Shall I trust the report? Variable performance of Sanger sequencing revealed by deep sequencing on HIV drug resistance mutation detection

Impact of pretreatment low-abundance HIV-1 drug resistance on virological failure after 1 year of antiretroviral therapy in China

OBJECTIVES

To assess the impact of pretreatment low-abundance HIV drug-resistant variants (LA-DRVs) on virological outcomes among ART-naive HIV-1-infected Chinese people who initiated ART.

METHODS

A nested case-control study was conducted among HIV-1-infected individuals who had pretreatment drug resistance (PDR) genotypic results. Cases were defined as individuals with virological failure (HIV-1 RNA viral load ≥1000 copies/mL) after 1 year of ART, and controls were individuals from the same cohort whose viral load was less than 1000 copies/mL. Next-generation sequencing was used to identify low-abundance PDR mutations at detection thresholds of 10%, 2% and 1%. The mutant load was calculated by multiplying the abundance of HIV-1 drug-resistant variants by the pretreatment viral load. The impact of pretreatment low-abundance mutations on virological failure was estimated in logistic regression models.

RESULTS

Participants (43 cases and 100 controls) were included in this study for the analysis. The proportion of participants with PDR was higher in cases than in controls at different detection thresholds (44.2% versus 22.0%, P = 0.007 at 10% threshold; 58.1% versus 31.0%, P = 0.002 at 2% threshold; 90.7% versus 69.0%, P = 0.006 at 1% threshold). Compared with participants without PDR, participants with ≥10% detectable PDR mutations were associated with an increased risk of virological failure (adjusted OR 8.0, 95% CI 2.4-26.3, P = 0.001). Besides this, individuals with pretreatment LA-DRVs (2%-9% abundance range) had 5-fold higher odds of virological failure (adjusted OR 5.0, 95% CI 1.3-19.6, P = 0.021). Furthermore, LA-DRVs at 2%-9% abundance resistant to NRTIs and mutants with abundance of ≥10% resistant to NNRTIs had a 4-fold and 8-fold risk of experiencing virological failure, respectively. It was also found that a mutant load of more than 1000 copies/mL was predictive of virological failure (adjusted OR 7.2, 95% CI 2.5-21.1, P = 0.0003).

CONCLUSIONS

Low-abundance PDR mutations ranging from 2% to 9% of abundance can increase the risk of virological failure. Further studies are warranted to define a clinically relevant threshold of LA-DRVs and the role of NRTI LA-DRVs.

HIV-1 Genotypic Resistance Testing Using Sanger and Next-Generation Sequencing in Adults with Low-Level Viremia in China

OBJECTIVE

In this study, we aimed to determine drug-resistance mutations (DRMs) in HIV-1 patients with low-level viremia (LLV) and explored the performance of next-generation sequencing (NGS) in detecting HIV DRMs by using LLV samples.

METHODS

Overall, 80 samples with LLV were amplified and sequenced using a commercial Sanger sequencing (SS) genotyping kit. Furthermore, 51 samples successfully sequenced using SS were simultaneously subjected to NGS. Genotyping success rates of various viremia categories by two sequencing methods were calculated. Stanford HIV-1 drug-resistance database (HIVdb version 8.9) was used to analyze the DRMs. In the NGS assay, a threshold of 5% was considered for reporting low-frequency variants, and the DRMs detected using SS and NGS were compared.

RESULTS

The overall success rate of PR/RT regions was 88.1% (67/80) using SS and 86.3% (44/51) using NGS. Furthermore, a significant linear trend was noted between viral load and the genotyping success rate. A total of 38.8% (26/67) participants harbored at least one mutation, as revealed through SS. Moreover, the prevalence of DRMs in persistent LLV was significantly higher than that in intermittent LLV (62.1% vs. 21.1%; P < 0.05). A total of 69 DRMs were detected using the two sequencing methods at the threshold of 5%. Moreover, 10 DRMs missed by SS were detected using NGS, whereas 8 DRMs missed by NGS were detected by SS.

CONCLUSION

Our data suggested that the genotyping resistance testing is necessary to guide antiretroviral therapy optimization in LLV patients.

Establishment and application of a method of tagged-amplicon deep sequencing for low-abundance drug resistance in HIV-1

In the latest HIV-1 global drug resistance report released by WHO, countries are advised to strengthen pre-treatment monitoring of drug resistance in AIDS patients. In this study, we established an NGS-based segmented amplification HIV-1 drug resistance mutation detection method. The pol region of HIV-1 was divided into three short fragments for NGS. The entire amplification and sequencing panel were more cost-effective and batched by using the barcode sequence corresponding to the sample. Each parameter was evaluated using samples with known resistance variants frequencies. The nucleotide sequence error rate, amino acid error rate, and noise value of the NGS-based segmented amplification method were both less than 1%. When the threshold was 2%, the consensus sequences of the HIV-1 NL4-3 strain were completely consistent with the Sanger sequences. This method can detect the minimum viral load of the sample at 10² copies/ml, and the input frequency and detection frequency of HIV-1 resistance mutations within the range of 1%–100% had good conformity (R² = 0.9963; R² = 0.9955). This method had no non-specific amplification for Hepatitis B and C. Under the 2% threshold, the incidence of surveillance drug resistance mutations in ART-naive HIV-infected patients was 20.69%, among which NRTIs class resistance mutations were mainly.

HIV-1 Drug Resistance Assay Using Ion Torrent Next Generation Sequencing and On-Instrument End-to-End Analysis Software

HIV-1 antiretroviral therapy management requires sequencing the protease, reverse transcriptase, and integrase portions of the HIV-1 pol gene. Most resistance testing is performed with Sanger sequencing, which has limited ability to detect minor variants. Next generation sequencing (NGS) platforms enable variant detection at frequencies as low as 1% allowing for earlier detection of resistance and modification of therapy. Implementation of NGS assays in the clinical laboratory is hindered by complicated assay design, cumbersome wet bench procedures, and the complexity of data analysis and bioinformatics. We developed a complete NGS protocol and companion analysis and reporting pipeline using AmpliSeq multiplex PCR, Ion Torrent S5 XL sequencing, and Stanford's HIVdb resistance algorithm. Implemented as a Torrent Suite software plugin, the pipeline runs automatically after sequencing. An optimum variant frequency threshold of 10% was determined by comparing Sanger sequences of archived samples from ViroSeq testing, resulting in a sensitivity of 98.2% and specificity of 99.0%. The majority (91%) of drug resistance mutations were detected by both Sanger and NGS, with 1.7% only by Sanger and 7.3% only by NGS. Variant calls were highly reproducible and there was no cross-reactivity to VZV, HBV, CMV, EBV, and HCV. The limit of detection was 500 copies/mL. The NGS assay performance was comparable to ViroSeq Sanger sequencing and has several advantages, including a publicly available end-to-end analysis and reporting plugin. The assay provides a straightforward path for implementation of NGS for HIV drug resistance testing in the laboratory setting without additional investment in bioinformatics infrastructure and resources.

Viral resistance burden and APOBEC editing correlate with virological response in heavily treatment-experienced people living with multi-drug resistant HIV

BACKGROUND

The impact of drug resistance mutational load and APOBEC editing in heavily treatment-experienced (HTE) people living with multidrug-resistant HIV has not been investigated.

MATERIAL AND METHODS

This study explored the HIV-DNA and HIV-RNA mutational load of drug resistance and APOBEC-related mutations through next-generation sequencing (NGS, Illumina MiSeq) in 20 failing HTE participants enrolled in the PRESTIGIO registry.

RESULTS

The patients showed high levels of both HIV-DNA (4.5 [4.0-5.2] log₁₀ copies/10⁶ T-CD4+ cell) and HIV-RNA (4.5 [4.1-5.0] log₁₀ copies/mL) with complex resistance patterns in both compartments. Among the 255 drug-resistant mutations found, 66.3% were concordantly detected in both HIV-DNA and HIV-RNA; 71.3% of mutations were already present in historical Sanger genotypes. At an intra-patient frequency > 5%, a considerable proportion of mutations detected through DNA-NGS were found in historical genotypes but not through RNA-NGS, and few patients had APOBEC-related mutations. Of 14 patients who switched therapy, the five who failed treatment had DNA resistance with higher intra-patient frequency and higher DNA/RNA mutational load in a context of tendentially less pronounced APOBEC editing compared with those who responded.

CONCLUSIONS

Using NGS in HIV-DNA and HIV-RNA together with APOBEC editing evaluation might help to identify HTE individuals with MDR who are more prone to experience virological failure.

Expanding the Genetic and Phenotypic Spectrum of Female Infertility Caused by TUBB8 Mutations

Tubulin beta eight class VIII (TUBB8) is a subtype of β-tubulin that only exists in primates. TUBB8 mutations have been reported to cause arrest of oocyte maturation and embryonic development. We aim to further investigate the mutational spectrum of TUBB8 and its relevance with female infertility. In our study, infertile patients were recruited, and their basal and clinical characteristics were analyzed. Genomic DNA was extracted from peripheral blood donated by patients. Candidate variants were identified by whole-exome sequencing, selected by relevant criteria, and validated by Sanger sequencing. We found five heterozygous variants: c.C208A(p.P70T), c.T907C(p.C303R), c.G173A(p.R58K), c.G326T(p.G109V), and c.C916T(p.R306C) in TUBB8 among six infertile patients characterized by abnormal phenotypes in oocyte maturation, fertilization, or embryo development. Most of oocytes retrieved from affected individuals were arrested at GV (germinal vesicle) stage and early embryos were arrested at variable stages. In vitro experiments were performed, and the relationship between variant c.G173A(p.R58K), c.C208A(p.P70T), and infertility phenotype was confirmed. We also discussed the possibility about patient II-1 from family 4 is affected by germinal/germline mosaicism. These results expand the kinds of variants and phenotypic spectrum of TUBB8 variants with regard to female infertility.

Prevalence of drug resistance mutations in HIV-infected individuals with low-level viraemia under combination antiretroviral therapy: an observational study in a tertiary hospital in Northern Taiwan, 2017-19

BACKGROUND

Effective ART is crucial for combating the HIV pandemic. Clinically, plasma viral load monitoring to achieve virological suppression is the guide for an optimal ART. The presence of low-level viraemia (LLV) below the definition level of virological failure is a risk factor for ART failure. However, there is no treatment consensus over LLV yet, mainly due to the limitation of standard HIV-RNA genotyping and the resultant insufficient understanding of LLV characteristics.

OBJECTIVES

To better profile drug resistance mutations (DRMs) and the associated factors in cases experiencing LLV.

METHODS

A prospective observational study was conducted from 2017 to 2019. HIV-DNA was used as an alternative to HIV-RNA for HIV genotyping coupled with deep sequencing for ART-naive and ART-failure cases, as well as those with LLV.

RESULTS

Eighty-one ART-naive, 18 ART-failure and 16 LLV cases received HIV genotyping in the study. Three-quarters (12/16) of cases experiencing LLV harboured DRMs. Cases with LLV had higher prevalence of DRMs to NNRTIs than the ART-naive group (69% versus 20%, P < 0.001), but lower DRM prevalence to NRTIs than the ART-failure group (25% versus 61%, P < 0.001). Approximately half of the LLV cases had issues of suboptimal ART compliance/ART interruption, and 68.8% (11/16) did not display drug resistance to their ART at the time of LLV.

CONCLUSIONS

HIV DRM profiles in LLV cases were significantly different to those in ART-naive and ART-failure cases. Approaches to consolidate ART compliance and early exploration of potential ART resistance may be needed for cases experiencing LLV episodes.

HIV Drug Resistance Mutations Detection by Next-Generation Sequencing during Antiretroviral Therapy Interruption in China

Patients with antiretroviral therapy interruption have a high risk of virological failure when re-initiating antiretroviral therapy (ART), especially those with HIV drug resistance. Next-generation sequencing may provide close scrutiny on their minority drug resistance variant. A cross-sectional study was conducted in patients with ART interruption in five regions in China in 2016. Through Sanger and next-generation sequencing in parallel, HIV drug resistance was genotyped on their plasma samples. Rates of HIV drug resistance were compared by the McNemar tests. In total, 174 patients were included in this study, with a median 12 (interquartile range (IQR), 6–24) months of ART interruption. Most (86.2%) of them had received efavirenz (EFV)/nevirapine (NVP)-based first-line therapy for a median 16 (IQR, 7–26) months before ART interruption. Sixty-one (35.1%) patients had CRF07_BC HIV-1 strains, 58 (33.3%) CRF08_BC and 35 (20.1%) CRF01_AE. Thirty-four (19.5%) of the 174 patients were detected to harbor HIV drug-resistant variants on Sanger sequencing. Thirty-six (20.7%), 37 (21.3%), 42 (24.1%), 79 (45.4%) and 139 (79.9) patients were identified to have HIV drug resistance by next-generation sequencing at 20% (v.s. Sanger, p = 0.317), 10% (v.s. Sanger, p = 0.180), 5% (v.s. Sanger, p = 0.011), 2% (v.s. Sanger, p < 0.001) and 1% (v.s. Sanger, p < 0.001) of detection thresholds, respectively. K65R was the most common minority mutation, of 95.1% (58/61) and 93.1% (54/58) in CRF07_BC and CRF08_BC, respectively, when compared with 5.7% (2/35) in CRF01_AE (p < 0.001). In 49 patients that followed-up a median 10 months later, HIV drug resistance mutations at >20% frequency such as K103N, M184VI and P225H still existed, but with decreased frequencies. The prevalence of HIV drug resistance in ART interruption was higher than 15% in the survey. Next-generation sequencing was able to detect more minority drug resistance variants than Sanger. There was a sharp increase in minority drug resistance variants when the detection threshold was below 5%.

Potential challenges to sustained viral load suppression in the HIV treatment programme in South Africa: a narrative overview

Background

South Africa, with one of the highest HIV prevalences in the world, introduced the universal test and treat (UTT) programme in September 2016. Barriers to sustained viral suppression may include drug resistance in the pre-treated population, non-adherence, acquired resistance; pharmacokinetics and pharmacodynamics, and concurrent use of alternative treatments.

Objective

The purpose of this review is to highlight potential challenges to achieving sustained viral load suppression in South Africa (SA), a major expectation of the UTT initiative.

Methodology

Through the PRISMA approach, published articles from South Africa on transmitted drug resistance; adherence to ARV; host genetic factors in drug pharmacokinetics and pharmacodynamics, and interactions between ARV and herbal medicine were searched and reviewed.

Results

The level of drug resistance in the pre-treated population in South Africa has increased over the years, although it is heterogeneous across and within Provinces. At least one study has documented a pre-treated population with moderate (> 5%) or high (> 15%) levels of drug resistance in eight of the nine Provinces. The concurrent use of ARV and medicinal herbal preparation is fairly common in SA, and may be impacting negatively on adherence to ARV. Only few studies have investigated the association between the genetically diverse South African population and pharmacokinetics and pharmacodynamics of ARVs.

Conclusion

The increasing levels of drug resistant viruses in the pre-treated population poses a threat to viral load suppression and the sustainability of first line regimens. Drug resistance surveillance systems to track the emergence of resistant viruses, study the burden of prior exposure to ARV and the parallel use of alternative medicines, with the goal of minimizing resistance development and virologic failure are proposed for all the Provinces of South Africa. Optimal management of the different drivers of drug resistance in the pre-treated population, non-adherence, and acquired drug resistance will be beneficial in ensuring sustained viral suppression in at least 90% of those on treatment, a key component of the 90-90-90 strategy.

Next-Generation Sequencing for HIV Drug Resistance Testing: Laboratory, Clinical, and Implementation Considerations

Higher accessibility and decreasing costs of next generation sequencing (NGS), availability of commercial kits, and development of dedicated analysis pipelines, have allowed an increasing number of laboratories to adopt this technology for HIV drug resistance (HIVDR) genotyping. Conventional HIVDR genotyping is traditionally carried out using population-based Sanger sequencing, which has a limited capacity for reliable detection of variants present at intra-host frequencies below a threshold of approximately 20%. NGS has the potential to improve sensitivity and quantitatively identify low-abundance variants, improving efficiency and lowering costs. However, some challenges exist for the standardization and quality assurance of NGS-based HIVDR genotyping. In this paper, we highlight considerations of these challenges as related to laboratory, clinical, and implementation of NGS for HIV drug resistance testing. Several sources of variation and bias occur in each step of the general NGS workflow, i.e., starting material, sample type, PCR amplification, library preparation method, instrument and sequencing chemistry-inherent errors, and data analysis options and limitations. Additionally, adoption of NGS-based HIVDR genotyping, especially for clinical care, poses pressing challenges, especially for resource-poor settings, including infrastructure and equipment requirements and cost, logistic and supply chains, instrument service availability, personnel training, validated laboratory protocols, and standardized analysis outputs. The establishment of external quality assessment programs may help to address some of these challenges and is needed to proceed with NGS-based HIVDR genotyping adoption.