutation at baseline. The other subject developed an additional secondary NNRTI-R mutation at virologic failure (K101E) but remained phenotypically susceptible to all NNRTIs (data not shown). K103S was also observed in two subjects at baseline; however, the mutation was present as a mixture with K103N in both subjects.
There was no association with the presence of NAMs or TAMs at baseline with virologic failure in either treatment group or both groups combined. Only one subject had the M184V mutation at baseline (along with primary NNRTI-R), no subjects had T215Y/F at baseline, and only a single subject (on 3TC + ZDV + EFV) with D67N and T215D at baseline experienced virologic failure with subsequent development of K70R, M184V, and EFV-R mutations. An unexpected finding of this analysis was that changes at RT codon Q278 were associated with virologic failure in the 3TC + ZDV + EFV group specifically (P = 0.02) and for both groups combined (P = 0.01). Twenty-eight subjects (all in the mITT population) had substitutions at codon Q278 at baseline, representing a frequency of 5.7%. Q278 substitutions observed at baseline included Q278A/E/H/N/S of which the most common was Q278H (13 of 28). Among the eight subjects with Q278 substitutions at baseline who were virologic failures (3TC + ZDV + EFV, n = 6; FTC + TDF + EFV, n = 2), Q278H was also the most common (n = 4); other substitutions at Q278 observed in this group of virologic failures included Q278N (n = 2), Q278E (n = 1), and Q278S (n = 1). The presence of Q278H alone at baseline was not significantly associated with virologic failure (P > 0.05, data not shown). Changes at RT codon Q278 represent polymorphic changes in RT, which have not previously been associated with either NRTI-R or NNRTI-R (http://hivdb.stanford.edu). Adjustment for multiple comparisons was not performed in this analysis and would likely result in a lack of statistical significance for this observation with RT codon Q278.
HIV-1 Subtypes and Treatment Response
The majority of subjects (greater than 94% in each group, data not shown) enrolled in Study 934 were infected with subtype B HIV-1. The next most common subtype was AG (1.6% in each group). Five subjects (2%) entered the study with subtype C HIV-1 and all were randomized to the FTC + TDF + EFV group. None of these five subjects with subtype C HIV-1 met resistance analysis criteria through Week 144; all achieved less than 400 copies/mL of plasma HIV-1 RNA on FTC + TDF + EFV through Week 144 or by the time of study discontinuation. In this small subset of subjects with subtype C HIV-1 and treated with FTC + TDF + EFV, there was no emergence of the K65R mutation.35,36 Only one subject in the 3TC + ZDV + EFV group with a non-B subtype of HIV-1 (subtype AG) required resistance analysis and developed M184V and EFV-R mutations. Overall, there was no association between baseline HIV-1 subtype and subsequent virologic outcome in this study nor was there notable development of resistance in subjects in either group with non-B subtypes of HIV-1.
GS-01-934 (Study 934) was a clinical trial in over 500 ARV treatment-naïve subjects comparing the once-daily regimen of FTC + TDF + EFV with 3TC + ZDV + EFV in which the 3TC + ZDV FDC was dosed twice daily. Baseline genotypic analyses conducted as part of this study revealed NNRTI-R (both primary and secondary NNRTI-R) and NRTI-R-associated mutations in 4.8% and 6.2% of analyzed subjects with genotypic data, respectively. Primary NNRTI-R, predominantly the K103N mutation, was present in 4.4% (subjects with genotypic data, n = 501) or 4.3% of enrolled subjects (n = 509 enrolled subjects). This is a similar frequency of transmitted NRTI-R mutations to that observed in ARV-naïve subjects enrolled in a previous TDF study, GS-99-903 (Study 903), in which 6.4% of ARV treatment-naïve subjects had HIV-1 carrying NRTI-R mutations.16 In contrast, the baseline HIV-1 genotypes from only 0.3% of subjects enrolled in Study 903 had detectable NNRTI-R mutations. The significantly increased prevalence of baseline NNRTI-R in Study 934 most likely reflects the increased use of NNRTIs (both EFV and NVP) in the 3 years separating the initiation of the two studies (June 2000 to August 2003). The prevalence of baseline NRTI-R and NNRTI-R in Study 934 subjects is also consistent with recently published studies of the prevalence of transmitted resistance in ARV-naïve and recently infected HIV-1 subjects from 1997-2006.1,2
Transmitted drug resistance, including EFV-R, has been shown to persist in HIV-1-infected subjects and be detectable by population sequencing in the absence of ARV drug pressure for periods of several years.27 In contrast, the M184V/I mutations, which are highly prevalent in ARV treatment-experienced subjects and in HIV-1 genotypic databases,30 are rarely detected in ARV-naïve subjects2; only one of 509 subjects enrolled in Study 934 had the M184V mutation detected at baseline. The reduced RC of viruses carrying the M184V/I mutation,30 as observed in this study, may inhibit their transmission, or if transmitted, they may revert to wild-type in the absence of drug selection pressure.
Consistent with other recently published studies, the presence of genotypic EFV-R at baseline was strongly associated with treatment failure on the EFV-containing regimens used in either of the two treatment groups in Study 934.33,34 Study 934 was among the first clinical trials in ARV-naïve subjects to prospectively perform baseline genotyping of subjects enrolling in the study, although these studies were not performed in “real time” during the screening process. Twenty-two subjects with primary NNRTI-R at baseline were identified early in the course of these studies. Although the investigators for these subjects were promptly informed, nine of 22 had experienced virologic failure before being discontinued from study medications. Only two study subjects (one in each treatment group) with the K103N mutation at baseline (and phenotypic resistance to EFV) achieved less than 400 copies/mL and less than 50 copies/mL of HIV-1 RNA through Week 144. Baseline plasma viral loads and CD4+ cells in these two subjects were 4.8 and 5.12 log10 copies/mL of HIV-1 RNA and 176 and 242 cells/mm3, respectively, similar to the median viral load and CD4+ cell values for the overall study population (5.0 log10 copies/mL of HIV-1 RNA in both groups and 233 and 241 CD4+ cells/mm3 in the FTC + TDF + EFV and 3TC + ZDV + EFV groups, respectively). The factors leading to virologic suppression in these two subjects with baseline EFV-R are unknown. The 2008 DHHS guidelines (www.aidsinfo.nih.gov/ContentFiles/AdultandAdolescentGL.pdf) recommend HIV-1 resistance testing before initiation of ARV therapy in HIV-1-infected ARV-naïve subjects with either acute or chronic infection. The results of Study 934 strongly support these current treatment recommendations.
In subjects without baseline NNRTI-R (the mITT population), treatment with FTC + TDF + EFV compared with 3TC + ZDV + EFV was significantly more effective at suppressing the plasma viral load to less than 400 copies/mL of HIV-1 RNA through Week 144.5 The superior efficacy of the FTC + TDF + EFV regimen translated into fewer subjects on FTC + TDF + EFV with virologic failure; as a result, fewer subjects on FTC + TDF + EFV versus 3TC + ZDV + EFV were eligible for postbaseline resistance analyses (7.8% versus 12.8%, P = 0.075) or developed any RT resistance mutations (5.3% versus 9.1%, P = 0.12), respectively. Significantly fewer subjects on FTC + TDF + EFV compared with 3TC + ZDV + EFV developed the FTC- or 3TC-associated M184V/I resistance mutation (0.8% versus 4.1% of subjects, respectively; P = 0.021). This result is in contrast with resistance data from an earlier tenofovir DF study, Study 903, in which approximately 6% of subjects developed M184V/I after 144 weeks of treatment with either 3TC + TDF + EFV or 3TC + d4T + EFV.16 The long plasma half-lives of FTC, TDF, and EFV (mean t1/2 of 11, 18.5, and 164 hours, respectively), and in particular their consequent dosing symmetry,37 may protect against the emergence of the M184V/I mutation. In contrast, 3TC and ZDV have shorter plasma half-lives (0.5-3.0 and 5-7 hours, respectively).38,39
The plasma half-life of NRTIs is an important indicator of their potential to be dosed once-daily. NRTIs also require intracellular phosphorylation by cellular kinases to either the diphosphate (tenofovir) or to the triphosphate (all other NRTIs) to become activated DNA chain terminators. The pharmacokinetics of the intracellular phosphorylated forms of NRTIs are therefore an additional relevant parameter to consider. FTC triphosphate and TFV diphosphate have been shown to have intracellular half-lives greater than 39 and greater than 60 hours, respectively, whereas the intracellular half-lives of 3TC and ZDV triphosphates have been determined to be 15 to 16 hours and 4 to 11 hours, respectively.38,39 Among these four NRTIs, ZDV appears to have both the shortest plasma half-life and the shortest half-life of the intracellular triphosphate. Notably, seven of 10 subjects in the 3TC + ZDV + EFV group and both subjects in the FTC + TDF + EFV group who developed M184V did so early in Study 934 (by Week 48) when issues of ZDV tolerability, or regimen compliance, might be most common. Adherence in Study 934 was assessed by pill counts; however, no consistent pattern of reduced adherence was observed among subjects in the resistance analysis population from both treatment groups who developed RT resistance mutations (data not shown).
Along with a significantly lower frequency of the M184V mutation, no subject on FTC + TDF + EFV in Study 934 developed the TDF-associated K65R RT mutation through 144 weeks and no subject developed the K70E RT mutation. This is in contrast to a previous TDF study in ARV-naïve subjects, Study 903, in which eight subjects (2.7%) developed the K65R mutation through 144 weeks of treatment with 3TC + TDF + EFV; an additional two subjects in the comparator arm, 3TC + d4T + EFV, also developed K65R.16 The mean time from the last time point with plasma HIV-1 RNA less than 400 copies/mL to the visit analyzed for genotypic resistance with 400 copies/mL or greater of plasma HIV-1 RNA was similar in the two TDF-containing groups in Studies 903 and 934 (median of 16 weeks in Study 934 versus median of 12 weeks in Study 903). The time from failure to analysis was marginally longer in Study 934 and would not explain the lack of observed K65R development in Study 934 nor the lower prevalence of M184V/I compared with Study 903. As a result of the open-label design of Study 934, subjects on FTC + TDF + EFV were taking fewer pills than subjects on 3TC + TDF + EFV in Study 903, the latter being a double-blind, placebo-controlled study. A combination of lower pill burden and the pharmacokinetic symmetry of FTC + TDF + EFV may have inhibited emergence of the K65R mutation in Study 934 compared with 3TC + TDF + EFV in Study 903. The longer plasma half-life of FTC compared with 3TC, coupled with greater potency40 and efficiency of incorporation of FTC versus 3TC41 may also have inhibited emergence of K65R-containing viruses in Study 934. The analyses of PR/RT resistance development conducted in Study 934 used a population-based sequencing methodology, which represents the current standard of care in terms of HIV-1 genotypic resistance monitoring. NRTI and/or NNRTI-resistant HIV-1 present at less than 20% of the viral quasi-species in virologic failure subjects may not have been detected in these analyses.
A significant predictor of development of the K65R mutation in multivariate statistical analyses of the previous TDF study, Study 903, was the presence of low CD4+ T-cell counts at baseline, specifically less than 50 CD4+ cells/mm.3,16 In Study 903, 17% and 14% of subjects in the 3TC + TDF + EFV and 3TC + d4T + EFV treatment groups, respectively, had less than 50 CD4+ cells/mm.3 at baseline.16,42 The eight TDF-treated subjects and two d4T-treated subjects whose HIV-1 developed the K65R mutation in study 903 had a median CD4+ cell count at baseline of 24 cells/mm3 and 26 cells/mm3, respectively.16 In Study 934, 15% and 11% of subjects enrolled in the FTC + TDF + EFV and 3TC + ZDV + EFV groups, respectively, had less than 50 CD4+ cells/mm3 at baseline.3-5 Thus, a similar proportion of subjects with low CD4+ T-cell counts enrolled in both Studies 903 and 934, but K65R did not emerge in Study 934. There was no significant difference in the mean baseline CD4+ cell counts of subjects in the FTC + TDF + EFV and 3TC + ZDV + EFV groups who were included in the mITT resistance analysis population (179 +/− 143 cells/mm3, n = 19 versus 247 +/− 231 cells/mm3, n = 29, respectively). Moreover, there was no significant difference in mean baseline CD4+ cell counts between the resistance analysis populations and that of the overall population of Study 934 (data not shown). Notably, both subjects in the FTC + TDF + EFV group whose HIV-1 developed M184V had less than 50 CD4+ cell/mm3 at baseline (3 and 19 CD4 cells/mm3). In contrast, the 10 subjects in the 3TC + ZDV + EFV group whose HIV-1 developed M184V had a mean CD4+ cell count at baseline of 200 cells/mm3 (median, 232 cells/mm3; range, 10-353 cells/mm3), similar to the mean CD4+ cell count for the overall study population. The relatively few subjects on FTC + TDF + EFV requiring resistance analysis, the lack of development of K65R, and the potential differences in regimen tolerability between the two treatment groups made it difficult to establish a role for low CD4+ cell counts at baseline in resistance development in Study 934. Furthermore, no statistically significant difference with respect to mean or median baseline viral load was observed between subjects in the mITT resistance analysis population (n = 50) and those in the mITT population who were considered responders (data not shown).
In the group of 22 patients with NNRTI-R at baseline, the mean number of baseline CD4+ cells was 310 cells/mm3 (median, 256 CD4 cells/mm3; range, 15-1105 cells/mm3) and mean baseline viral load was 5.0 +/− 0.45 log10 copies/mL of HIV-1 RNA (median, 5.0 copies/mL; range, 4.04-5.88 log10 copies/mL), respectively; neither parameter was statistically different from the mITT population. Therefore, subjects in Study 934 with baseline NNRTI-R did not appear to have more advanced HIV-1 disease compared with the mITT population. None of the four subjects in this group with virologic failure on FTC + TDF + EFV developed K65R, although all developed the M184V mutation. All subjects who were virologic failures on FTC + TDF + EFV, regardless of whether they were in the ITT or mITT populations, remained phenotypically susceptible to TDF and all other NRTIs; the exceptions were those subjects who developed M184V, resulting in reduced susceptibility to both FTC and 3TC.
In two Phase 3 clinical trials investigating TDF combined with either 3TC or FTC plus EFV in ARV-naïve subjects (Studies 903 and 934, respectively), K65R has been observed to develop in less than 2% of TDF-exposed subjects overall. The K65R mutation has also continued to remain at a relatively low frequency (less than 3%) in large HIV-1 genotypic databases.30 The low frequency of development of NRTI mutations on FTC + TDF + EFV in Study 934 (two of 244 subjects enrolled [0.8%]) is similar to that observed when the two NRTIs are combined with ritonavir-boosted PIs. In the ABT-0418, BATON, and ARTEMIS studies, in which ARV-naïve subjects were treated with FTC + TDF combined with ritonavir-boosted PIs (lopinavir twice a day, atazanavir once per day, and darunavir twice a day, respectively), the M184V/I mutation emerged in less than 2% of enrolled subjects overall, and the K65R mutation was not observed to develop.43-45 The subsequent management of these subjects with virologic failure may therefore include a wide variety of treatment options.
Consistent with other studies in ARV-naïve subjects, NNRTI-R was found to be the most clinically relevant form of transmitted resistance with regard to risk of virologic failure on an EFV-containing regimen.33,34 This finding highlights the importance of baseline genotyping when initiation of ARV therapy is being considered, especially when considering an NNRTI-containing regimen. Nevertheless, it was notable that a small proportion of patients enrolled in Study 934 (approximately 2%) entered the study with evidence of genotypic and phenotypic resistance to PIs. Therefore, resistance testing is warranted for all ARV-naïve subjects regardless of the first-line regimen under consideration. An effective and simple ARV regimen tailored to the genotypic/phenotypic profile of the subject's virus will lead to rapid suppression of viral replication to undetectable levels, reducing opportunities for both resistance development and viral transmission. Active monitoring for the presence of ARV drug resistance is essential to ensure maximum treatment benefit from the potent and effective ART regimens now available, including fixed-dose combinations.
We thank Katyna Borroto-Esoda, Kathryn Kitrinos, Jenny Svarovskaia, and Kirsten White for their thoughtful contributions and revisions to this manuscript; and Margaret Benton and Susan Edl for technical assistance in the preparation of this manuscript. Finally, we extend our thanks to all the patients and investigators involved in Study GS-01-934.
1. Daar ES, Richman DD. Confronting the emergence of drug-resistant HIV type 1: impact of antiretroviral therapy on individual and population resistance. AIDS Res Hum Retroviruses. 2005;21:343-357.
2. Geretti AM. Epidemiology of antiretroviral drug resistance in drug-naive persons. Curr Opin Infect Dis. 2007;20:22-32.
3. Gallant JE, DeJesus E, Arribas JR, et al. Tenofovir DF, emtricitabine, and efavirenz vs zidovudine, lamivudine, and efavirenz for HIV. N Engl J Med 2006;354:251-260.
4. Pozniak AL, Gallant JE, DeJesus E, et al. Tenofovir disoproxil fumarate, emtricitabine, and efavirenz versus fixed-dose zidovudine/lamivudine and efavirenz in antiretroviral-naive patients: virologic, immunologic, and morphologic changes: a 96-week analysis. J Acquir Immune Defic Syndr. 2006;43:535-540.
5. Arribas JR, Pozniak AL, Gallant JE, et al. Tenofovir disoproxil fumarate, emtricitabine, and efavirenz compared with zidovudine/lamivudine and efavirenz in treatment-naive patients: 144-week analysis. J Acquir Immune Defic Syndr. 2008;47:74-78.
6. Schinazi RF, Lloyd RM Jr, Nguyen M-H, et al. Characterization of human immunodeficiency viruses resistant to oxathiolane-cytosine nucleosides. Antimicrobial Agents Chemother. 1993;37:875-881.
7. Tisdale M, Kemp SD, Parry NR, et al. Rapid in vitro selection of human immunodeficiency virus type 1 resistant to 3'-thiacytidine inhibitors due to a mutation in the YMDD region of reverse transcriptase. Proc Natl Acad Sci U S A. 1993;90:5653-5656.
8. Margot NA, Waters JM, Miller MD. In vitro human immunodeficiency virus type 1 resistance selections with combinations of tenofovir and emtricitabine or abacavir and lamivudine. Antimicrobial Agents Chemother. 2006;50:4087-4095.
9. Borroto-Esoda K, Parkin N, Miller MD. A comparison of the phenotypic susceptibility profiles of emtricitabine and lamivudine. Antiviral Chemistry and Chemotherapy. 2007;18:297-300.
10. Gu Z, Salomon H, Cherrington JM, et al. K65R mutation of human immunodeficiency virus type 1 reverse transcriptase encodes cross-resistance to 9-(2-phosphonylmethoxyethyl)adenine. Antimicrobial Agents Chemother. 1995;39:1888-1891.
11. Wainberg MA, Miller MD, Quan Y, et al. In vitro selection and characterization of HIV-1 with reduced susceptibility to PMPA. Antiviral Therapy. 1999;4:87-94.
12. Margot NA, Isaacson E, McGowan I, et al. Genotypic and phenotypic analyses of HIV-1 in antiretroviral-experienced patients treated with tenofovir DF. AIDS. 2002;16:1227-1235.
13. White KL, Margot NA, Wrin T, et al. Molecular mechanisms of resistance to human immunodeficiency virus type 1 with reverse transcriptase mutations K65R and K65R+M184V and their effects on enzyme function and viral replication capacity. Antimicrobial Agents Chemother. 2002;46:3437-3446.
14. Margot NA, Isaacson E, McGowan I, et al. Extended treatment with tenofovir disoproxil fumarate in treatment-experienced HIV-1-infected patients: genotypic, phenotypic, and rebound analyses. J Acquir Immune Defic Syndr Hum Retrovirol. 2003;33:15-21.
15. McColl DJ, Margot NA, Wulfsohn M, et al. Patterns of resistance emerging in HIV-1 from antiretroviral-experienced patients undergoing intensification therapy with tenofovir disoproxil fumarate. J Acquir Immune Defic Syndr Human Retrovirol. 2004;37:1340-1350.
16. Margot NA, Lu B, Cheng A, et al. Resistance development over 144 weeks in treatment-naive patients receiving tenofovir disoproxil fumarate or stavudine with lamivudine and efavirenz in Study 903. HIV Med. 2006;7:442-450.
17. Ly JK, Margot NA, MacArthur H, et al. The balance between NRTI discrimination and excision drives the susceptibility of HIV-1 RT mutants K65R, M184V and K65R+M184V. Antiviral Chemistry and Chemotherapy. 2007;18:307-316.
18. Delaugerre C, Roudiere L, Peytavin G, et al. Selection of a rare resistance profile in an HIV-1-infected patient exhibiting a failure to an antiretroviral regimen including tenofovir DF. J Clin Virol. 2005;32:241-244.
19. Larder BA, Darby G, Richman DD. HIV with reduced sensitivity to zidovudine (AZT) isolated during prolonged therapy. Science. 1989;243:1731-1734.
20. Kellam P, Boucher CA, Larder BA. Fifth mutation in human immunodeficiency virus type 1 reverse transcriptase contributes to the development of high-level resistance to zidovudine. Proc Natl Acad Sci U S A. 1992;89:1934-1938.
21. Harrigan PR, Kinghorn I, Bloor S, et al. Significance of amino acid variation at human immunodeficiency virus type 1 reverse transcriptase residue 210 for zidovudine susceptibility. J Virol. 1996;70:5930-5934.
22. Whitcomb JM, Parkin NT, Chappey C, et al. Broad nucleoside reverse-transcriptase inhibitor cross-resistance in human immunodeficiency virus type 1 clinical isolates. J Infect Dis. 2003;188:992-1000.
23. Miller MD, Margot N, Lu B, et al. Genotypic and phenotypic predictors of the magnitude of response to tenofovir disoproxil fumarate treatment in antiretroviral-experienced patients. J Infect Dis. 2004;189:837-846.
24. Bacheler L, Jeffrey S, Hanna G, et al. Genotypic correlates of phenotypic resistance to efavirenz in virus isolates from patients failing nonnucleoside reverse transcriptase inhibitor therapy. J Virol. 2001;75:4999-5008.
25. Antinori A, Zaccarelli M, Cingolani A, et al. Cross-resistance among nonnucleoside reverse transcriptase inhibitors limits recycling efavirenz after nevirapine failure. AIDS Res Human Retrovirus. 2002;18:835-838.
26. Casado JL, Moreno A, Hertogs K, et al. Extent and importance of cross-resistance to efavirenz after nevirapine failure. AIDS Res Human Retrovirus. 2002;18:771-775.
27. Little SJ, Frost SDW, Wong JK, et al. Persistence of transmitted drug resistance among subjects with primary human immunodeficiency virus infection. J Virol. 2008;82:5510-5518.
28. Petropoulos CJ, Parkin NT, Limoli KL, et al. A novel phenotypic drug susceptibility assay for human immunodeficiency virus type 1. Antimicrobial Agents Chemother. 2000;44:920-928.
29. Johnson VA, Brun-Vezinet F, Clotet B, et al. Update of the drug resistance mutations in HIV-1: 2004. Topics in HIV Medicine. 2004;12:119-124.
30. McColl DJ, Chappey C, Parkin NT, et al. Prevalence, genotypic associations and phenotypic characterization of K65R, L74V and other HIV-1 RT resistance mutations in a commercial database. Antiviral Therapy. 2008;13:189-197.
31. Deval J, Navarro J-M, Selmi B, et al. A loss of viral replicative capacity correlates with altered DNA polymerization kinetics by the human immunodeficiency virus reverse transcriptase bearing the K65R and L74V dideoxynucleoside resistance substitutions. J Biol Chemy. 2004;279:25489-25496.
32. Frankel FA, Invernizzi CF, Oliveira M, et al. Diminished efficiency of HIV-1 reverse transcriptase containing the K65R and M184V drug resistance mutations. AIDS. 2007;21:665-675.
33. Borroto-Esoda K, Waters JM, Bae AS, et al. Baseline genotype as a predictor of virological failure to emtricitabine or stavudine in combination with didanosine and efavirenz. AIDS Res Hum Retroviruses. 2007;23:988-995.
34. Kuritzkes DR, Lalama CM, Ribaudo HJ, et al. Preexisting resistance to nonnucleoside reverse-transcriptase inhibitors predicts virologic failure of an efavirenz-based regimen in treatment-naive HIV-1-infected subjects. J Infect Dis. 2008;197:867-870.
35. Brenner BG, Oliveira M, Doualla-Bell F, et al. HIV-1 subtype C viruses rapidly develop K65R resistance to tenofovir in cell culture. AIDS. 2006;20:F9-F13.
36. Doualla-Bell F, Avalos A, Brenner B, et al. High prevalence of the K65R mutation in human immunodeficiency virus type 1 subtype C isolates from infected patients in Botswana treated with didanosine-based regimens. Antimicrob Agents Chemother. 2006;50:4182-4185. [E-pub October 2006].
37. Mathias AA, Hinkle J, Menning M, et al. Bioequivalence of efavirenz/emtricitabine/tenofovir disoproxil fumarate single-tablet regimen. J Acquir Immune Defic Syndr. 2007;46:167-173.
38. Piliero PJ. Pharmacokinetic properties of nucleoside/nucleotide reverse transcriptase inhibitors. J Acquir Immune Defic Syndr. 2004;37(Suppl 1):S2-S12.
39. Back DJ, Burger DM, Flexner CW, et al. The pharmacology of antiretroviral nucleoside and nucleotide reverse transcriptase inhibitors: implications for once-daily dosing. J Acquir Immune Defic Syndr. 2005;39(Suppl 1):S1-23; quiz S24-S25.
40. Rousseau FS, Wakeford C, Mommeja-Marin H, et al. Prospective randomized trial of emtricitabine versus lamivudine short-term monotherapy in human immunodeficiency virus-infected patients. J Infect Dis. 2003;188:1652-1658.
41. Feng JY, Shi J, Schinazi RF, et al. Mechanistic studies show that (-)-FTC-TP is a better inhibitor of HIV-1 reverse transcriptase than 3TC-TP. FASEB J. 1999;13:1511-1517.
42. Gallant JE, Staszewski S, Pozniak AL, et al. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA. 2004;292:191-201.
43. Molina JM, Gathe J, Lim PL, et al. Comprehensive resistance testing in antiretroviral-naïve patients treated with once-daily lopinavir/ritonavir plus tenofovir DF and emtricitabine: 48-week results from Study 418 [Poster WePeB5701]. Presented at XV International AIDS Conference; July 11-16, 2004; Bangkok, Thailand.
44. Elion R, Cohen C, Ward D, et al. BATON Study Group. Evaluation of efficacy, safety, pharmacokinetics, and adherence in HIV-1-infected, antiretroviral-naïve patients treated with ritonavir-boosted atazanavir plus fixed-dose tenofovir DF/emtricitabine given once daily. HIV Clin Trials. 2008;9:213-224.
45. Ortiz R, Dejesus E, Khanlou H, et al. Efficacy and safety of once-daily darunavir/ritonavir versus lopinavir/ritonavir in treatment-naive HIV-1-infected patients at week 48. AIDS. 2008;22:1389-1397.
© 2009 Lippincott Williams & Wilkins, Inc.