Dynamics of Raltegravir Resistance Profile in an HIV Type 2-Infected Patient

Abstract

The evolutionary dynamics of RAL resistance in the HIV-2 virus were examined through population and clonal sequence analysis of the IN from baseline, during treatment, and after stopping RAL therapy. The treatment failure of an RAL regimen in the HIV-2 patient studied was associated with the emergence of mutations via the N155H resistance pathway and subsequent switching to the Y143C mutational route. This study has also identified four novel secondary mutations, Q91R, S147G, A153G, and M183I, not previously reported in HIV-1 patients failing RAL therapy. Resistant variants involving the Y143C pathway were noted to have persisted beyond 4 weeks following the cessation of RAL therapy. All resistance-associated mutations were lost at 20 weeks after stopping RAL therapy. Our findings provide evidence supporting the supposition that substantial cross-resistance between strand transfer IN-Is is likely in HIV-2 as shown in HIV-1.

T he HIV integrase (IN) is an essential enzyme to viral replication. Integrase inhibitors (IN-Is) represent a new class of antiviral drugs that prevents the replication of HIV by blocking the IN strand transfer pathway. The first series of strand transfer IN-Is are diketo acids (DKA), such as L731,988 and S-1360.^1,2 The naphthyridine carboxamide (NCA) strand transfer inhibitors are DKA derivatives, represented by L-870,810 and its follow-up compound raltegravir (RAL). Both have shown potent antiviral activity³ and additional IN-Is such as elvitegravir (EGV) are currently being investigated in clinical studies.^4,5 RAL, the first-in-class strand transfer IN-I to reach approval, is a potent IN-I against multidrug-resistant HIV-1 infection.⁶ Successful clinical outcomes of RAL-treated HIV-2 patients have also been reported.^7,8 Despite the impressive levels of virological efficacy in HIV-1 patients, resistance to RAL has been noted in both in vivo and in vitro studies.^3,9,10

In RAL-failing HIV-1 patients, the development of resistance is associated with one of two signature mutations in the IN: N155H or Q148K/R/H, although a possible third pathway involving mutations at position 143 and other positions has been described.^9

–12 The two major pathways appear to be mutually exclusive; no viruses carrying both mutations on the same viral genome having been described to date.⁹ The selection of both N155H and Q148K/R/H results in phenotypic RAL resistance, while the Q148 pathway leads to a further reduction in IN-I susceptibility. The mechanisms for selection of the different resistant pathways and their clinical significance remain unclear. Furthermore, in HIV-1, the level of resistance is dramatically enhanced when add-on secondary mutations emerged.⁹ Viruses with primary mutations N155 or Q148 show extensive cross-resistance to diverse IN-Is including EGV.^13,14

Recently, we reported the first successful treatment of an HIV-2 patient with RAL and subsequently identified the N155H mutation in association with HIV-2 viral load rebound.⁷ The aim of this study was to evaluate the extent to which RAL resistance pathways developed in this HIV-2 patient during the course of RAL combination therapy. In the study presented here, detailed population and clonal sequence analysis of the IN from both baseline and on treatment samples has allowed us to assess the dynamics of selection of RAL-resistant variants in vivo and offered an insight into the unique nature of HIV-2 RAL resistance.

The RA- treated HIV-2 patient had a baseline viral load of 55,400 copies/ml and a CD4 count of 80 cells/μl; the initial treatment history and subsequent RAL combination therapy including ABC, AZT, and DRV/r has been reported previously.⁷ Plasma samples were collected for sequence analysis at baseline and during RAL treatment (DR1 at week 20 and DR2 at week 24). Samples (SR1 and SR2), taken 4 and 20 weeks after stopping RAL therapy, were also analyzed.

The isolation of viral RNA from plasma and cDNA synthesis was performed as reported previously.¹⁵ The entire coding region (299 amino acids) of the IN gene was amplified using nested reverse transcriptase polymerase chain reaction (RT-PCR); this method has been previously described.¹⁶ To obtain PCR products for cloning, the published procedure was amended.¹⁷ Briefly, three independent nested PCRs were carried out using aliquots of cDNA products and a 2 μl sample of each PCR product was then cloned independently. Purified PCR products were ligated into the TOPO TA cloning vector (Invitrogen) and transformed into competent Top 10 Escherichia coli. A total of 80–100 individual bacterial colonies were picked and the clones carrying an IN insertion were first identified by PCR-based screening using the second round IN-specific primers. Plasmid DNA from representative clones was purified (Qiaprep Spin Miniprep kit, Qiagen) and the inserts covering the complete coding region of the IN were sequenced. Detailed population and clonal sequence analyses of the IN were performed on all samples. All population and clonal sequences determined in this study have been deposited in GenBank under accession numbers FJ441970 to FJ442185.

To gain a more detailed picture of the evolutionary dynamics of mutations in the IN, samples from baseline, during RAL treatment, and after stopping RAL therapy were subjected to both population and clonal sequencing analysis and the results were compared. The full length IN from a total of 266 molecular clones was sequenced. Forty-three of the clones were defective; these had a termination codon within the IN-coding region. The remaining 223 clonal sequences, 20–66 clones per time point, were subjected to analysis.

In the baseline samples, IN active site residues (D64, D116, and E152), the HIV-1 RAL resistance-associated amino acid positions both at primary (N155H, Q148H/R/K, and Y143R/C/H), and the secondary (92, 97, 140, 151, and 263) mutation sites were completely conserved in both population and clonal samples. In contrast, six HIV-1 secondary mutations (L74I, I203M, E138T, Q157H, G163D, and S230G) were consistent with the wild-type HIV-2 sequences by both population and clonal sequencing. No further polymorphisms were detected in more than one of 20 clonal sequences analyzed (results are not shown).

The results of clonal and population sequence analysis from samples taken during treatment and after stopping RAL are summarized in Table 1. Two distinct mutational pathways involving either N155H or Y143C were identified by population sequencing of the samples from weeks 20 and 24, respectively; the N155H mutation was accompanied by an A153G substitution, whereas Q91R and T97A evolved with an Y143C. N155H and Y143C changes were mutually exclusive and no evidence of both mutations having occurred in the same genome was seen.

Table 1.

Genotypic Changes in the IN Observed during Follow-up of RAL Therapy a

Genotype b	Weeks on RAL treatment (sample/number of total clones)		Weeks after stopping RAL treatment (sample/number of total clones)
Genotype b	20 (DR1/39) c	24 (DR2/37)	4 (SR1/66) d	20 SR2 (61)
1	N155H, A153Ge 20 ( 51%)f	Y143C, T97A, Q91R 20 (54%)	Y143C, T97A, Q91R 54 (82%)	Wild type 61 (100%)
2	N155H 2 (5%)	N155H, A153G, E92Q 8 (22%)	Y143C, T97A, Q91R, N144T/P 4 (6%)
3	N155H, E92G 10 (26%)	N155H, H157R M183I 5 (14%)
4	N155H, A153G, E92G 2 (5%)	Wild type 2 (5%)
5	N155H, A153G, S147G 4 (10%)

Both population and clonal sequence analysis.

Genotypes found only in a single clone are not listed in the table, as they could represent PCR-introduced mutations.

DR1, during RAL treatment (week 20).

SR, stopping RAL therapy (week 4).

Mutations identified in both population and clonal sequences are highlighted in bold.

The results are presented as the number of clones and the parentheses indicate the percentage of the mutant clones.

To identify minority variants with genotypes different from that present in the dominant viral population, 39 and 37 molecular clones were analyzed at weeks 20 and 24. At week 20, in addition to confirmation of the presence of the dominant resistant genotype N155H/A153G, clonal sequences revealed four further resistant variants: N155H, N155H/E92G, N155H/E92G/A153G, and N155H/A153G/S147G. N155H and E92G/A153G appeared simultaneously, whereas E92G and A153G seemed to be present on separate genomes except for two clones, which carried both mutations (Table 1).

The switch from predominantly N155H to the Y143C resistance pathway was observed at week 24 by both population and clonal sequencing. The variant containing N155H and A153G found at week 20 persisted in the clonal sequences at week 24 with a third mutation E92Q being added. It is noteworthy that clonal sequencing raised the possibility that genotype N155H/A153G/E92Q at week 24 arose in the minority viruses harboring N155H/A153G/E92G at week 20. A double-nucleotide substitution (GGA to CAA) was required for the acquisition of the G92Q substitution. Alternatively, the 92Q evolved from predominant viruses bearing N155H and A153G, with a further accumulation of E92Q achieved by point mutation (GAA to CAA). Interestingly, minority viruses (5 of 37 clones) carrying N155H, H157R, and M183I mutations were also identified at week 24. Although a single mutation N155H was carried by two clones at week 20, no clones harboring only the Y143C mutation were isolated at week 24.

Sample SR1, taken 4 weeks after RAL therapy was stopped, showed that the only RAL resistance genotype persisting was Y143C/Q91R/T97A. The fixation of the Y143C/Q91R/T97A was shown by both population sequencing and in all 66 of the clones analyzed. In addition, N144T/P was detected in 4 out of 66 clones in association with the Y143C/Q91R/T97A genotype (Table 1).

At 20 weeks post-RAL therapy, all RAL resistance mutations found in week 20 (DR1)and 24 (DR2) samples as well as those remaining in the sample SR1 taken 4 weeks after stopping RAL therapy had completely disappeared in the sample of all 61 SR2 clones analyzed. This suggested that there had been selection and outgrowth of a fitter, wild-type virus.

The development of RAL resistance in HIV-1 through the two signature pathways involving either N155H or Q148K/R/H is well documented both in vitro and in clinical studies.^3,9 This study examined the kinetics of emergence of RAL resistance in an HIV-2 patient with early VL rebound.

At baseline, both population and clonal sequences showed that the IN active site residues and the three primary mutation positions in HIV-1 (N155, Q148, and Y143) were all completely conserved. Although there was little variation seen at the 10 secondary mutation codons (74, 92, 97, 138, 140, 151, 157, 163, 203, and 263), there were some preexiting substitutions specific to the HIV-2 IN (L74I, I203M, E138T, Q157H, G163D, and S230G) when compared to HIV-1 IN of the HXB-2 reference strain. Our study further supports the assumption that the wide-ranging polymorphisms observed in HIV-2 IN rarely occurred at the key mutation sites associated with RAL resistance in HIV-1 and therefore are unlikely to have a significant effect on phenotypic susceptibility to RAL.¹⁸

Six secondary mutational positions (74, 92, 97, 151, 157, and 163) have been shown to be associated with 155 resistance pathway in HIV-1 (Table 2). The novel and unexpected finding in our study was the association of the A153G secondary substitution with the N155H primary mutation as the prevailing majority viral population in sample DR1 taken at week 20 of RAL treatment. The existence of a single clone carrying the N155H mutation alone implies that N155H may be the first resistance mutation to develop under RAL drug pressure. This interpretation would accommodate similar findings that have been reported in some RAL-treated HIV-1 patients.⁹ The continuous RAL pressure exerted on the N155H mutant virus appears to have resulted in the subsequent selection of the double mutant variants, N155H/E92G, found in some clonal sequences and the eventual dominance of the N155H/A153G genotype. The analysis of molecular clones confirms that A153G is present on the same genome as the N155H mutation, indicating a preferential association with the 155 mutational pathway.

Table 2.

IN Mutations Developed in HIV-1 and HIV-2 Patients During RAL Therapy a

N155H pathway		Q148H/K/R pathway		Y143C/H/R pathway
HIV-1	HIV-2 b	HIV-1	HIV-2	HIV-1	HIV-2
N155H	N155H	Q148H/K/R	Q148Rc	E92Q	Q91R
L74M/I	E92G/Q	L74M/I		T97A	T97A
E92Q	S147G d	E138A/K
T97A	A153G	G140A/S
V151I	H157R
E157Q	M183I
G163R/K

Mutations associated with the three resistance pathways (condons 148, 155, and 143) are included and other resistance pathways involving E92Q, T66A, and E157Q have also been described.^4,18

Mutations identified by both populational and clonal sequencing.

Q148R was reported in one HIV-2 patient failing RAL,²⁵ but no secondary mutations were identified.

Novel mutations identified in this study are highlighted in bold.

The analysis of amino acids of HIV-1 and HIV-2 IN showed a similar high degree of conservation at position 153 with 1–6% polymorphism rates being reported.^18,19 S153 and A153 are the consensus wild-type sequence for the HIV-1 and HIV-2 virus, respectively. In contrast to the in vitro selected DKA and EGV-resistant mutants carrying S153A/Y,^2,20,21 changes at codon 153 have not been reported during in vitro studies on NCA-resistant variants, nor isolated in RAL and EVG failing HIV-1 patients. The secondary mutation A153G, coupled with N155H, identified in this study provides the first clinical evidence of RAL resistance mediated by a change at codon 153. Viruses bearing N155H and A153G double mutations may demonstrate higher levels of RAL resistance than those carrying the N155H single mutant. Consistent with this view, data from Hazuda and colleagues on RAL-resistant HIV-1 showed that the N155H mutation by itself led to 10-fold resistance. Levels of resistance increased still further when additional mutations (L74M, T97A, and E92Q) were selected, with the N155H and E92Q double mutations conferring the highest resistance of up to 70-fold.⁹ Interestingly, with regard to the phenotypic effects of mutations at codon 153 on several strand transfer inhibitors, Zahm and co-authors recently demonstrated that S153Y potentiates susceptibility to L-870,810.²² In contrast, other studies have shown that the S153Y substitution may confer greater resistance to EGV than RAL and L-870,810.^20,23 Finally, it has been postulated that S153Y may play a role in indirectly interrupting the binding of strand transfer IN-Is.²⁴

In clinical trials of RAL, Y143C appeared to represent an alternate, rarely utilized pathway to RAL resistance that may be linked to the non-B subtypes of HIV-1. This mutation has frequently been reported as occurring with at least one other mutation including N155H, E92Q, and T97A.^9,10 The secondary mutation Q91R (evolved with Y143C and T97) identified in this study at week 24 has not previously been associated with both in vivo and in vitro resistance to RAL, nor any other IN-Is. Q91R, like E92Q, is located in the catalytic core domain, close to the active-site residues (D64, D116, and E152). Consequently, the possible mechanisms for RAL resistance mediated by the Q91R may involve its interaction with RAL or the viral cDNA. However, further studies are needed to determine the precise contribution of Q91R to phenotypic RAL resistance and fitness of the virus. Despite the fact that the HIV-2 virus in this patient initially developed RAL resistance at week 20 via the N155H pathway as demonstrated in HIV-1, the subsequent switch to the 143 route at week 24 was unexpected. In HIV-1 the most frequently reported resistance pathway incorporates mutations at codon 148 and the resistance to RAL was superior with Q148 mutations (especially with additional changes at G140).⁹ The two major resistance pathways (155 and 148) in HIV-1 appear to be mutually exclusive in clinical trials⁹ and their coexistence in quasispecies was described in a more recent study.¹¹ Furthermore, Q148K/R was also reported in one RAL-failing HIV-2 patient.²⁵ The mechanism involved in the HIV-2 INs switch from the 155 to the 143 pathway instead of the 148 pathway as described in this study remains to be defined. Additionally, the key secondary mutations (G140S and E138A/K) associated with the148 pathway in HIV-1 were not present in either dominant or clonal variants of HIV-2 isolated from on-treatment samples.

Both clinical data and in vitro studies point to cross-resistance among the different strand transfer IN-Is developed to date, consistent with the shared mechanism of these agents.^{3,13,14,24,26} This study has demonstrated the potential for rapid evolution of diverse RAL resistance genotypes present in both predominant and minority HIV-2 variants. On one hand, the Y143C resistance pathway (Y143C/Q91R/T97A) that was utilized by the dominant viruses at week 24 and persisted beyond 4 weeks of stopping RAL therapy has not been shown to be cross-resistant to EVG and other IN-Is. It remains possible that RAL-resistant viruses may remain fully susceptible to other strand transfer IN-Is such as EVG. On the other hand, the heterogeneous nature of resistance profiles found during RAL treatment could have a significant clinical relevance for subsequent treatment regimens with the minority variants independently following different evolutionary pathways and subsequently emerging as the prevailing population in response to a new drug pressure. In the patient studied an unusually high number of mutations (five) was observed at four positions (E92G/Q, T97A, N155H, and H157R) in the genome. These mutations have previously been associated with resistance to both RAL and EGV.^9,13 Elsewhere S147G was found in 30% of EVG failing HIV-1 patients,¹³ while S153Y was shown to generate EVG resistance and cross-resistance to RAL in HIV-1 in vitro.^13,20 The association of S147G and A153G with RAL resistance, as shown for the first time in this study, may further impact the susceptibility of HIV-2 to EVG and other IN-Is of the same class. As a secondary mutation accompanying N155H conferring RAL resistance and one of the primary mutations upon failure of EVG in a clinical trial,^9,13 E92Q has been shown to be of paramount importance for phenotypic cross-resistance among diverse strand-transfer IN-Is.^14,24 It is possible that preexisting resistant minority populations carrying E92Q found in the RAL-failing HIV-2 patient may rapidly emerge as the dominant species after they are subjected to strong selective pressure by an alternative drug of the same class such as EVG and thus a transient and incomplete response to the new IN-Is-containing regimens may be observed. This assumption is consistent with previous studies with PIs and NNRTIs that have shown that preexisting quasispecies with genotypes different from that present in the dominant viral population may contribute to virological failure following the substitution of alternative same class agents in a failing regimen.^27,28 The possible role of Q91R and M183I in cross-resistance among different strand transfer IN-Is remains unknown.

In conclusion, the ability of HIV IN to develop resistance to RAL and a wide range of strand transfer IN-Is poses a great challenge for therapies with IN-Is. Analysis of variant sequences from sequential posttreatment samples and multiple clones in this study provides additional insight into the dynamics of RAL resistance developed in HIV-2. In contrast to studies with RAL resistance in HIV-1, the results presented here suggest that the development of RAL mutations in HIV-2 could be more complex. The treatment failure of the RAL regimen in the HIV-2 patient was associated with the emergence of resistance mutations via two resistant pathways, stemming from N155H and Y143C. Four novel secondary mutations including Q91R, S147G, A153G, and M183I, all located in the catalytic core domain of the IN and not previously reported in RAL-failing HIV-1 patients, have been identified. Furthermore, this is the first time that minority viral populations expressing a variety of resistant mutations in IN, selected during RAL treatment, have been documented, although their clinical relevance to subsequent RAL failure requires further investigation. Taken together, these findings provide evidence supporting the supposition that substantial cross-resistance between IN-Is is likely in HIV-2 as shown in HIV-1. Further studies involving a larger study population are warranted to determine the influence of the HIV-2 IN backbone on the pathways leading to RAL resistance and clinical outcomes.

Footnotes

Acknowledgments

This work was supported in part by a research grant from the Investigator Initiated Studies Program of Merck. We thank the Functional Genomics Laboratory, University of Birmingham (BBSRC grant 6/JIF 13209), for DNA sequencing.

Disclosure Statement

No competing financial interests exist.

References

Billich

. S-1360 Shionogi-GlaxoSmithKline. Curr Opin Investig Drugs, 2003; 4,2:206–209.

Hazuda

, Felock

, Witmer

et al. Inhibitors of strand transfer that prevent integration and inhibit HIV-1 replication in cells. Science, 2000; 287,5453:646–650.

Hazuda

, Anthony

, Gomez

et al. A naphthyridine carboxamide provides evidence for discordant resistance between mechanistically identical inhibitors of HIV-1 integrase. Proc Natl Acad Sci USA, 2004; 101,31:11233–11238.

Sato

, Motomura

, Aramaki

et al. Novel HIV-1 integrase inhibitors derived from quinolone antibiotics. J Med Chem, 2006; 49,5:1506–1508.

DeJesus

, Berger

, Markowitz

et al. Antiviral activity, pharmacokinetics, and dose response of the HIV-1 integrase inhibitor GS-9137 (JTK-303) in treatment-naive and treatment-experienced patients. J Acquir Immune Defic Syndr, 2006; 43,1:1–5.

Grinsztejn

, Nguyen

, Katlama

et al. Safety and efficacy of the HIV-1 integrase inhibitor raltegravir (MK-0518) in treatment-experienced patients with multidrug-resistant virus: A phase II randomised controlled trial. Lancet, 2007; 369,9569:1261–1269.

Garrett

, Xu

, Smit

et al. Raltegravir treatment response in an HIV-2 infected patient: A case report. AIDS, 2008; 22,9:1091–1092.

Damond

, Lariven

, Roquebert

et al. Virological and immunological response to HAART regimen containing integrase inhibitors in HIV-2-infected patients. AIDS, 2008; 22,5:665–666.

Hazuda

, Miller

, Nguyen

et al. Resistance to the HIV-integrase inhibitor raltegravir: Analysis of protocol 005, a Phase II study in patients with triple-class resistant HIV-1 infection. Antiviral Ther, 2007; 12:S10(Abstract 8).

10.

Cooper

, Steigbigel

, Gatell

et al. Subgroup and resistance analyses of raltegravir for resistant HIV-1 infection. N Engl J Med, 2008; 359,4:355–365.

11.

Charpentier

, Karmochkine

, Laureillard

et al. Drug resistance profiles for the HIV integrase gene in patients failing raltegravir salvage therapy. HIV Med, 2008; 9,9:765–770.

12.

Malet

, Delelis

, Valantin

et al. Mutations associated with failure of raltegravir treatment affect integrase sensitivity to the inhibitor in vitro. Antimicrob Agents Chemother, 2008; 52,4:1351–1358.

13.

McColl

, Fransen

, Gupta

et al.

Resistance and cross-resistance to first generation integrase inhibitors: Insights from a phase II study of elvitegravir (GS-9137)

Antiviral Ther, 2007; 12:S11(Abstract 9).

14.

Goethals

, Clayton

, Van

et al. Resistance mutations in human immunodeficiency virus type 1 integrase selected with elvitegravir confer reduced susceptibility to a wide range of integrase inhibitors. J Virol, 2008; 82,21:10366–10374.

15.

, Hue

, Taylor

et al. Minimal variation in T-20 binding domain of different HIV-1 subtypes from antiretroviral-naive and -experienced patients. AIDS, 2002; 16,12:1684–1686.

16.

, Anderson

, Ferns

et al. Genetic diversity of integrase (IN) sequences in antiretroviral treatment-naive and treatment-experienced HIV type 2 patients. AIDS Res Hum Retroviruses, 2008; 24,7:1003–1007.

17.

Bacheler

, Anton

, Kudish

et al. Human immunodeficiency virus type 1 mutations selected in patients failing efavirenz combination therapy. Antimicrob Agents Chemother, 2000; 44,9:2475–2484.

18.

Roquebert

, Damond

, Collin

et al. HIV-2 integrase gene polymorphism and phenotypic susceptibility of HIV-2 clinical isolates to the integrase inhibitors raltegravir and elvitegravir in vitro. J Antimicrob Chemother, 2008; 62,5:914–920.

19.

Rhee

, Liu

, Kiuchi

et al. Natural variation of HIV-1 group M integrase: Implications for a new class of antiretroviral inhibitors. Retrovirology, 2008; 5:74.

20.

Jones

, Ledford

, Yu

et al.

Resistance profile of HIV-1 mutants in vitro selected by the HIV-1 integrase inhibitor, GS-9137 (JTK-303)

Program and abstracts of the 14th Conference on Retroviruses and Opportunistic Infections, Los Angeles, CaliforniaAbstract 627, 2007.

21.

Fikkert

, Hombrouck

, Van

et al. Multiple mutations in human immunodeficiency virus-1 integrase confer resistance to the clinical trial drug S-1360. AIDS, 2004; 18,15:2019–2028.

22.

Zahm

, Bera

, Pandey

et al. Mechanisms of human immunodeficiency virus type 1 concerted integration related to strand transfer inhibition and drug resistance. Antimicrob Agents Chemother, 2008; 52,9:3358–3368.

23.

Marinello

, Marchand

, Mott

et al. Comparison of raltegravir and elvitegravir on HIV-1 integrase catalytic reactions and on a series of drug-resistant integrase mutants. Biochemistry, 2008; 47,36:9345–9354.

24.

Hombrouck

, Voet

, Van

et al. Mutations in human immunodeficiency virus type 1 integrase confer resistance to the naphthyridine L-870,810 and cross-resistance to the clinical trial drug GS-9137. Antimicrob Agents Chemother, 2008; 52,6:2069–2078.

25.

Roquebert

, Blum

, Collin

et al. Selection of the Q148R integrase inhibitor resistance mutation in a failing raltegravir containing regimen. AIDS, 2008; 22,15:2045–2046.

26.

Shimura

, Kodama

, Sakagami

et al.

Broad antiretroviral activity and resistance profile of the novel human immunodeficiency virus integrase inhibitor elvitegravir (JTK-303/GS-9137)

J Virol, 2008; 82,2:764–774.

27.

Lecossier

, Shulman

, Morand-Joubert

et al. Detection of minority populations of HIV-1 expressing the K103N resistance mutation in patients failing nevirapine. J Acquir Immune Defic Syndr, 2005; 38,1:37–42.

28.

Charpentier

, Dwyer

, Mammano

et al. Role of minority populations of human immunodeficiency virus type 1 in the evolution of viral resistance to protease inhibitors. J Virol, 2004; 78,8:4234–4247.