First-generation nonnucleoside reverse transcriptase inhibitors (NNRTIs) such as efavirenz and nevirapine have a low genetic barrier to resistance that limits their efficacy. Such NNRTI-based treatment regimens frequently lead to the selection of single mutations such as K103N, V106M, Y181C, and Y188L in HIV-1 reverse transcriptase (RT), which confer resistance against this class of antiretroviral (ARV). Indiscriminate use of the first-generation NNRTIs as part of first-line treatment regimens in clinical practice has led to the development of widespread resistance among patients displaying treatment failure.1,2 Etravirine (ETR) is a new next-generation NNRTI that is active against NNRTI-resistant HIV-1. ETR is a highly flexible diarylpyrimidine compound that targets NNRTI-resistant as well as wild-type virus. ETR possess an enhanced genetic barrier to resistance than the first-generation NNRTIs because multiple mutations are required to overcome ETR susceptibility, whereas only one mutation is typically needed to confer high-level resistance to the first-generation NNRTIs. Three or more NNRTI-associated baseline mutations have been shown to negatively influence the effectiveness of treatment with ETR,3,4 whereas some studies reported that two mutations were sufficient to reduce susceptibility to ETR. Nevertheless, ETR's durable efficacy and favorable safety profile has been demonstrated in double-blind, placebo-controlled trials that involved treatment-experienced patients (phase III study with TMC125 to Demonstrate Undetectable viral load in patients Experienced with ARV Therapy [DUET] studies).5,6
The main focus of this work is to evaluate the effectiveness of ETR as a salvage therapy in a patient population that is failing ARV therapy and drug-naïve to ETR. We used the pre-existing genotyping profile of NNRTI resistance mutations in virus isolates from such a patient population to infer ETR's therapeutic potential. This study provides a comprehensive overview of the NNRTI resistance mutations in virus isolates from patients attending AIDS clinics in Brazil who are experiencing treatment failure.
All samples were genotyped from 2000 to 2008 and were retrieved from the Brazilian National Genotyping Network (RENAGENO). Genotyping assay used was ViroSeq HIV-1 genotyping system (Celera Diagnostic; Abbott Laboratories, Abbott Park, IL) using clip reaction. In total we analyzed 1632 sequences from patients failing first- and second-line highly active antiretroviral therapy regimens. Sequences were segregated by the presence of NNRTI-related mutations and overall 1018 sequences (62.4%) of our data set were selected for further analysis. Only sequences showing homogeneous subtype call within the first 235 amino acids of RT were selected for analysis. Mosaic isolates were excluded to avoid bias in subtype segregation analysis. Isolates represent different states and also a recent pattern of ARV treatment failure resulting from previous and current scheme prescriptions.
In vitro and in vivo studies identified mutational patterns conferring resistance to ETR. Studies done with selection in culture correlated high-level resistance with the presence of F227C, Y181I, or M230L mutations alone or L100I plus K103N plus Y181C mutations or the presence of two or more first-generation NNRTI resistance-associated mutations (RAMS) associated with K101P, V179D/E/F/I, Y181I/V, or G190S. On the other hand, two first-generation NNRTI RAMs plus Y181I/V or V179D/E/I/F or K101E/P or Y188L were considered intermediate resistance.7 Vingerhoets et al found a correlation between 13 RAMs and clinical response to ETR in the Phase III DUET trials, and they identified the following mutations: V90I, A98G, L100I, K101E/P, V106I, V179D/F, Y181C/I/V, and G190A/S.7 In summary, patients with one or two ETR RAMs displayed a 19% decrease in virus load response, whereas the virus load response dropped to less than 75% in patients with three or more RAMs. Therefore, the presence of one to two of these ETR RAMs was considered as partial or low-level resistance and the presence of three or more RAMs as high-level resistance.
In this work 1632 sequences presented homogeneous subtype call within the first 235 amino acids of RT. Subtype assessment were done by submitting the sequences to the REGA HIV subtyping tool available at the BioAfrica web site (www.bioafrica.net).8 Of the total samples analyzed 1257 samples, 882 isolates clustered with subtype B (77%), and 237 with subtype F1 (14.5%) and 138 with subtype C (8.4%), respectively. As expected, a high proportion of these sequences carried mutations related to resistance to NNRTIs. Actually, these isolates represent 62.4% (n = 1021) from the total samples studied, and they clustered 48.8% (n = 798), 8.7% (n = 143), and 4.9% (n = 80) with subtypes B, F, and C, respectively. The frequency of NNRTI-associated mutations was analyzed. Mutation K103N was the major substitution selected by the patients in our data set accounting for approximately 32% of samples followed by Y181C and G190A regardless of the subtype call. Interestingly, mutations V106M and E138A were more prevalent in subtype C specimens when compared with B and F counterparts (Fig. 1).
Nevertheless, a high percentage of the patients' population under treatment failure still does not have detectable isolates carrying any mutations that confer high-level cross-resistance to ETR. To better understand the association of primary mutations for first-generation NNRTI and crossresistance to ETR, the isolates carrying any NNRTI primary mutations were segregated and simultaneous presence of a different number of ETR related mutations were calculated (data not shown). It is clear that there was a high proportion of these isolates carrying mutations related to ETR resistance. However, only a low proportion of samples (10% or less) had more than two mutations related to ETR resistance and solely less than 4% showed genotypes related to a high level of resistance to ETR. These crossresistance patterns seem not to be related to subtype assignment of isolates studied.
Overall, 37.6% of isolates showed some level of genotypic compatible to crossresistance to ETR and this number breaks down to 28.2%, 5.8%, and 3.6% from subtypes B, F1, and C, respectively.
After segregating the sequences into three groups in accordance to subtype assignment within the RT region analyzed, we could address the correlation of main primary mutations to first-generation NNRTIs such as K103N, Y181 C/I, and G190A/S with genotype related to ETR resistance. It is clear that K103N is the major NNRTI-related mutation found in our cohort accounting for more than 50% of isolates followed by G190A/S and Y181C/I. There was a slight tendency of higher accumulation of G190A/S and Y181C/I mutations in isolates from subtype C when compared with B and F counterparts. In terms of ETR crossresistance, K103N was associated with a low proportion of isolates carrying high level of resistance to ETR (approximately 15%) regardless of the subtype. Conversely, G190A/S and Y181C/I substitution was highly associated with high-level resistance to ETR and the presence of these mutations in 72% and 67% of these isolates, respectively (Table 1).
NNRTIs are potent and well-tolerated ARVs and constitute major components of the first-line treatment regimens in Brazil. In this work, we evaluated the effectiveness of ETR, the first US Food and Drug Administration-approved second-generation NNRTI, among patients failing ARV regimens containing first-generation NNRTIs. ETR displays a high genetic barrier to developing drug resistance and hence the number of RAMs that may arise in response to this new compound is crucial to the effectiveness of this drug. At least three or more baseline mutations are associated with decreased response to ETR in clinical trials. Because ETR is recommended for use by highly treatment-experienced patients, it is imperative to evaluate the prevalence of these mutations in viral isolates from those patients failing first-generation NNRTI treatment regimens. There are 10 different amino acid changes, which can influence susceptibility to ETR. We found K103N to be the most prevalent NNRTI mutation in all the subtypes under study followed by G190A/S and Y181C/I. This mutation pattern reflects higher efavirenz use than nevirapine among our patients. The prevalence of 13 specific ETR RAMs in subtype B samples were V179T 2.0%, G190S 3.7%, Y181V 0.5%, V106I 6.0%, V179D 2.6%, K101P 3.0%, K101E 5.3%, Y181C 12.0%, A98G 6.9%, V90I 6.9%, Y181I 3.6%, G190A 15.0%, and L100I 6.1%. These findings slightly differ from those reported by Llibre et al in Spain9 and may reflect country-specific differences in the use of efavirenz and nevirapine. Interestingly, the V106M mutation was more prevalent in subtype C than in others, a difference that could be explained by the difference in RT sequence. Brenner et al10 demonstrated that clade C isolates from sub-Saharan countries developed the V106M mutation (GTG →ATG) more frequently than the subtype B counterpart as a result of a difference in codon use. Most subtype B isolates harbored GTA (Valine) at codon 106, whereas the GTG (Valine) polymorphism was generally present in many clade C viruses.10 Our data show that a comparable phenomenon is displayed by Brazilian subtype C, which is also related to differences in codon use at RT codon 106. Samples with low-to-intermediate resistance were much more prevalent. However, when only NNRTIs that were associated with decreased virologic response in the DUET trials were considered, the prevalence of etravirine resistance is lower than previously reported during the drug development stages. As expected, genotypic resistance to ETR was associated more with isolates carrying G190A/S and Y181C/I mutations than those harboring K103N. Indeed, only 14% of samples carrying the K103N mutation accumulated additional mutations related to intermediated level of ETR crossresistance. This number increased to 68% and 72% in samples that harbored G190A/S and Y181C/I mutations, respectively. Combinations known to cause high levels of resistance to ETR such as Y181C/I + V179F with or without F227C were absent in our samples. The concomitant presence of V179F and F227C was very rare and was only present in one subtype B isolate in combination with K103N and V106I. We used the Stanford HIV-1 drug resistance tool (http://hivdb.stanford.edu) to predict the resistance levels to all NNRTIs and found a high incidence of drug resistance that results from a combination of mutations in our database. Combinations of five RAMs to ETR: A98G-K101E-V106I-Y181C-G190S and A98G-K101E-V179D-Y181C-G190A, which confer high levels of resistance to all NNRTIs, were found in two isolates. Other genotypes that are associated with a high level of resistance to all NNRTIs such as A98G-K101E-Y181C-G190A, A98G-K101H-Y181C-G190A, and K101E-V106I-Y181C-G190S were found in four, five, and three isolates, respectively. Another prevalent mutation pattern, K101E-Y181C-G190A, that causes high level of resistance to ETR was encountered in 15 isolates. Some other combinations of three or four mutations, which encode an intermediate resistance level to ETR such as A98G-Y181C-G190A, K101E-E138A-G190A, K101H-Y181C-G190A, and A98G-V106I-Y181C-G190A, were present in seven, seven, nine, and two isolates, respectively. These findings reiterate the importance of the genotyping test before the introduction of ETR in patient populations with failing ARV regimens and/or using first-generation NNRTIs. Although we detected an intermediate and high level of primary resistance to ETR, this prevalence was low. Therefore, our analysis suggests that ETR is an effective drug to use as a salvage therapy in Brazilian patients failing previous ARV.
The authors thank Muthukumar Balasubramaniam for review of the manuscript.
1. Wensing AM, van der Vijver DA, Angarano G, et al. Prevalence of drug-resistant HIV-1 variants in untreated individuals in Europe: implications for clinical management. J Infect Dis
2. Richman DD, Morton SC, Wrin T, et al. The prevalence of antiretroviral drug resistance in the United States. Acquir Immune Defic Syndr
3. Andries K, Azijn H, Thielemans T, et al. TMC125, a novel next-generation nonnucleoside reverse transcriptase inhibitor active against nonnucleoside reverse transcriptase inhibitor-resistant human immunodeficiency virus type 1. Antimicrob Agents Chemother
4. Deeks SG. International perspectives on antiretroviral resistance. Nonnucleoside reverse transcriptase inhibitor resistance. J Acquir Immune Defic Syndr
. 2001;26(suppl 1):S25-S33.
5. Madruga JV, Cahn P, Grinsztejn B, et al. Efficacy and safety of TMC125 (etravirine) in treatment-experienced HIV-1 infected patients in DUET-1: 24-week results from a randomised, double-blind, placebocontrolled trial. Lancet
6. Lazzarin A, Campbell T, Clotet B, et al. Efficacy and safety of TMC125 (etravirine) in treatment-experienced HIV-1 infected patients in DUET-2: 24-week results from a randomised, double-blind, placebocontrolled trial. Lancet
7. Vingerhoets J, Azijn H, Tambuyzer L, et al. Activity of etravirine on different HIV type 1 subtypes: in vitro susceptibility in treatment-naive patients and week 48 pooled DUET study data. AIDS Res Hum Retroviruses
8. de Oliveira T, Deforche K, Cassol S, et al. An automated genotyping system for analysis of HIV-1 and other microbial sequences. Bioinformatics
9. Llibre JM, Santos JR, Puig T, et al. Prevalence of etravirine-associated mutations in clinical samples with resistance to nevirapine and efavirenz. J Antimicrob Chemother
10. Brenner B, Turner D, Oliveira M, et al. A V106M mutation in HIV-1 clade C viruses exposed to efavirenz confers cross-resistance to non-nucleoside reverse transcriptase inhibitors. AIDS