A Highly Sensitive and Specific Model for Predicting HIV-1 Tropism in Treatment-Experienced Patients Combining Interpretation of V3 Loop Sequences and Clinical Parameters

Evolution Driven By A Varying Host Environment Selects For Distinct HIV-1 Entry Phenotypes and Other Informative Variants

HIV-1 generates remarkable intra- and inter-host viral diversity during infection. In response to dynamic selective pressures of the host environment, HIV-1 will evolve distinct phenotypes - biological features that provide fitness advantages. The transmitted form of HIV-1 has been shown to require a high density of CD4 on the target cell surface (as found on CD4+ T cells) and typically uses CCR5 as a co-receptor during entry. This phenotype is referred to as R5 T cell-tropic (or R5 T-tropic); however, HIV-1 can switch to a secondary co-receptor, CXCR4, resulting in a X4 T cell-tropic phenotype. Macrophage-tropic (or M-tropic) HIV-1 can evolve to efficiently enter cells expressing low densities of CD4 on their surface (such as macrophages/microglia). So far only CCR5-using M-tropic viruses have been found. M-tropic HIV-1 is most frequently found within the central nervous system, and infection of the CNS has been associated with neurological impairment. It has been shown that interferon resistance phenotypes have a selective advantage during transmission, but the underlying mechanism of this is still unclear. During untreated infection, HIV-1 evolves under selective pressure from both the humoral/antibody response and CD8+ T cell killing. Sufficiently potent antiviral therapy will suppress viral replication, but if the antiviral drugs are not sufficiently potent to stop replication then the replicating virus will evolve drug resistance. HIV-1 phenotypes are highly relevant to treatment efforts, clinical outcomes, vaccine studies, and cure strategies. Therefore, it is critical to understand the dynamics of the host environment that drive these phenotypes and how they affect HIV-1 pathogenesis. This review will provide a comprehensive discussion of HIV-1 entry, transmission, and drug resistance phenotypes. Finally, we will assess the methods used in previous and current research to characterize these phenotypes.

Performance of Geno2Pheno[coreceptor] to infer coreceptor use in human immunodeficiency virus type 1 (HIV-1) subtype A

BACKGROUND

Assessment of human immunodeficiency virus type 1 (HIV-1) coreceptor usage is required prior to treatment with the CCR5 antagonist maraviroc to exclude the presence of CXCR4-using (X4) strains. Genotype-based interpretation systems are mostly designed on subtype B and have been reported to be less accurate for subtype A/CRF02_AG.

OBJECTIVES

To evaluate the performance of the widely used Geno2Pheno[coreceptor] (G2P[c]) algorithm for prediction of coreceptor usage with subtype A/CRF02_AG vs. subtype B.

STUDY DESIGN

Co-receptor tropism of 24 subtype A/CRF02_AG and 24 subtype B viruses was measured phenotypically by a homebrew single-cycle assay and genotypically by using G2P[c]. Samples with discrepant genotype-phenotype results were analyzed by next generation sequencing (NGS) and interpreted by the NGS Geno2Pheno algorithm (G2P[454]).

RESULTS

At 10% false positive rate (FPR), the G2P[c]/phenotype discordance rate was 12.5% (n = 3) for subtype A/CRF02_AG and 8.3% (n = 2) for subtype B. Minority X4 species escaping detection by bulk sequencing but documented by NGS explained the two subtype B and possibly one subtype A/CRF02_AG discordant case. The other two subtype A/CRF02_AG miscalled by G2P[c] could be explained by X4 overcalling at borderline FPR and/or by algorithm failure.

DISCUSSION

Our study did not demonstrate relevantly higher G2P[c] inaccuracy with subtype A/CRF02_AG with respect to subtype B. Genotype/phenotype discordances can be due to different reasons, including but not limited to, algorithm inaccuracy. Very large genotype/phenotype correlation panels are required to detect and explain the reason for any consistent difference in genotypic tropism prediction for subtype A/CRF02_AG vs. subtype B.

Deep Sequencing of the HIV-1 env Gene Reveals Discrete X4 Lineages and Linkage Disequilibrium between X4 and R5 Viruses in the V1/V2 and V3 Variable Regions

HIV-1 requires the CD4 receptor and a coreceptor (CCR5 [R5 phenotype] or CXCR4 [X4 phenotype]) to enter cells. Coreceptor tropism can be assessed by either phenotypic or genotypic analysis, the latter using bioinformatics algorithms to predict tropism based on the env V3 sequence. We used the Primer ID sequencing strategy with the MiSeq sequencing platform to reveal the structure of viral populations in the V1/V2 and C2/V3 regions of the HIV-1 env gene in 30 late-stage and 6 early-stage subjects. We also used endpoint dilution PCR followed by cloning of env genes to create pseudotyped virus to explore the link between genotypic predictions and phenotypic assessment of coreceptor usage. We found out that the most stringently sequence-based calls of X4 variants (Geno2Pheno false-positive rate [FPR] of ≤2%) formed distinct lineages within the viral population, and these were detected in 24 of 30 late-stage samples (80%), which was significantly higher than what has been seen previously by using other approaches. Non-X4 lineages were not skewed toward lower FPR scores in X4-containing populations. Phenotypic assays showed that variants with an intermediate FPR (2 to 20%) could be either X4/dual-tropic or R5 variants, although the X4 variants made up only about 25% of the lineages with an FPR of <10%, and these variants carried a distinctive sequence change. Phylogenetic analysis of both the V1/V2 and C2/V3 regions showed evidence of recombination within but very little recombination between the X4 and R5 lineages, suggesting that these populations are genetically isolated.

IMPORTANCE

Primer ID sequencing provides a novel approach to study genetic structures of viral populations. X4 variants may be more prevalent than previously reported when assessed by using next-generation sequencing (NGS) and with a greater depth of sampling than single-genome amplification (SGA). Phylogenetic analysis to identify lineages of sequences with intermediate FPR values may provide additional information for accurately predicting X4 variants by using V3 sequences. Limited recombination occurs between X4 and R5 lineages, suggesting that X4 and R5 variants are genetically isolated and may be replicating in different cell types or that X4/R5 recombinants have reduced fitness.

V3 net charge: additional tool in HIV-1 tropism prediction

Genotype-based algorithms are valuable tools for the identification of patients eligible for CCR5 inhibitors administration in clinical practice. Among the available methods, geno2pheno[coreceptor] (G2P) is the most used online tool for tropism prediction. This study was conceived to assess if the combination of G2P prediction with V3 peptide net charge (NC) value could improve the accuracy of tropism prediction. A total of 172 V3 bulk sequences from 143 patients were analyzed by G2P and NC values. A phenotypic assay was performed by cloning the complete env gene and tropism determination was assessed on U87_CCR5(+)/CXCR4(+) cells. Sequences were stratified according to the agreement between NC values and G2P results. Of sequences predicted as X4 by G2P, 61% showed NC values higher than 5; similarly, 76% of sequences predicted as R5 by G2P had NC values below 4. Sequences with NC values between 4 and 5 were associated with different G2P predictions: 65% of samples were predicted as R5-tropic and 35% of sequences as X4-tropic. Sequences identified as X4 by NC value had at least one positive residue at positions known to be involved in tropism prediction and positive residues in position 32. These data supported the hypothesis that NC values between 4 and 5 could be associated with the presence of dual/mixed-tropic (DM) variants. The phenotypic assay performed on a subset of sequences confirmed the tropism prediction for concordant sequences and showed that NC values between 4 and 5 are associated with DM tropism. These results suggest that the combination of G2P and NC could increase the accuracy of tropism prediction. A more reliable identification of X4 variants would be useful for better selecting candidates for Maraviroc (MVC) administration, but also as a predictive marker in coreceptor switching, strongly associated with the phase of infection.

[Methods for determination of HIV tropism and their clinical use]

Determination of HIV-1 tropism is mandatory before using CCR5 antagonists in clinical practice. One drug of this class, maraviroc, has been approved for the treatment of HIV infection. The phenotypic assay, TrofileTM, was clinically validated in the clinical development program of maraviroc and has been widely used to select candidates for maraviroc therapy. Phenotypic tests, however, have the disadvantage of being complex, are costly and time-consuming, and their accessibility is limited, which hampers their routine use in clinical diagnosis. Genotypic assays, based on sequencing the third hypervariable (V3 loop) of the viral gene env, interpreted according to various genotypic bioinformatic tools, such as geno2pheno and PSSM, are faster and cheaper than phenotypic assays, and are also more accessible. In retrospective analyses of the maraviroc pivotal trials, genotypic methods using either conventional ("bulk") or deep-sequencing technology predicted virologic response to maraviroc similarly to phenotypic assays and are now included within several European recommendations to guide the clinical use of CCR5 antagonists.

Correlation of the virological response to short-term maraviroc monotherapy with standard and deep-sequencing-based genotypic tropism prediction methods

Genotypic tropism testing methods are emerging as the first step before prescription of the CCR5 antagonist maraviroc (MVC) to HIV-infected patients in Europe. Studies validating genotypic tests have included other active drugs that could have potentially convoluted the effects of MVC. The maraviroc clinical test (MCT) is an in vivo drug sensitivity test based on the virological response to a short-term exposure to MVC monotherapy. Thus, our aim was to compare the results of genotypic tropism testing methods with the short-term virological response to MVC monotherapy. A virological response in the MCT was defined as a ≥ 1-log(10) decrease in HIV RNA or undetectability after 8 days of drug exposure. Seventy-three patients undergoing the MCT were included in this study. We used both standard genotypic methods (n = 73) and deep sequencing (n = 27) on MCT samples at baseline. For the standard methods, the most widely used genotypic algorithms for analyzing the V3 loop sequence, geno2pheno and PSSM, were used. For deep sequencing, the geno2pheno algorithm was used with a false-positive rate cutoff of 3.5. The discordance rates between the standard genotypic methods and the virological response were approximately 20% (including mostly patients without a virological response). Interestingly, these discordance rates were similar to that obtained from deep sequencing (18.5%). The discordance rates between the genotypic methods (tropism assays predictive of the use of the CCR5 coreceptor) and the MCT (in vivo MVC sensitivity assay) indicate that the algorithms used by genotypic methods are still not sufficiently optimized.

Evaluation of genotypic tropism prediction tests compared with in vitro co-receptor usage in HIV-1 primary isolates of diverse subtypes

OBJECTIVES

To evaluate the sensitivity and specificity of genotypic methods for predicting the co-receptor usage of subtypes B and non-B HIV-1 primary isolates, using as gold standard the infectivity of each primary isolate in GHOST cells stably expressing HIV-1 co-receptors.

METHODS

Primary isolates were obtained by co-culturing either patient's peripheral blood mononuclear cells (PBMCs) or ultracentrifuged plasma with donor-activated PBMCs. In vitro co-receptor usage was determined by infecting GHOST cells. Tropism prediction, based on V3 sequences, was determined with simple rules and bioinformatic tools (Geno2pheno[coreceptor] and WebPSSM).

RESULTS

This study includes 102 HIV-1 primary isolates; 23 (22.5%) subtype B and 79 (77.5%) non-B genetic forms. V3 sequences were classified into six subtypes (A-G), although 32 (31.4%) were circulating recombinant forms and 21 (20.6%) were unique recombinant forms. Sixty-nine isolates were R5, 27 R5X4 and 6 X4. The highest levels of sensitivity and specificity for the detection of X4 strains among V3 sequences, between 91% and 100%, were obtained by using PSSM(x4r5), PSSM(si/nsi) and the 11/25 rule for sequences of subtypes A, B and G, but not for subtype F. Establishing the recommended cut-off for clinical settings of a 10% false positive rate for Geno2pheno, we obtained 93% specificity and 97% sensitivity.

CONCLUSIONS

Comparing genotypic assays for HIV-1 co-receptor use with a cell-culture phenotypic assay could provide more reliable results of sensitivity and specificity for the detection of X4 strains than comparing them with recombinant assays, considered as gold standard. In general, except for subtype F isolates, there is a good correlation for tropism prediction.

Genotypic resistance testing in routine clinical care

PURPOSE OF REVIEW

Genotypic resistance testing has become part of routine clinical management of HIV-infected patients. Focussing on observational studies, this review looks at recent advances in this area.

RECENT FINDINGS

Translation of the nucleotide sequence generated by the resistance test into clinically useful information remains a major challenge. A recent key development is the availability of therapy optimization tools to predict regimens that are most likely to achieve virological suppression. Standard genotypic resistance testing only examines protease and part of reverse transcriptase; as drugs are licensed to further targets, it has become necessary to expand the repertoire for testing. Traditionally, genotypic testing has not been attempted at viral loads less than 1000 copies/ml, but recent studies indicate that major mutations are often detected at much lower levels. Similarly, various methods have been developed for the detection of minority variants including allele-specific PCR, single-genome sequencing, and ultra-deep sequencing.

SUMMARY

The technology and interpretation of genotypic resistance tests is in a phase of rapid development. It remains uncertain which of these developments will become part of routine clinical practice.

[Laboratory diagnosis of HIV infection, viral tropism and resistance to antiretrovirals]

The accurate diagnosis of HIV infection demands that to consider a positive result, at least three assays with different antigenic base should be used, one of them, Western-Blot being mandatory for confirmation. Fourth generation ELISAs shorten the window phase to 13-15 days, as they now include p24 antigen detection. Proviral DNA or Viral RNA detection by molecular methods have proved useful for addressing complex situations in which serology was inconclusive. Viral load (HIV-RNA) is routinely used to follow-up HIV infected patients and is used for treatment initiation decisions. It is also used to monitor viral failure. When this happens, resistance tests are needed to guide treatment changes. Resistance is also used to assess the transmission of drug resistance to newly diagnosed patients. Finally, before using an anti-CCR5 drug, viral tropism needs to be determined. This can be done using genotypic tests, widely available in many HIV labs, or phenotypic tests, only available at certain sites.

Performance of genotypic algorithms for predicting HIV-1 tropism measured against the enhanced-sensitivity Trofile coreceptor tropism assay

The objectives of this study were to assess the performance of genotypic algorithms for predicting CXCR4-using virus, with enhanced sensitivity Trofile HIV coreceptor tropism assay (ES Trofile) as the reference, and to compare the concordance/accuracy of genotypic tests with ES Trofile and with the original Trofile assay. Paired phenotypic and genotypic determinations of HIV-1 coreceptor usage were compared in plasma samples from HIV-1-infected patients. Sequencing of the third hypervariable (V3) loop of the viral gene and phenotypic assays were performed for each sample. Genotypic rules used to predict tropism were Geno2pheno (false-positive rate at 1 to 20%), position-specific scoring matrix X4R5 (PSSM(X4R5)) and PSSM(sinsi) (where "sinsi" stands for syncytium inducing and non-syncytium inducing), and the 11/25, 11/24/25, and net charge rules. Two hundred forty-four phenotypic and genotypic samples were tested. Coreceptor usage was obtained from ES Trofile for 145 (59%) samples and from Trofile for 99 (41%) samples. The highest concordance (82.6%) was obtained with PSSM(X4R5) when ES Trofile was used as the reference. Geno2pheno at a 20% false-positive rate showed the highest sensitivity (76.7%) for CXCR4-using virus detection with ES Trofile. Samples from naïve subjects and those with CD4 cell counts between 200 and 500 cells/mm(3) showed the best predictive performance. Overall, the accuracy of the bioinformatics tools to detect CXCR4-using virus was similar for ES Trofile and Trofile; however, the negative predictive values for genotypic tools with ES Trofile were slightly higher than they were with Trofile. The accuracy of genotypic algorithms for detecting CXCR4-using viruses is high when using ES Trofile as the reference. Results are similar to those obtained with Trofile. The concordance with ES Trofile is better with higher CD4 cell counts and nonexposure to antiretroviral therapy.