Performance comparison of next-generation sequencing platforms for determining HIV-1 coreceptor use

New aspects of the Virus Life Cycle and Clinical Utility of Next Generation Sequencing based HIV-1 Resistance Testing in the Genomic, the Proviral and the Viral Reservoir of Peripheral Blood Mononuclear Cells

BACKGROUND

Typically, genotypic resistance testing is recommended at the start of antiretroviral therapy and is even mandatory in cases of virologic failure. The material of choice is plasma viral RNA. However, in patients with low viremia (viral load < 500 copies/ml), resistance testing by population-based sequencing is very difficult.

OBJECTIVE

Therefore, we aimed to investigate whether next generation sequencing (NGS) from proviral DNA and RNA could be an alternative.

MATERIAL AND METHODS

EDTA blood samples (n = 36) from routine clinical viral load testing were used for the study. Viral loads ranged from 96 to 390,000 copies/mL, with 100% of samples having low viremia. Distribution of subtypes A (n = 2), B (n = 16), C (n = 4), D (n = 2), G (1), CRF02 AG (n = 5), CRF01 AE (n = 5), undefined/mixed (n = 4). The extracted consensus sequences were uploaded to the Stanford HIV Drug Resistance Data Base and Geno2pheno for online analysis of drug resistance mutations and resistance factors.

RESULTS

A total of 2476 variants or drug resistance mutations (DRMs) were detected with Sanger sequencing, compared with 2892 variants with NGS. An average of 822/1008 variants were identified in plasma viral RNA by Sanger or NGS sequencing, 834/956 in cellular viral RNA, and 820/928 in cellular viral DNA.

CONCLUSIONS

Both methods are well suited for the detection of HIV substitutions or drug resistance mutations. Our results suggest that cellular RNA or cellular viral DNA is an informative alternative to plasma viral RNA for variant detection in patients with low viremia, as shown by the high correlation of variants in the different viral pools. And we show that by using UDS, a plus of two DRMs per patient becomes visible and that can make a big difference in the assessment of the expected resistance behavior of the virus.

Comparative analyses of error handling strategies for next-generation sequencing in precision medicine

Next-generation sequencing (NGS) offers the opportunity to sequence millions and billions of DNA sequences in a short period, leading to novel applications in personalized medicine, such as cancer diagnostics or antiviral therapy. Nevertheless, sequencing technologies have different error rates, which occur during the sequencing process. If the NGS data is used for diagnostics, these sequences with errors are typically neglected or a worst-case scenario is assumed. In the current study, we focused on the impact of ambiguous bases on therapy recommendations for Human Immunodeficiency Virus 1 (HIV-1) patients. Concretely, we analyzed the treatment recommendation with entry blockers based on prediction models for co-receptor tropism. We compared three different error handling strategies that have been used in the literature, namely (i) neglection, (ii) worst-case assumption, and (iii) deconvolution with a majority vote. We could show that for two or more ambiguous positions per sequence a reliable prediction is generally no longer possible. Moreover, also the position of ambiguity plays a crucial role. Thus, we analyzed the error probability distributions of existing sequencing technologies, e.g., Illumina MiSeq or PacBio, with respect to the aforementioned error handling strategies and it turned out that neglection outperforms the other strategies in the case where no systematic errors are present. In other cases, the deconvolution strategy with the majority vote should be preferred.

Long-term evolution of transmitted CXCR4-using HIV-1 under effective antiretroviral therapy

OBJECTIVE

To study the long-term evolution of the transmitted CXCR4-using viruses. CCR5-using viruses (R5 viruses) predominate during primary HIV-1 infections (PHI) while CXCR4-using viruses are isolated in less than 10% of PHI.

DESIGN

Six patients infected with an R5X4 virus, detected by a sensitive phenotypic assay during PHI, were matched with six patients infected with a pure R5 virus for sex, Fiebig stage, time of antiretroviral initiation and duration of follow-up.

METHODS

We used MiSeq ultra-deep sequencing to determine the composition of the virus quasispecies during PHI and at the end of follow-up (median time of follow-up: 12.5 years).

RESULTS

X4 viruses were detected by genetic analysis in three of six samples from the R5X4 group, accounting for 1.3-100% of the virus quasispecies, during PHI, and in four of six samples (accounting for 6.7-100%) at the end of follow-up. No X4 virus was detected in the R5 group during PHI and in only one patient (accounting for 1.2%) at the end of follow-up. The complexity of the virus quasispecies at the stage of PHI was higher in the R5X4 group than in the R5 group. Complexity increased from PHI to the end of follow-up in the R5 group but remained stable in the R5X4 group.

CONCLUSION

CXCR4-using viruses persisted in the peripheral blood mononuclear cells of several patients on suppressive antiretroviral therapy for a median duration of 12.5 years after PHI. The genetic complexity of HIV-1 evolved differently post-PHI in patients infected with R5X4 viruses from those infected with R5 viruses.

Prevalence and clinical impact of minority resistant variants in patients failing an integrase inhibitor-based regimen by ultra-deep sequencing

Background

Integrase strand transfer inhibitors (INSTIs) are recommended by international guidelines as first-line therapy in antiretroviral-naive and -experienced HIV-1-infected patients.

Objectives

This study aimed at evaluating the prevalence at failure of INSTI-resistant variants and the impact of baseline minority resistant variants (MiRVs) on the virological response to an INSTI-based regimen.

Methods

Samples at failure of 134 patients failing a raltegravir-containing (n = 65), an elvitegravir-containing (n = 20) or a dolutegravir-containing (n = 49) regimen were sequenced by Sanger sequencing and ultra-deep sequencing (UDS). Baseline samples of patients with virological failure (VF) (n = 34) and of those with virological success (VS) (n = 31) under INSTI treatment were sequenced by UDS. Data were analysed using the SmartGene platform, and resistance was interpreted according to the ANRS algorithm version 27.

Results

At failure, the prevalence of at least one INSTI-resistant variant was 39.6% by Sanger sequencing and 57.5% by UDS, changing the interpretation of resistance in 17/134 (13%) patients. Among 53 patients harbouring at least one resistance mutation detected by both techniques, the most dominant INSTI resistance mutations were N155H (45%), Q148H/K/R (23%), T97A (19%) and Y143C (11%). There was no difference in prevalence of baseline MiRVs between patients with VF and those with VS. MiRVs found at baseline in patients with VF were not detected at failure either in majority or minority mutations.

Conclusions

UDS is more sensitive than Sanger sequencing at detecting INSTI MiRVs at treatment failure. The presence of MiRVs at failure could be important to the decision to switch to other INSTIs. However, there was no association between the presence of baseline MiRVs and the response to INSTI-based therapies in our study.

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.

Accurate quantification of CXCR4-using HIV-1 variants by Illumina deep-sequencing

: Next-generation sequencing is a sensitive method for determining HIV-1 tropism but there is a lack of data on the quantification of X4 variants. We evaluated MiSeq and 454 GS-Junior platforms for determining HIV-1 tropism and for quantifying X4 variants. Both platforms were 93% concordant for determining HIV-1 tropism and correlated well for determining the proportion of X4 variants (Spearman correlation, ρ = 0.748; P < 0.0001). MiSeq Illumina sequencing seems to be well adapted for characterizing X4-containing samples.

CasPER, a method for directed evolution in genomic contexts using mutagenesis and CRISPR/Cas9

Here we describe a method for robust directed evolution using mutagenesis of large sequence spaces in their genomic contexts. The method employs error-prone PCR and Cas9-mediated genome integration of mutant libraries of large-sized donor variants into single or multiple genomic sites with efficiencies reaching 98-99%. From sequencing of genome integrants, we determined that the mutation frequency along the donor fragments is maintained evenly and successfully integrated into the genomic target loci, indicating that there is no bias of mutational load towards the proximity of the double strand break. To validate the applicability of the method for directed evolution of metabolic gene products we engineered two essential enzymes in the mevalonate pathway of Saccharomyces cerevisiae with selected variants supporting up to 11-fold higher production of isoprenoids. Taken together, our method extends on existing CRISPR technologies by facilitating efficient mutagenesis of hundreds of nucleotides in cognate genomic contexts.

Performance comparison of deep sequencing platforms for detecting HIV-1 variants in the pol gene

The present study compares the performances of an in-house sequencing protocol developed on MiSeq, the Sanger method, and the 454 GS-FLX for detecting and quantifying drug-resistant mutations (DRMs) in the human immunodeficiency virus polymerase gene (reverse transcriptase [RT] and protease [PR]). MiSeq sequencing identified all the resistance mutations detected by bulk sequencing (n = 84). Both the MiSeq and 454 GS-FLX platforms identified 67 DRMs in the RT and PR regions, but a further 25 DRMs were identified by only one or other of them. Pearson's analysis showed good concordance between the percentage of drug-resistant variants determined by MiSeq and 454 GS-FLX sequencing (ρ = .77, P < .0001). The MiSeq platform is as accurate as the 454 GS-FLX Roche system for determining RT and PR DRMs and could be used for monitoring human immunodeficiency virus type 1 drug resistance.

Multicenter validation of cancer gene panel-based next-generation sequencing for translational research and molecular diagnostics

The simultaneous detection of multiple somatic mutations in the context of molecular diagnostics of cancer is frequently performed by means of amplicon-based targeted next-generation sequencing (NGS). However, only few studies are available comparing multicenter testing of different NGS platforms and gene panels. Therefore, seven partner sites of the German Cancer Consortium (DKTK) performed a multicenter interlaboratory trial for targeted NGS using the same formalin-fixed, paraffin-embedded (FFPE) specimen of molecularly pre-characterized tumors (n = 15; each n = 5 cases of Breast, Lung, and Colon carcinoma) and a colorectal cancer cell line DNA dilution series. Detailed information regarding pre-characterized mutations was not disclosed to the partners. Commercially available and custom-designed cancer gene panels were used for library preparation and subsequent sequencing on several devices of two NGS different platforms. For every case, centrally extracted DNA and FFPE tissue sections for local processing were delivered to each partner site to be sequenced with the commercial gene panel and local bioinformatics. For cancer-specific panel-based sequencing, only centrally extracted DNA was analyzed at seven sequencing sites. Subsequently, local data were compiled and bioinformatics was performed centrally. We were able to demonstrate that all pre-characterized mutations were re-identified correctly, irrespective of NGS platform or gene panel used. However, locally processed FFPE tissue sections disclosed that the DNA extraction method can affect the detection of mutations with a trend in favor of magnetic bead-based DNA extraction methods. In conclusion, targeted NGS is a very robust method for simultaneous detection of various mutations in FFPE tissue specimens if certain pre-analytical conditions are carefully considered.

High levels of local inter- and intra-host genetic variation of West Nile virus and evidence of fine-scale evolutionary pressures

West Nile virus (WNV; Flaviviridae, Flavivirus) has been endemic in New York State (NYS) since its 1999 introduction, yet prevalence in Culex mosquitoes varies substantially over small spatial and temporal scales. It is unclear if viral genetics plays a role in this variability, as genetic and phenotypic characterization on local scales has generally been lacking. In addition, intrahost diversity of circulating strains have not been fully characterized despite the documented role of minority variants in viral fitness and virulence. In an effort to characterize WNV variability within epidemiologically relevant scales, we performed phylogenetic analyses on NYS isolates from 1999 to 2012. In addition, we performed full-genome, deep-sequencing and genetic analyses on 15 WNV strains isolated in 2012 from Cx. pipiens in an endemic focus of Suffolk County, NY. Our results indicate continued evolution and seasonal maintenance in NYS, yet also widespread mixing and high levels of genetic diversity within geographic foci and individual seasons. Well supported local clusters with shared amino acid differences were identified and suggest local evolutionary pressures and the potential for phenotypic variability. Intrahost diversity of focal isolates was also high, with polymorphism at levels >1.0% identified in approximately 10% of the WNV genome. Although most minority mutations were unique, mutational hotspots shared among local isolates were identified, particularly in C, NS1 and NS2A genes. The most polymorphic region, positions 3198-3388 of the NS1 gene, was comprised predominately of non-synonymous mutations, suggesting a selective advantage for amino acid diversity in this region.