SEARCH

SEARCH BY CITATION

Keywords:

  • CCR5;
  • coreceptor;
  • CXCR4;
  • genotypic testing;
  • HIV-1;
  • Thailand;
  • tropism

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Objectives

Chemokine (C-C motif) receptor 5 (CCR5) inhibitors are a novel class of antiretroviral agents that are promising for treatment of patients who harbour the HIV-1 R5 strain. Data on coreceptor tropism in non-B HIV-1 subtypes are limited. We studied coreceptor tropism in HIV-1 circulating in Thailand, where CRF01_AE predominates, using a genotypic assay.

Methods

We compiled V3 sequences of HIV-1 strains circulating in Thailand during 2010–2012. Coreceptor tropism was predicted based on V3 sequences using geno2pheno version 2.5 (http://coreceptor.bioinf.mpi-inf.mpg.de).

Results

One hundred and fifty-five HIV-1-infected patients were enrolled in this study. Ninety-nine patients (63.9%) were antiretroviral-naïve, and the remainder had virological failure. The median (interquartile range) CD4 cell count and HIV-1 RNA were 220 (74–379) cells/μL and 75 374 (14 127–226 686) HIV-1 RNA copies/mL, respectively. Of the sequences obtained from these patients, 119 (76.8%) were CRF01_AE and 22 (14.2%) were subtype B. At a false positive rate of < 5%, 61 (39.4%) HIV-1-infected individuals were predicted to harbour the X4 phenotype. X4 viruses were detected more frequently in the treatment-failure group compared with the treatment-naïve group (30.3 vs. 55.4%, respectively; P = 0.002). Those with CRF01_AE had a higher proportion of X4 viruses compared with non-AE subtypes (47.9 vs. 11.1%, respectively; P < 0.001). By multivariate logistic regression, CRF01_AE and treatment failure were independently associated with predicted X4 phenotype [odds ratio (OR) 7.93; 95% confidence interval (CI) 2.57–24.50; P < 0.001, and OR 3.10; 95% CI 1.50–6.42; P = 0.002, respectively].

Conclusions

CRF01_AE and treatment failure are associated with the predicted X4 phenotype. In regions where CRF01_AE predominates, use of CCR5 inhibitors must be considered with caution. The phenotypic assay and its correlation with genotypes should be further investigated in CRF01_AE.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

HIV-1 infection requires entry into the target cells as a first step to establish infection. The chemokine receptors chemokine (C-C motif) receptor 5 (CCR5) and chemokine (C-X-C motif) receptor 4 (CXCR4) are usually the main coreceptors used by HIV-1, together with CD4, to enter target cells [1]. HIV-1 isolates that use CCR5 and CXCR4 are called R5 and X4 strains, respectively. Dual tropic R5X4 strains can use both CCR5 and CXCR4 as coreceptors. R5 strains play an important role in transmission and can be detected in the vast majority of acute and recent infections [2]. X4 strains emerge later and are associated with accelerated disease progression [3].

CCR5 inhibitors are a novel class of antiretroviral agents that are promising for treatment of patients who harbour the R5 strain of HIV-1. Maraviroc (MVC) is the first antiretroviral agent in this group that has been approved for use in both treatment-naïve and treatment-experienced patients in whom only CCR5-tropic virus is detected. The high central nervous system penetration of MVC [4] and R5-dominant neurological strain [5] makes MVC appealing as an intensification regimen in HIV-associated neurocognitive disorder. Also, the safety and tolerability of the drug [6] make it attractive for use as a switch drug in nucleoside reverse transcriptase inhibitor (NRTI)-sparing regimens. Salvage therapy with MVC in combination with other antiretroviral agents in patients who had multi-drug-class resistance demonstrated sustained reductions in plasma viral load [7].

Coreceptor tropism testing should be performed prior to initiation of therapy with CCR5 inhibitors. HIV-1 tropism can be assessed with either phenotypic or genotypic methods. The Trofile assay (Monogram Biosciences, South San Francisco, CA) generates pseudovirions from the patient-derived envelope (env) gene [8]. These virions are used to infect human cell lines expressing either CCR5 or CXCR4. The enhanced sensitivity Trofile assay (ESTA) can detect X4 clones with 100% sensitivity when these clones make up 0.3% of the population. However, its use is limited by its restricted availability and long turnaround time. Genotypic testing of coreceptor usage is based on sequencing of the V3-encoding region of the HIV-1 env gene [9]. Various algorithms and programs are available to predict coreceptor usage from the sequence. Their accessibility, low cost and rapid turnaround time make genotypic assays a more feasible alternative to phenotypic assays [9] .

The majority of HIV-1 strains circulating in Thailand are CRF01_AE, followed by subtype B [10, 11]. However, the prevalences of R5 and X4 strains are not known. Data on coreceptor usage of HIV-1 CRF01_AE are also limited. Characterization of tropism of HIV-1 circulating in Thailand is crucial for formulating treatment regimens, especially CCR5 inhibitor-containing regimens, for treatment of HIV-1 CRF01_AE infection. We studied CCR5 inhibitor susceptibility in HIV-1 circulating in Thailand using a genotypic assay.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study population

We conducted a cross-sectional study of HIV-1 tropism in HIV-1-infected individuals in Thailand. Plasma samples from treatment-naïve and treatment-experienced HIV-1-infected individuals, age ≥ 15 years with viral HIV RNA levels of > 1000 HIV-1 RNA copies/ml, were consecutively collected at a medical school hospital in Bangkok, Thailand from April 2010 to August 2012. Patients whose history of antiretroviral exposure could not be determined were excluded from the study. Subjects in the treatment-naïve group had never been exposed to any antiretroviral agents prior to specimen collection. Patients in the treatment-failure group had evidence of genotypic resistance to nonnucleoside reverse transcriptase inhibitor (NNRTI)-based or protease inhibitor (PI)-based regimens corresponding to the current antiretroviral regimen used. Documented resistance in reverse transcriptase (RT) and protease (PR) regions was obtained in the treatment-failure group during routine resistance monitoring using TRUGENE™ (Siemens, Erlangen, Germany). The minimum duration of receipt of antiretroviral therapy (ART) was 6 months. All patients were initiated on ART according to the 2008 Thai national guidelines [12]. None of the patients were MVC exposed. The study was approved by the Institutional Review Board of Ramathibodi Hospital.

Genotypic study and data analysis

Ethylenediaminetetraacetic acid (EDTA) plasma samples were used for RNA extraction using the NucliSENS easyMAG platform (bioMérieux, Marcy l'Etoile, France) according to the manufacturer's protocol. The extracted RNA was then subjected to a two-step reverse transcriptase−polymerase chain reaction (RT-PCR) with two sets of primers: forward primer, 5′ GAG CCA ATT CCC ATA CAT TAT TGT and reverse primer, 5′ GCC CAT AGT GCT TCC TGC TGC TCC CAA GAA CC for RT-PCR and forward primer, 5′ TGT GCC CCA GCT GGT TTT GCG AT and reverse primer, 5′ TAT AAT TCA CTT CTC CAA TTG TCC for nested PCR. The sequencing primers were 5′ AAT GTC AGY ACA GTA CAA TGT ACA C (forward) and 5′ GAA AAA TTC CCT TCC ACA ATT AAA (reverse). The sequencing reaction was performed in triplicate using Big-Dye® Terminator v1.1 with the ABI PRISM 3100 genetic analyser (Applied Biosystems, Foster City, CA). The V3 sequences were analysed using SeqScape® v2.7 (Applied Biosystems).

Viral tropism was predicted using the genotypic tool geno2pheno version 2.5 [13] (http://coreceptor.bioinf.mpi-inf.mpg.de). The lowest false positive rate (FPR) for the replicates was used to determine the overall tropism of the patient's virus. Based on German guidelines, tropisms were divided into three groups according to the FPR (the likelihood of incorrectly identifying an R5 virus as X4). The cut-offs were set at 5 and 15%. An FPR of ≤ 5% predicts the virus to be X4, and thus CCR5 inhibitors are not likely to be effective. An FPR of > 15% predicts the virus to be R5, and hence CCR5 inhibitors are likely to be effective. An FPR between 5 and 15% predicts that the drug may be useful.

Virus subtypes were determined based on V3 nucleotide sequences using the geno2pheno webpage. Subtype analysis based on RT was performed using the comet tool (available at http://comet.retrovirology.lu). The V3 consensus sequence and characteristics of each subtype or circulating recombinant form (CRF) were obtained using WebLogo3 (available at http://weblogo.threeplusone.com/create.cgi).

Statistical analysis

Mean and standard deviation (SD), median and interquartile range (IQR), and frequencies (%) were used to describe patients' characteristics. The χ2 test and independent t-test or Mann−Whitney U-test were used to compare categorical variables and continuous variables between the two groups, respectively. Univariate logistic regression was used to determine the factors associated with reduced CCR5 inhibitor susceptibility. Variables that presented P < 0.10 were considered in a multivariate logistic regression model after assessment of multicollinearity of variance inflation factors. The selected variables were included in a multiple logistic regression model with forward selection and those that attained significance (P < 0.05) were retained in the model. The odds ratio (OR) and its 95% confidence interval (CI) were estimated. A P-value < 0.05 was considered statistically significant. All statistical analyses were performed using the spss statistical software version 18 (SPSS Inc., Chicago, IL).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

For this study, 155 HIV-1-infected individuals were recruited. Sixty-six (42.6%) of the individuals were female, and their mean (SD) age was 37.1 ± 11.2 years. The median (IQR) CD4 cell count and HIV-1 RNA viral load were 220 (74–379) cells/μL and 75 374 (14 127–226 686) HIV-1 RNA copies/mL, respectively. Of the sequences obtained from these patients, 119 (76.8%) were CRF01_AE, 22 (14.2%) were subtype B, and 14 (9.0%) were subtype A. Ninety-nine patients (63.9%) were antiretroviral-naïve, and the remainder had experienced virological failure of NNRTI- or PI-based antiretroviral regimens. Age, HIV-1 subtype and CD4 cell count were not different between the treatment-naïve and treatment-failure groups. However, gender and viral load and were different between the groups. The characteristics of individuals recruited for this study are shown in Table 1.

Table 1. Characteristics of HIV-1-infected patients
CharacteristicsTreatment naïveTreatment failure
CD4 ≥ 350 cells/μL (n = 31)CD4 < 350 cells/μL (n = 68)Before ART initiation (n = 56)At tropism testing (n = 56)
  1. ART, antiretroviral therapy; IQR, interquartile range.

Male gender [n (%)]19 (61.3)45 (66.2)25 (44.6)25 (44.6)
Age (years) [median (IQR)]34.5 (29.0–44.6)36.6 (29.7–45.4)36.6 (30.1–44.6)36.9 (30.8–44.5)
CD4 count (cells/μL) [median (IQR)]491 (443–648)107 (41–252)155 (78–301)188 (69–297)
HIV-1 RNA (copies/ml) [median (IQR)]19,580 (7,748–116,950)162,494 (64,566–314,303)43,200 (11,773–135,208)29,050 (7,326–109,621)
Subtype [n (%)]    
A3 (9.7)9 (13.2) 2 (3.6)
B4 (12.9)8 (11.8) 10 (17.9)
CRF01_AE24 (77.4)51 (75.0) 44 (78.6)

Greater than 80% agreement between subtyping based on the V3 loop and subtyping based on RT was found. The sequence characteristics of the HIV-1 V3 loop of each subtype in this study were examined (Fig. 1). Comparing CRF01_AE with subtype B, there were amino acid differences in the consensus sequence at the following positions: 5, N in subtype B but S in CRF01_AE; 10, K in subtype B but T in CRF01_AE; 13, H in subtype B but T in CRF01_AE; 18, R in subtype B but Q in CRF01_AE; 19, A in subtype B but V in CRF01_AE; 22, A in subtype B but R in CRF01_AE; 32, Q in subtype B but K in CRF01_AE; and 34, H in subtype B but Y in CRF01_AE. A basic amino acid at position 11 and/or 25 is associated with the X4 phenotype [14]. Of note, 10.1 and 1.3% of CRF01_AE sequences harboured R at position 11 and K at position 25, respectively. In contrast, only 2.8% of subtype A sequences and none of the subtype B sequences were positive for the predicted X4 phenotype using this ‘11/25’ criterion.

figure

Figure 1. Sequence logos of V3 residue sequences in this study. The size of each logo represents the proportion of patients with the amino acid at a specific site.

Download figure to PowerPoint

The lowest FPR for each patient was calculated (Fig. 2). Comparing the treatment-naïve and treatment-failure groups, the FPR was significantly lower in the treatment-failure group (P = 0.018). Stratified by subtype, the FPR was significantly lower in CRF01_AE than in subtypes A and B (P = 0.002). The prediction of coreceptor tropism is shown in Figure 3. CCR5 inhibitors were predicted to be effective in only 25.0% of the treatment-failure group and 44.4% of the treatment-naïve group (P = 0.007). There was no difference in predicted susceptibility between the higher and lower CD4 cell count subgroups in treatment-naïve patients (Fig. 3a). Stratifying by subtype (Fig. 3b), CRF01_AE showed significantly reduced susceptibility compared with subtypes A and B (P = 0.001). There was no difference in the distribution of CD4 cell count between CRF01_AE and other subtypes (P = 0.895; Mann−Whitney U-test). Within CRF01_AE (Fig. 3c), the treatment-failure group also showed reduced susceptibility to CCR5 inhibitors compared with the ART-naïve group (P = 0.018).

figure

Figure 2. The lowest false positive rate (FPR) for each patient calculated by geno2pheno stratified by (a) ART naïve or experienced and (b) subtype.

Download figure to PowerPoint

figure

Figure 3. Percentage of HIV-1-infected patients with predicted chemokine (C-C motif) receptor 5 (CCR5) inhibitor susceptibility based on a genotypic assay performed using geno2pheno stratified by (a) ART naïve or experienced, (b) subtype, and (c) ART naïve or experienced within those infected with CRF01_AE. The cut-offs were set at false positive rates (FPRs) of 5 and 15% based on German guidelines.

Download figure to PowerPoint

At an FPR of 5%, 61 (39.4%) HIV-1 infected individuals were predicted to harbour the X4 phenotype. There was no change in tropism prediction resulted from different FPR values from triplicate testing. X4 viruses were detected more frequently in the treatment-failure group compared with the treatment-naïve group (55.4 vs. 30.3%, respectively; P = 0.002). Those with CRF01_AE had a higher proportion of X4 viruses compared with non-AE subtypes (47.9 vs. 11.1%, respectively; P < 0.001). By univariate logistic regression (Table 2), CRF01_AE, treatment failure and gender were associated with X4 phenotype. However, multiple logistic regression analysis (Table 3) showed that only CRF01_AE and treatment failure were independently associated with reduced susceptibility to CCR5 inhibitors (OR 7.93; 95% CI 2.57–24.50; P < 0.001, and OR 3.10; 95% CI 1.50–6.42; P = 0.002, respectively).

Table 2. Univariate analysis of factors associated with predicted X4 phenotype
FactorOR95% CIP value
  1. CI, confidence interval; OR, odds ratio.

CRF01_AE7.362.45–22.09< 0.001
Treatment failure2.851.45–5.630.002
Sex0.570.30–1.100.096
Age, per 5 years1.040.90–1.200.616
CD4 count, per 100 cells/μL0.950.81–1.100.452
Viral load, per 10 000 copies/ml1.000.99–1.000.531
Table 3. Multivariate analysis of factors associated with predicted X4 phenotype
FactorOR95% CIP value
  1. CI, confidence interval; OR, odds ratio

CRF01_AE7.932.57–24.50< 0.001
Treatment failure3.101.50–6.420.002

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We estimated HIV-1 coreceptor tropisms of plasma sequences of HIV-1 circulating in Thailand using a genotypic interpretation tool. At the FPR cutoff set to 5% by geno2pheno tool, we revealed that 39.4% of viruses harboured the predicted X4 phenotype. X4 viruses were detected more frequently in the treatment-failure group compared with the treatment-naïve group. A limited data set from a viral phenotypic study in Thailand using MT-2 cells also demonstrated a high percentage of syncytium-inducing viruses, which use CXCR4 for cell entry [15].

Previous studies in Western countries, where HIV-1 subtype B predominates, reported that 80–90% of untreated HIV-1 infected patients [16, 17], and 50–60% of those exposed to ART [18, 19] harboured R5 strains. These studies also showed a strong association between CD4 count at baseline and detection of X4 virus. This association was not found in our study or in another previous study [20]. This discrepancy may be explained by the subtype-specific characteristic that CRF01_AE virus prefers CXCR4 coreceptor usage. Data on coreceptor tropism of non-B subtypes are limited. A study in ART-naïve Ugandan women showed that only 64% of those infected with HIV-1 subtype D harboured R5 strains compared with 100% of those infected with subtype A [21] . Another study showed a lower frequency of X4 strains in those infected with subtype C [22]. Our study revealed a higher proportion of X4 viruses in those infected with CRF01_AE compared with subtypes A and B. Multivariate analysis showed that CRF01_AE and treatment failure were the significant predictors of the X4 phenotype. X4 viruses may emerge as a result of ART. A previous report found a switch from R5 to X4 viruses during effective ART in about 50% of patients who harboured R5 viruses at baseline [23]. Reports from other countries with limited CRF01_AE samples also demonstrated a high prevalence of X4 tropism in HIV-1 CRF01_AE: 31 and 40% of patients harboured X4 virus in studies from Singapore and Vietnam, respectively [24, 25] . Sequence compiling in China showed that 22% of patients harboured X4 and 40% R5X4 [26]. A study on patients with recently diagnosed infection showed significant higher number of predicted CXCR4 use in CRF01_AE compared with subtypes A, C and CRF01_AG [27]. In addition to ART experience, these data suggest that coreceptor usage may be influenced by viral subtype. No association between baseline CD4 count or viral load and detection of X4 viruses was found in our study.

Many genotypic tools and algorithms are available for the prediction of coreceptor tropisms. Although some studies have suggested that genotypic tools have a lower sensitivity to detect X4 virus compared with phenotypic assays [28, 29], our study demonstrates a large proportion of HIV-infected individuals harbour X4 virus. At an FPR of 5%, tropism prediction in those infected with HIV-1 CRF01_AE was shown to have 94.5% accuracy with 92.6% specificity and 97.9% sensitivity [30] .

This study has some limitations. First, a phenotypic assay was not performed because of restricted availability. However, previous studies have shown geno2pheno to provide good accuracy in tropism prediction [30]. Secondly, there was a limited number of non-AE HIV-1 subtypes in this sample set.

In conclusion, our study shows CRF01_AE and treatment failure to be significant factors associated with predicted X4 phenotype. In regions where CRF01_AE predominates and in patients with a history of treatment failure, use of CCR5 inhibitors must be considered with caution. Tropism testing should be performed whenever considering the use of CCR5 inhibitors, and this test should not be omitted even in resource-limited settings such as Thailand. Further studies on correlating phenotypes and viral response to CCR5 inhibitors should be performed.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We are grateful to the laboratory of Richard Harrigan, BC Center for Excellence in HIV/AIDS, Vancouver, BC, Canada for providing the sequencing primers and PCR protocol. This work was supported by a research grant from the Faculty of Medicine Ramathibodi Hospital, Mahidol University.

Conflicts of interest

None of the authors has a conflict of interest to declare.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Berger EA, Doms RW, Fenyo EM et al. A new classification for HIV-1. Nature 1998; 391: 240.
  • 2
    Zhu T, Mo H, Wang N et al. Genotypic and phenotypic characterization of HIV-1 patients with primary infection. Science 1993; 261: 11791181.
  • 3
    Shepherd JC, Jacobson LP, Qiao W et al. Emergence and persistence of CXCR4-tropic HIV-1 in a population of men from the multicenter AIDS cohort study. J Infect Dis 2008; 198: 11041112.
  • 4
    Garvey L, Nelson M, Latch N et al. CNS effects of a CCR5 inhibitor in HIV-infected subjects: a pharmacokinetic and cerebral metabolite study. J Antimicrob Chemother 2012; 67: 206212.
  • 5
    Soulie C, Tubiana R, Simon A et al. Presence of HIV-1 R5 viruses in cerebrospinal fluid even in patients harboring R5X4/X4 viruses in plasma. J Acquir Immune Defic Syndr 2009; 51: 6064.
  • 6
    Abel S, Davis JD, Ridgway CE, Hamlin JC, Vourvahis M. Pharmacokinetics, safety and tolerability of a single oral dose of maraviroc in HIV-negative subjects with mild and moderate hepatic impairment. Antivir Ther 2009; 14: 831837.
  • 7
    Patterson P, Magneres C, Sued O et al. A phase 4, single-arm, open-label, pilot study of maraviroc, raltegravir and darunavir/r in HIV-1 adults with triple class failure: TERCETO study. J Int AIDS Soc 2012; 15: 18268.
  • 8
    Whitcomb JM, Huang W, Fransen S et al. Development and characterization of a novel single-cycle recombinant-virus assay to determine human immunodeficiency virus type 1 coreceptor tropism. Antimicrob Agents Chemother 2007; 51: 566575.
  • 9
    Vandekerckhove LP, Wensing AM, Kaiser R et al. European guidelines on the clinical management of HIV-1 tropism testing. Lancet Infect Dis 2011; 11: 394407.
  • 10
    Auswinporn S, Jenwitheesuk E, Vibhagool A, Sookpranee M, Leechawengwong M, Chantratita W. Genotypic subtyping of the HIV-1 polymerase gene in 30 naave patients from Thailand. Southeast Asian J Trop Med Public Health 2001; 32: 346350.
  • 11
    Wirachsilp P, Kantakamalakul W, Foongladda S et al. Surveillance of subtype and genetic variation of the circulating strains of HIV-1 in Thailand. Southeast Asian J Trop Med Public Health 2007; 38: 814827.
  • 12
    Sungkanuparph S, Anekthananon T, Hiransuthikul N et al. Guidelines for antiretroviral therapy in HIV-1 infected adults and adolescents: the recommendations of the Thai AIDS Society (TAS) 2008. J Med Assoc Thai 2008; 91: 19251935.
  • 13
    Lengauer T, Sander O, Sierra S, Thielen A, Kaiser R. Bioinformatics prediction of HIV coreceptor usage. Nat Biotechnol 2007; 25: 14071410.
  • 14
    de Jong JJ, Goudsmit J, Keulen W et al. Human immunodeficiency virus type 1 clones chimeric for the envelope V3 domain differ in syncytium formation and replication capacity. J Virol 1992; 66: 757765.
  • 15
    Yu XF, Wang Z, Beyrer C et al. Phenotypic and genotypic characteristics of human immunodeficiency virus type 1 from patients with AIDS in northern Thailand. J Virol 1995; 69: 46494655.
  • 16
    Brumme ZL, Goodrich J, Mayer HB et al. Molecular and clinical epidemiology of CXCR4-using HIV-1 in a large population of antiretroviral-naive individuals. J Infect Dis 2005; 192: 466474.
  • 17
    Moyle GJ, Wildfire A, Mandalia S et al. Epidemiology and predictive factors for chemokine receptor use in HIV-1 infection. J Infect Dis 2005; 191: 866872.
  • 18
    Gulick RM, Lalezari J, Goodrich J et al. Maraviroc for previously treated patients with R5 HIV-1 infection. N Engl J Med 2008; 359: 14291441.
  • 19
    Wilkin TJ, Su Z, Kuritzkes DR et al. HIV type 1 chemokine coreceptor use among antiretroviral-experienced patients screened for a clinical trial of a CCR5 inhibitor: AIDS Clinical Trial Group A5211. Clin Infect Dis 2007; 44: 591595.
  • 20
    Saracino A, Monno L, Cibelli DC et al. Co-receptor switch during HAART is independent of virological success. J Med Virol 2009; 81: 20362044.
  • 21
    Huang W, Eshleman SH, Toma J et al. Coreceptor tropism in human immunodeficiency virus type 1 subtype D: high prevalence of CXCR4 tropism and heterogeneous composition of viral populations. J Virol 2007; 81: 78857893.
  • 22
    Tscherning C, Alaeus A, Fredriksson R et al. Differences in chemokine coreceptor usage between genetic subtypes of HIV-1. Virology 1998; 241: 181188.
  • 23
    Delobel P, Sandres-Saune K, Cazabat M et al. R5 to X4 switch of the predominant HIV-1 population in cellular reservoirs during effective highly active antiretroviral therapy. J Acquir Immune Defic Syndr 2005; 38: 382392.
  • 24
    Luu QP, Dean J, Do TT et al. HIV type 1 coreceptor tropism, CCR5 genotype, and integrase inhibitor resistance profiles in Vietnam: implications for the introduction of new antiretroviral regimens. AIDS Res Hum Retroviruses 2012; 28: 13441348.
  • 25
    Ng KY, Chew KK, Kaur P et al. High prevalence of CXCR4 usage among treatment-naive CRF01_AE and CRF51_01B-infected HIV-1 subjects in Singapore. BMC Infect Dis 2013; 13: 90.
  • 26
    Zhang C, Xu S, Wei J, Guo H. Predicted co-receptor tropism and sequence characteristics of China HIV-1 V3 loops: implications for the future usage of CCR5 antagonists and AIDS vaccine development. Int J Infect Dis 2009; 13: e212e216.
  • 27
    Chalmet K, Dauwe K, Foquet L et al. Presence of CXCR4-using HIV-1 in patients with recently diagnosed infection: correlates and evidence for transmission. J Infect Dis 2012; 205: 174184.
  • 28
    Low AJ, Dong W, Chan D et al. Current V3 genotyping algorithms are inadequate for predicting X4 co-receptor usage in clinical isolates. AIDS 2007; 21: F17F24.
  • 29
    Sanchez V, Masia M, Robledano C, Padilla S, Ramos JM, Gutierrez F. Performance of genotypic algorithms for predicting HIV-1 tropism measured against the enhanced-sensitivity Trofile coreceptor tropism assay. J Clin Microbiol 2010; 48: 41354139.
  • 30
    Shoombuatong W, Hongjaisee S, Barin F, Chaijaruwanich J, Samleerat T. HIV-1 CRF01_AE coreceptor usage prediction using kernel methods based logistic model trees. Comput Biol Med 2012; 42: 885889.