Format

Send to

Choose Destination
J Clin Microbiol. 2018 Nov 27;56(12). pii: e01443-18. doi: 10.1128/JCM.01443-18. Print 2018 Dec.

Limited Marginal Utility of Deep Sequencing for HIV Drug Resistance Testing in the Age of Integrase Inhibitors.

Author information

1
Department of Epidemiology, University of Washington, Seattle, Washington, USA.
2
Department of Laboratory Medicine, University of Washington, Seattle, Washington, USA.
3
Vaccine and Infectious Disease Division, Fred Hutchinson Cancer Research Center, Seattle, Washington, USA.
4
Department of Medicine, University of Washington, Seattle, Washington, USA.
5
Department of Laboratory Medicine, University of Washington, Seattle, Washington, USA agrening@uw.edu.

Abstract

HIV drug resistance genotyping is a critical tool in the clinical management of HIV infections. Although resistance genotyping has traditionally been conducted using Sanger sequencing, next-generation sequencing (NGS) is emerging as a powerful tool due to its ability to detect low-frequency alleles. However, the clinical value added from NGS approaches to antiviral resistance testing remains to be demonstrated. We compared the variant detection capacity of NGS versus Sanger sequencing methods for resistance genotyping in 144 drug resistance tests (105 protease-reverse transcriptase tests and 39 integrase tests) submitted to our clinical virology laboratory over a four-month period in 2016 for Sanger-based HIV drug resistance testing. NGS detected all true high-frequency drug resistance mutations (>20% frequency) found by Sanger sequencing, with greater accuracy in one instance of a Sanger-detected false positive. Freely available online NGS variant callers HyDRA and PASeq were superior to Sanger methods for interpretations of allele linkage and automated variant calling. NGS additionally detected low-frequency mutations (1 to 20% frequency) associated with higher levels of drug resistance in 30/105 (29%) protease-reverse transcriptase tests and 4/39 (10%) integrase tests. In clinical follow-up of 69 individuals for a median of 674 days, we did not find a difference in rates of virological failure between individuals with and without low-frequency mutations, although rates of virological failure were higher for individuals with drug-relevant low-frequency mutations. However, all 27 individuals who experienced virological failure reported poor adherence to their drug regimen during the preceding follow-up time, and all 19 who subsequently improved their adherence achieved viral suppression at later time points, consistent with a lack of clinical resistance. In conclusion, in a population with low antiviral resistance emergence, NGS methods detected numerous instances of minor alleles that did not result in subsequent bona fide virological failure due to antiviral resistance.

KEYWORDS:

AIDS; HIV; HyDRA; PASeq; amplicon sequencing; antiretroviral; antiviral resistance; deep sequencing; next-generation sequencing; resistance

PMID:
30305383
PMCID:
PMC6258839
DOI:
10.1128/JCM.01443-18
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for HighWire Icon for PubMed Central
Loading ...
Support Center