Next-Generation Sequencing for Detecting HIV-1 Drug Resistance: Ready for Clinical Diagnostics?

Resistance of Human Immunodeficiency Virus (HIV) to antiretroviral drugs is the most common cause for therapeutic failure in people infected with HIV. The ability of HIV to mutate and reproduce itself in the presence of antiretroviral drugs (HIV drug resistance or HIVDR) compromises the effectiveness of treatment and limits therapeutic options. According to “Global report on early warning indicators of HIV drug resistance” published by World Health Organization (WHO) in July 2016, “since 2011, monitoring and reporting of early warning indicators (EWI) of HIVDR has decreased globally. In 2015, only a handful of countries reported implementing EWI monitoring – a fact that may place many countries at serious risk of unknowingly creating situations favourable to the emergence of preventable HIVDR” [1]. However, almost all molecular diagnostics tests for HIVDR are currently off the in vitro diagnostics (IVD) market. The remaining commercial Sanger sequencing-based test dates from the early 2000s requires 1–2 weeks to produce results, and has low sensitivity for drug resistance mutations that occur at a frequency below 15–20% of total viral population.

We have developed first Next-Generation Sequencing (NGS)-based molecular diagnostics test (Sentosa® SQ HIV-1 Genotyping Assay) for detecting Drug Resistance Mutations (DRMs) in the HIV-1 reverse transcriptase, protease and integrase genes (Figure). The assay is a part of the automated integrated NGS workflow which is comprised of 1) a robotic liquid handling system for RNA extraction and NGS library preparation; 2) Ion Torrent’s instruments for deep sequencing; 3) kits for RNA extraction, HIV NGS library preparation and sequencing, and 4) data analysis and reporting software. Reporting includes 272 amino acid variants across the target regions. The turnaround time for the HIV NGS workflow is about 27 hours (or 2.5 x 8-hour working days) with hands-on time just about 3.5 hours. The assay is able to process up to 15 plasma samples simultaneously.

In our pilot study we compared performance of a Sanger sequencing-based (TruGene HIV-1 Genotyping Kit) and our NGS-based assay. 111 HIV-1 clinical samples were tested by both methods. In total, 634 drug resistance mutations were detected (199 mutations in the protease gene and 435 mutations in the reverse transcriptase gene). The NGS assay demonstrated very high sensitivity, detecting 100% of all DRMs in the protease gene compared to 90.45% detected by the TruGene, and identifying 98.16% of all DRMs in the reverse transcriptase gene compared to 74.48% identified by the TruGene. In total, 130 DRMs were detected by the NGS test and not found by the TruGene, while the TruGene only found 8 drug resistance mutations that our NGS assay missed. This discrepancy can be explained by heterogeneity of viral population and preferential annealing of HIV primers. Overall mutation detection rates aggregated were 98.74% for the NGS assay and 79.5% for the TruGene kit (Table).

Another advantage of our assay is the ability to detect drug resistance mutations in the HIV integrase gene that is not covered by other HIVDR assays. The integrase gene is becoming increasingly important as a drug target in the U.S.

Timely detection and reporting of DRMs is critical for drug regiment and can minimize the development of resistance to antiviral drugs. In this perspective the NGS-based workflow appears as a promising new tool for detecting clinically relevant variants in HIV-1. Given its high sensitivity (up to 5% mutation frequency) compared to Sanger sequencing-based systems and the comparatively short turnaround time of 2.5 days the workflow provides comprehensive, clinically relevant information for optimal selection of HIV treatment regimens. Such comprehensive HIV NGS workflow that can help fill some existing gaps in the quality of antiretroviral therapy services will take the necessary steps towards improving patient outcomes.

1. WHO. Global report on early warning indicators of HIV drug resistance: technical report, July 2016. 64pp.

Table. Mutation detection rates for the TruGene HIV-1 Genotyping and Sentosa® SQ HIV Genotyping Assays.

Figure. Regions of HIV-1 genome targeted by the Sentosa® SQ HIV Genotyping Assay.