Recent studies have suggested that in HIV-1-infected patients, the populations of drug-resistant viruses can be quite heterogeneous and that minority variants expressing distinct genotypes, although not detectable by standard genotyping, can contribute to the evolution of drug resistance.1-6 There are several situations in which such minority populations with resistance to nonnucleoside reverse transcriptase inhibitors (NNRTI) could influence the use of this class of drugs.
First, minority populations could play a role in cross-resistance, a common problem for NNRTIs. Because of small differences in the interaction between NNRTIs and the hydrophobic pocket in the reverse transcriptase (RT) targeted by these drugs, the mutations that emerge most commonly are somewhat drug-dependent.7 Resistance to nevirapine is often associated with the appearance of mutation Y181C.8-10 This mutation, unlike the K103N mutation, does not produce high-level resistance to efavirenz.11 In view of these differences in cross-resistance, it was hoped that efavirenz could be used effectively in the treatment of at least some patients failing nevirapine. Indeed, transient responses to efavirenz were sometimes observed for patients with the Y181C mutation on standard genotyping, but these responses were short-lived because of the emergence of viruses expressing the K103N mutation.12,13 This rapid selection of K103N raises the possibility that even in cases in which viruses expressing Y181C are the dominant viral population after nevirapine failure, minority populations expressing the K103N mutation may also be present.
Second, transient selection of minority populations of resistant viral species after NNRTI treatment interruption could also predispose to later treatment failure using this class of drugs. Nevirapine has a long half-life, and after its discontinuation, prolonged subtherapeutic plasma levels can favor the emergence of resistant strains,14,15 which could compromise treatment if NNRTIs are again used. The development of nevirapine resistance has been extensively documented in African women receiving single-dose prophylaxis at the time of delivery.16-19 In patients receiving long-term suppressive therapy by means of an antiretroviral regimen that includes nevirapine, however, emergence of viral strains with nevirapine resistance after treatment interruption has never been documented.
To evaluate these possibilities, we have developed a technique based on sequence-selective polymerase chain reaction (PCR) that permits the detection of minority K103N mutant viruses, and we used this approach to evaluate the presence of minority populations expressing this mutation in patients failing nevirapine and in patients interrupting treatment with this drug.
MATERIALS AND METHODS
Patients Failing Treatment With Nevirapine
To evaluate the frequency with which minority viral populations expressing the K103N mutation could be detected in patients failing treatment with nevirapine, we evaluated patients followed at Stanford University Medical Center and meeting the following criteria: (1) age >18 years; (2) chronic HIV-1 infection; (3) the patient had been continuously receiving nevirapine for at least 3 months before evaluation; and (4) standard population-based genotyping had been performed in the setting of virologic treatment failure, and the presence of the Y181C resistance mutation but not the K103N resistance mutation was reported. Plasma samples from 6 patients failing antiretroviral treatment who had never received an NNRTI were used as controls.
Genotype was determined by standard techniques using an ABI platform (Applied Biosystems, Foster City, CA). Six overlapping sequencing reactions (3 forward primers and 3 reverse primers) were performed for each sample, such that each position was evaluated in at least 2 (but usually 3) separate reactions. Mixtures with minority populations were reported if a distinct peak whose amplitude was at least 30% of the majority sequence was observed in 2 independent reactions.
Patients Interrupting Treatment With Nevirapine
To evaluate the possibility that minority viral populations expressing the K103N mutation could emerge after the abrupt discontinuation of treatment with nevirapine, a subgroup of patients participating in the Agence Nationale de Recherches sur le SIDA (ANRS) 081 clinical trial20 was evaluated. This study was a prospective trial comparing treatment with nevirapine plus stavudine plus indinavir and lamivudine plus stavudine plus indinavir, and all patients met the following criteria: (1) ≥18 years old; (2) CD4 cell count ≥100 cells/μL; (3) plasma HIV-1 RNA level ≥5000 copies/mL; and (4) no prior exposure to stavudine, lamivudine, or any NNRTI or protease inhibitor (PI). For inclusion in the current study, patients meeting the following criteria were identified: (1) the patient had been randomized to the group receiving nevirapine plus stavudine plus indinavir, (2) the patient had abruptly stopped taking nevirapine for >7 days, and (3) samples of plasma obtained within 60 days of discontinuing nevirapine were available and had a viral load >1000 copies. Six patients from this group (7 samples) were evaluated. The patients had been taking nevirapine for 134 ± 135 (range: 35-394) days. All patients discontinued all study drugs. Viral load immediately before discontinuing treatment was <2000 copies/mL in all patients and was undetectable (<20 copies/mL) in 3 patients. The samples evaluated had been obtained 33 ± 23 (range: 7-60) days after stopping nevirapine and had a viral load of 5.35 ± 0.42 log10 copies/mL (mean ± SD).
Screening for the K103N Mutation Using Sequence-Selective Real-Time Polymerase Chain Reaction
The techniques based on sequence-selective real-time PCR used to quantify viral populations in plasma expressing specific resistance mutations have been described in detail elsewhere.1 Briefly, 1 mL of plasma was centrifuged (23,500 g for 1 hour at 4°C), and RNA was extracted from the pellet (QIAamp Viral RNA Mini Kit; Qiagen, Valencia, CA). RT-PCR was performed using oligonucleotides RTA (5′-GGCCTGAAAATCCATACAATACTC) and RTB (5′-TGATATCTAATCCCTGGTGTCTCAT). The reaction products were diluted, and real-time PCR was performed in parallel reactions using primers and probes permitting the quantification of all viral sequences or the preferential quantification of sequences containing the K103N mutation. For quantification of all viral sequences, the following primers and probe were used: F1 (5′-TTCAGAGAACTTAATAAGAGAACTCAAGA) coupled with R1 (5′-CCCACATCCAGTACTGTTACTGATTT) and probe 5′-(6-Fam)TTCTGGGAAGTTCAATTAGGAATACCACATCC (Tamra)(phosphate). To quantify sequences selectively with the K103N mutation as a result of the AAC codon change, the primer M4 (5′-CCACATCCAGTACTGTTACTGATTTG) was used in place of R1. To quantify sequences selectively with the K103N mutation as a result of the AAT codon change, the primer M6 (5′-CCACATCCAGTACTGTTACTGATTTA) was used. The percentage of mutated sequences was then calculated as follows: % mutated sequences = [(quantity of mutated sequences in the sample)/(quantity of total sequences in the sample)] × 100. Preparations of linearized plasmid DNA (pNL4-3) in which the AAC or AAT codon changes at position 103 of the RT had been introduced by site-directed mutagenesis were used as standards.
When samples containing only the wild-type K103 sequence were evaluated as described previously, the apparent presence of 0.03% ± 0.03% (n = 8) mutated sequences was detected using the system designed to detect the K103N as a result of the AAC codon change, and the apparent presence of 0.002% ± 0.003% mutated sequences was detected using the system designed to detect the K103N mutation as a result of the AAT codon change. When complementary DNA (cDNA) with AAA or AAT codon changes was tested, with the system detecting, respectively, the AAT and AAA codon changes, the apparent percentage of mutated sequences was also <0.1%. Based on these findings, a value of 0.1% (>3 SD greater than the mean of the background) was used to define the lower limit of sensitivity of these assays.
Cloning of Reverse Transcriptase Sequences With and Without the K103N Mutation
To confirm the results obtained in the initial screening of plasma for the presence of minority populations expressing the K103N mutation, we cloned a portion of the RT from selected patients, identified clones with and without the K103N mutation, and sequenced selected clones. To do so, molecular clones were evaluated using a previously described procedure.2 Briefly, aliquots of the initial RT-PCR reaction were reamplified using primers, RTC (5′-ATACAATACTCCAGTATTTGCCATAA) and RTD (5#-CTGGTGTCTCATTGTTTATACTAGGT), and the amplification products were cloned into pCR4-TOPO. To identify clones with and without the K103N mutation, individual bacterial colonies were grown in Luria-Bertani medium. After overnight culture, medium was resuspended and diluted 1:100 with water, and 10-μL aliquots were evaluated by real-time PCR as described previously. Wild-type and mutated clones were distinguished by comparing the number of cycles required to reach threshold fluorescence (Ct) for reactions performed using the non-selective and sequence-selective PCR conditions (mutant clones, ΔCt <2 cycles; wild-type clones, ΔCt >10 cycles). Plasmid DNA from representative clones was purified, and the inserts were sequenced. Sequencing confirmed the results of the screening by real-time PCR in all cases. For 2 patients, standard population-based genotyping was repeated by amplifying the RT region from plasma RNA by RT-PCR and sequencing the amplification products.
Nucleotide sequences were aligned using the CLUSTALW program. Phylogenetic trees were constructed by maximum likelihood using DNAML (DNA maximum likelihood), and bootstrap analysis was performed using SEQBOOT (Bootstrap resampling of molecular sequence data) (PHYLIP, version 3.5).
Detection of Minority K103N Mutants in Patients Selecting the Y181C Mutation After Nevirapine Failure
Sixteen patients failing a regimen containing nevirapine and whose genotype reported the presence of the Y181C mutation but the absence of the K103N mutation were evaluated for the presence of minority K103N mutants in plasma. In the initial screening of cDNA prepared from plasma viral RNAs, sequences expressing the K103N mutation that represented >1% of total sequences were detected in 4 of 16 patients using sequence-selective real-time PCR but in none of 6 control samples (Table 1).
To confirm these findings, a 257-base pair fragment of the pol gene (spanning codons 64-140 of the RT coding sequence) was amplified from the RT-PCR product used in the initial screening by nested PCR. These amplification products were cloned, and the colonies were screened by sequence-selective real-time PCR for the presence of the K103N mutation. Representative clones were then sequenced, and phylogenetic analysis was performed. Clones containing the K103N mutation were detected for all 4 patients identified in the initial screening, and sequencing of selected clones confirmed the results of sequence-selective PCR in all cases. Phylogenetic analysis indicated that clones with and without the K103N from a given patient clustered together on branches of the phylogenetic tree that were separated from those of other patients by high bootstrap scores, as would be expected for viruses derived from the same patient (Fig. 1).
In 3 patients, the K103N mutation resulted from an AAA→AAC codon change, and in 1 patient, the K103N mutation resulted from an AAA→AAT codon change. Clones with the K103N mutation represented 1%, 5%, 33%, and 76% of total clones (see Table 1). For these patients, the sequences with and without the K103N mutation usually contained the same amino acid polymorphisms and were phylogenetically closely related to each other and to the majority population established by the original genotype (see Fig. 1). Of note, mutations at position 101 (1 patient each with K101E and K101R) were only found in clones without the K103N mutation (data not shown).
Comparison of the Results Obtained by Population Genotyping and the Evaluation of Clones
As indicated previously, a relatively high percentage of clones expressing K103N was detected in 2 patients (76% and 33%), which are proportions that might be expected to be detected by standard genotyping. To evaluate this question further, we repeated the genotyping for these patients and re-examined the original chromatograms. For patient 16 (76% clones with K103N), the repeat genotype and all 3 original chromatograms demonstrated a mixed population at position 103, with peaks corresponding to AAA (K103) and AAC (K103N) of approximately equal intensity. Thus, this result should have been reported on the original interpretation and represents a laboratory error. For patient 11 (33% clones with K103N), repeat genotyping revealed a small but distinct peak corresponding to AAC (K103N) codon use at position 103, which represented approximately 30% of the total signal by planimetry. In retrospect, minority AAC bands were also seen on the original chromatograms, but the peak height was insufficient to be read as minority populations according to the software program used in the clinical laboratory at the time of analysis. The newer version of the software program readily detected the mixture. For patients 6 and 10 (1% and 5% clones with K103N), no evidence of minority populations expressing K103N was seen on the original chromatograms.
Clinical Follow-Up on Patients Failing Nevirapine
Two of the 4 patients with populations expressing K103N subsequently received efavirenz, including patient 16 (76% clones with K103N). Both patients failed treatment, and the emergence of resistant viruses with K103N was identified by standard genotyping. For both patients, codon use at position 103 was the same in the clones obtained at the time of nevirapine failure and in the genotype obtained after efavirenz failure. For patient 6, the genotype of the majority population with K103N observed after efavirenz failure was identical to that of the 1% minority population with K103N detected earlier, except that an additional resistance mutation (L74V) was also present.
Five patients failing nevirapine for whom minority populations with K103N were not detected received a salvage regimen that included efavirenz (see Table 1). One patient responded to treatment. The remaining patients failed the new regimen, and new NNRTI resistance mutations emerged in all 4, including the K103N mutation in 2 patients.
Detection of Minority K103N Mutants in Patients Interrupting Nevirapine Treatment
We also evaluated plasma from a separate group of patients who had begun treatment with nevirapine (but then discontinued treatment) for the possible emergence of the K103N mutation in samples obtained 7 to 60 days subsequent to stopping the drug. Viral cDNA sequences expressing the K103N mutation were detected in a single sample obtained 26 days after stopping treatment. In this patient, the K103N mutation (AAA→AAC codon change) was detected in 18 (27%) of 67 clones from this sample. No other mutations associated with NNRTI resistance were identified in the clones sequenced. The K103N mutation was not detectable in plasma obtained before initiating treatment with nevirapine or in a sample obtained 1 week before stopping treatment (viral load of 1380 copies/mL), suggesting that the mutation had emerged after treatment interruption.
This study was based on the hypothesis that failure of efavirenz-based salvage regimens after the development of HIV-1 resistance to nevirapine could be related to the presence of minority viral species with cross-resistance to efavirenz in the viral population selected by nevirapine. Our results reveal that minority populations expressing K103N can be detected in patients failing nevirapine, despite the absence of this mutation on standard genotyping. For 1 patient evaluated here, subsequent failure of a regimen containing efavirenz was associated with the emergence of a viral population with K103N whose RT genotype closely resembled that of the 1% minority species detected while the patient was receiving nevirapine, further emphasizing the potential importance of such minority populations in clinical practice. Recently, Mellors et al21 have also reported detecting minority variants expressing K103N in 2 of 10 nevirapine-experienced patients by single genome sequencing. These findings are consistent with those of our recent study showing that preexisting minority populations with genotypes different from that present in the dominant viral population can contribute to virologic failure after the replacement of a PI in a failing regimen by an alternative agent in this class.2
Although this study supports the idea that the selection of minority populations with distinct resistance genotypes can contribute to the treatment failure regularly observed in this situation, our observations cast some doubt on the idea that the characterization of such populations using a technique that detects these species at the 1% level is likely to be useful in predicting treatment outcome. First, viral populations with K103N were identified in only 4 of 16 nevirapine-experienced patients, and in 2 of these cases, re-examination of the original genotyping chromatograms indicated that subpopulations with K103N had been missed because of laboratory error or because stringent criteria were used for the identification of minority populations. Importantly, 5 of the 12 patients for whom minority populations expressing K103N were not detected at the 1% level in our study were subsequently treated with efavirenz, and in 2 of these patients, virologic failure of that salvage regimen was associated with the emergence of the K103N mutation. Although we cannot exclude the possibility that the K103N mutation appeared de novo in these cases after treatment with efavirenz, it is also possible that minority populations expressing K103N, possibly associated with other mutations providing resistance to efavirenz, had been selected in these patients but were present at levels below the threshold obtained in this study. It is also noteworthy that for several patients treated with efavirenz, virologic failure was accompanied by the emergence of resistance mutations other than K103N, suggesting that comprehensive screening for mutations associated with NNRTI resistance would probably be required if detection of minority variants were to be considered for use in clinical practice.21 Taken together, these observations suggest that even if minority populations present at the 1% level were comprehensively characterized, the diversity of minority species able to contribute to subsequent failure might still be underestimated. It should be emphasized, however, that a relatively small number of patients were evaluated in our study. Only prospective studies involving a large group of subjects are likely to be able to define the benefit, if any, of screening for minority populations in identifying NNRTI-experienced patients for whom NNRTIs can be effectively used in salvage regimens.
The development of nevirapine resistance has been extensively documented in African women receiving single-dose nevirapine at the time of delivery for the purpose of lowering the risk of mother-to-child HIV transmission.16-19 This is thought to result from the long half-life of nevirapine, which produces prolonged subtherapeutic plasma levels, providing selective pressure for the emergence of resistant strains.14-17 In these usually antiretroviral treatment-naive women, emergence of resistant strains after a single dose of nevirapine is correlated with viral load.15 Thus, patients on long-term antiretroviral therapy with nevirapine in whom viremia is low or undetectable may be at lower risk for this occurrence if treatment is discontinued (eg, for reasons of toxicity or during structured treatment interruptions). Nevertheless, we identified 1 patient with a low but detectable viral load at the time therapy was discontinued in whom minority populations expressing K103N seemed be selected in this setting. Further studies in which plasma samples are obtained more frequently after the discontinuation of treatment would be required to document the frequency of this occurrence. Nevertheless, our results underscore the advisability of providing alternative antiretroviral therapy for patients discontinuing nevirapine to prevent the emergence of such resistant clones and the seeding of the latent viral reservoir with these resistant viruses.
In conclusion, this study documents that minority viral populations expressing the K103N mutation in RT can be selected in patients failing nevirapine and, although not detected by standard genotyping, may compromise the use of efavirenz in a salvage regimen. These findings add further evidence supporting the idea that minority populations can create important obstacles to the accurate assessment of viral resistance. The development of strategies for the evaluation and treatment of patients with drug resistance that adequately take into consideration this additional level of complexity might produce clinical benefit. Because populations representing <1% of total plasma virus may be important, however, optimal approaches to this task remain to be defined.
The authors thank the investigators participating in the Trianon study and especially Odile Launay and Jean-Pierre Aboulker for their help in obtaining specimens used in this study.
1. Hance AJ, Lemiale V, Izopet J, et al. Changes in human immunodeficiency virus type 1 populations after treatment interruption in patients failing antiretroviral therapy. J Virol.
2. Charpentier C, Dwyer DE, Mammano F, et al. Role of minority populations
of HIV-1 in the evolution of viral resistance to protease inhibitors. J Virol.
3. Clavel F, Hance AJ. HIV drug resistance
. N Engl J Med.
4. Coffin JM. HIV population dynamics in vivo: implications for genetic variation, pathogenesis, and therapy. Science.
5. Deeks SG. Treatment of antiretroviral-drug-resistant HIV-1 infection. Lancet.
6. Metzner KJ, Bonhoeffer S, Fischer M, et al. Emergence of minor populations of human immunodeficiency virus type 1 carrying the M184V and L90M mutations in subjects undergoing structured treatment interruptions. J Infect Dis.
7. Boyer PL, Currens MJ, McMahon JB, et al. Analysis of nonnucleoside drug-resistant variants of human immunodeficiency virus type 1 reverse transcriptase. J Virol.
8. Havlir DV, Eastman S, Gamst A, et al. Nevirapine-resistant human immunodeficiency virus: kinetics of replication and estimated prevalence in untreated patients. J Virol.
9. Richman DD, Havlir D, Corbeil J, et al. Nevirapine resistance mutations of human immunodeficiency virus type 1 selected during therapy. J Virol.
10. Torti C, Pozniak A, Nelson M, et al. Distribution of K103N and/or Y181C HIV-1 mutations by exposure to zidovudine and non-nucleoside reverse transcriptase inhibitors. J Antimicrob Chemother.
11. Bacheler LT, Anton ED, Kudish P, et al. Human immunodeficiency virus type 1 mutations selected in patients failing efavirenz combination therapy. Antimicrob Agents Chemother.
12. Antinori A, Zaccarelli M, Cingolani A, et al. Cross-resistance among nonnucleoside reverse transcriptase inhibitors limits recycling efavirenz after nevirapine failure. AIDS Res Hum Retroviruses.
13. Shulman NS, Zolopa AR, Passaro DJ, et al. Efavirenz- and adefovir dipivoxil-based salvage therapy in highly treatment-experienced patients: clinical and genotypic predictors of virologic response. J Acquir Immune Defic Syndr.
14. Mirochnick M, Fenton T, Gagnier P, et al. Pharmacokinetics of nevirapine in human immunodeficiency virus type 1-infected pregnant women and their neonates. Pediatric AIDS Clinical Trials Group Protocol 250 Team. J Infect Dis.
15. Mirochnick M, Siminski S, Fenton T, et al. Nevirapine pharmacokinetics in pregnant women and in their infants after in utero exposure. Pediatr Infect Dis J.
16. Jackson JB, Becker-Pergola G, Guay LA, et al. Identification of the K103N resistance mutation in Ugandan women receiving nevirapine to prevent HIV-1 vertical transmission. AIDS.
17. Eshleman SH, Mracna M, Guay LA, et al. Selection and fading of resistance mutations in women and infants receiving nevirapine to prevent HIV-1 vertical transmission (HIVNET 012). AIDS.
18. Eshleman SH, Jones D, Guay L, et al. HIV-1 variants with diverse nevirapine resistance mutations emerge rapidly after single-dose nevirapine: HIVNET 012 [abstract 79]. Presented at: XII International HIV Drug Resistance
Workshop; 2003; Los Cabos, Mexico.
19. Kantor R, Lee E, Johnston E, et al. Rapid flux in non-nucleoside reverse transcriptase inhibitor resistance mutations among subtype C HIV-1-infected women after single dose nevirapine [abstract 78]. Presented at: XII International HIV Drug Resistance
Workshop; 2003; Los Cabos, Mexico.
20. Launay O, Gerard L, Morand-Joubert L, et al. Nevirapine or lamivudine plus stavudine and indinavir: examples of 2-class versus 3-class regimens for the treatment of human immunodeficiency virus type 1. Clin Infect Dis.
21. Mellors J, Palmer S, Nissley D, et al. Low frequency non-nucleoside reverse transcriptase inhibitor (NNRTI)-resistant variants contribute to failure of efavirenz-containing regimens in NNRTI-experienced patients with negative standard genotypes for NNRTI mutations [abstract 134]. Presented at: XII International HIV Drug Resistance
Workshop; 2003; Los Cabos, Mexico.
22. Lecossier D, Shulman N, Zolopa AR, et al. Resistance genotypes in patients failing nevirapine: co-existence of majority viral populations expressing Y181C and minority populations
expressing K103N [abstract 143]. Presented at: XII International HIV Drug Resistance
Workshop; 2003; Los Cabos, Mexico.