WHO HIVResNet meeting report: Johannesburg, South Africa, 11-12 November 2017

M EETING REPORT WHO HIVRESNET MEETING REPORT JOHANNESBURG, SOUTH AFRICA, 11–12 NOVEMBER 2017 HIV DRUG RESISTANCE

WHO HIVRESNET MEETING REPORT JOHANNESBURG, SOUTH AFRICA, 11–12 NOVEMBER 2017 WHO/CDS/HIV/18.11 © World Health Organization 2018 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial- ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-ncsa/ 3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization. Suggested citation. WHO HIVResNet Meeting Report. Geneva: World Health Organization; 2018 (WHO/CDS/ HIV/18.11). Licence: CC BYNC- SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/ licensing. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third- party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. Layout by minimum graphics 1CONTENTS Executive summary 2 1. Background and objectives 6 2. Landscape: integrase inhibitor resistance and considerations for optimal treatment sequencing strategies 7 3. Role of HIV drug resistance testing for clinical management in low- and middle-income countries 11 4. Point-mutation assays: technology landscape, priority mutations and prospects for point-of-care testing 12 5. Assessment of resistance testing to integrase inhibitors 14 6. Clinical significance of low-abundance drug-resistant variants 15 7. Rationale, advantages and disadvantages of next-generation sequencing in low- and middle-income countries: technology landscape and bioinformatics 16 8. DBS and HIV drug resistance 17 9. HIV drug resistance and PrEP 18 10. Research priorities for HIV drug resistance 20 References 21 Annex 1. List of participants 22 Annex 2. Meeting agenda 24 2EXECUTIVE SUMMARY Use and sequencing of dolutegravir (DTG) in low- and middle-income countries n DTG is a potent integrase inhibitor. DTG is highly potent for treatment-naive individuals, and little HIV drug resistance is expected to emerge if adequate adherence is maintained. n Data are lacking on DTG use in settings with limited or no viral load monitoring. Programmatic issues such as drug stock-outs can result in poor adherence and can subsequently lead to emergence of DTG resistance. n The substitution of DTG for efavirenz (EFV) for people already taking EFV-based first-line antiretroviral therapy (ART) is ideally accompanied by viral load testing. The risk of viral failure and the subsequent emergence of HIV drug resistance among people with viral suppression is likely to be minimal. However, for people with viral non-suppression, the probability of dual resistance to both tenofovir (TDF) and lamivudine (3TC) or emtricitabine (FTC) may be high (50% based on the TenoRes study). Because of lack of data on long-term viral suppression outcomes among people with TDF + 3TC resistance who are receiving DTG-based ART, switching to a ritonavir-boosted protease inhibitor (PI) is preferable. Alternatively, using DTG with an optimized backbone (based on the results of HIV drug resistance testing, if available) may be considered. n In settings in which viral load test results are unavailable at the time of a planned substitution of TDF + 3TC plus EFV to TDF + 3TC plus DTG, the most prudent choice is to wait for viral load test results to avoid a change to functional monotherapy or to use DTG with an optimized nucleoside reverse-transcriptase inhibitor (NRTI) backbone (Table 1). n Based on the available evidence, the genetic barrier to resistance of DTG appears to be higher than that of non-nucleoside reverse- transcriptase inhibitors (NNRTI) but lower than that of ritonavir- boosted PIs. However, this genetic barrier may not necessarily correlate directly with the drug’s effectiveness, and more research is needed to answer this question. Recommended approach to DTG transition for people receiving TDF + 3TC plus EFV Viral load Recommendation from HIVResNet <1000 copies/mla Stay on EFV-based ART or replace EFV with DTG (TDF + 3TC plus DTG) Unknown Stay on EFV until viral load is determined or use zidovudine (AZT) + 3TC plus DTG Unknown but clinically failing Switch to a PI plus AZT + lamivudine (3TC) (or a PI plus TDF + 3TC) or DTG plus AZT + 3TC >1000 copies/ml Switch to a PI plus AZT + 3TC (or a PI plus TDF + 3TC) or DTG plus AZT + 3TC a The group noted a potential issue among people with viraemia with viral load <1000 copies/ml and with K65R. The WHO HIVResNet met on 11–12 November 2017 in Johannesburg, South Africa. The following list summarizes major consensus points. 3n As DTG is introduced in low- and middle-income countries, viral and drug resistance outcomes need to be closely monitored; thus, HIV drug resistance surveillance remains critical to inform care and treatment guidelines and programme functioning. — Surveillance of pretreatment HIV drug resistance is important to establish baseline polymorphic profiles of the integrase gene in HIV-1 non-subtype B, which may be associated with DTG clinical response or non-response in various subtypes and populations. Ongoing assessment of NRTI pretreatment drug resistance will remain important and relevant both for its possible and as yet unknown impact on ART based on DTG plus two NRTIs and to measure the population burden of NRTI resistance caused by pre- exposure prophylaxis with TDF + 3TC. — Surveillance of acquired HIV drug resistance and robust nationally representative measures of viral load suppression and retention are critical to assess programme functioning with respect to DTG roll-out and viral and HIV drug resistance outcomes. Potential use of HIV drug resistance testing for managing people living with HIV in low- and middle- income countries WHO currently recommends using HIV drug resistance testing for people for whom second-line ART based on darunavir/ritonavir is failing after using DTG-based first-line ART. In the future, as HIV drug resistance testing becomes more widely available, its use should be linked to clearly definable actions by health-care workers. In addition, consideration should be given to whether or not the same action could be taken without resistance testing. A future role of HIV drug resistance testing could be considered in the following priority groups of people. Other potential groups were considered for future HIV drug resistance testing: (1) people starting DTG-based ART, to minimize functional DTG monotherapy; (2) people starting PI-based ART, to minimize functional PI monotherapy; and (3) people starting DTG-based ART with previous use of raltegravir (RAL) who may have acquired DTG resistance due to RAL exposure. Point-mutation assays and point-of-care assays n A point-mutation assay developed as a point-of-care assay would provide several theoretical advantages to people living with HIV, health-care providers and health-care programmes. n Pilot studies and clinical outcome data are needed to establish whether the potential advantages of point-mutation assays, implemented following a centralized point-of-care assay or near-point-of-care assay model, outweigh the disadvantages for various applications. n Because point-mutation assays detect only a subset of possible resistance mutations, they are unlikely to be suitable for surveillance applications in which the full-length sequence is needed to capture data on all mutations and to detect transmission networks. n Further research areas identified for point-of-care assays and point- mutation assays include: — impact assessment studies and operational research to guide the placement of point-of-care resistance tests; — evaluating training needs, developing monitoring and implementation guidelines for using point-of-care point- mutation assays in low- and middle-income countries, developing global quality assurance and quality control parameters for point-mutation assays and developing external quality assurance for validating assays; — considering sustainability, supply-chain management and market demands, especially in countries investing heavily in centralized HIV drug resistance testing; and — integrating viral load and drug resistance tests into point-of- Potential use of HIV drug resistance testing for managing people living with HIV in low- and middle-income countries ART scenario Purpose Failure of DTG-based ART To minimize unnecessary switching (if there is no integrase inhibitor drug resistance) Failure of PI-based ART To minimize unnecessary switching (if there is no PI drug resistance) Start of EFV-based ART To assess EFV resistance in countries with no access to DTG 4care point-mutation assays, in which the viral load is measured first and drug resistance is reflexively tested among individuals with viral failure. This testing algorithm could reduce costs, since most of the people living with HIV are expected to have viral suppression. HIV-1 integrase resistance testing n Increasing laboratory capacity within the WHO HIVResNet laboratory network to genotype the integrase region of HIV-1 for surveillance purposes is important, since integrase inhibitors are rolled out widely in low- and middle-income countries. n The minimum region of HIV-1 integrase to be genotyped for surveillance purposes was defined as codons 51 to 263. n Participants generally agreed that, when conducting surveillance, there is no cost benefit in excluding the protease region during sequencing, since data on PI resistance still needs to be accumulated. Next-generation sequencing n Within the WHO HIVResNet laboratory network, any laboratory considering a transition to next-generation sequencing should first assess the need. No laboratory needs to rush to implement it. Any assessment for readiness for transition to next-generation sequencing should include: testing volume, laboratory capacity and the development by the WHO HIVResNet of standard operating procedures and methods for HIVResNet labs. n Laboratories should not report variants presenting at less than 15% in a specimen using next-generation sequencing for the purpose of surveillance. Use of this standard threshold will facilitate data comparability between laboratories using Sanger sequencing and laboratories using next-generation sequencing. As additional data become available, this threshold may be revised. n Designation criteria and external quality assurance need to be developed for HIVResNet laboratories using next- generation sequencing for the purpose of generating data for HIV drug resistance surveillance. WHO will lead the development of these designation criteria. n A guidance document should be developed to elaborate standardized next-generation sequencing methods (wet laboratory and data analysis) to enable the comparability of results between laboratories. WHO is well positioned to lead this activity, and this document should build on the strengths and limitations of next- generation sequencing outlined above. A subgroup of HIVResNet laboratory members and researchers have begun this work. n Opportunity exists to engage in dialogue with manufacturers of next-generation sequencing to reduce costs. Dried blood spot (DBS) and HIV drug resistance testing n Plasma can be promoted as the gold standard; however, when use of plasma is not feasible, DBS may be used as specimens for an HIV drug resistance survey. However, when DBS are used for surveys, the survey sample size should be increased to account for an amplification failure rate that is larger than anticipated. n DBS are an important alternative to plasma for genotyping. WHO has developed guidance on collecting, processing and storing DBS. Although WHO’s standard operating procedures on DBS collection, processing, storage and handling are largely adequate, proper implementation in the field must be more strongly emphasized. Current guidance recommends that DBS be stored at room temperature with desiccant and humidity indicator for a maximum of 14 days from the day of collection to storage at –20 °C or –80 °C. Based on survey data, it was suggested that the guidance document can be strengthened by clarifying that DBS should be stored at the lowest temperature possible (–80 °C is preferred but –20 °C is acceptable if a –80 °C freezer is unavailable) for the shortest amount of time with minimal (and ideally no) additional freeze-thaw cycles. As new evidence becomes available on long-term DBS stability under various temperature and humidity storage conditions, time durations and transport conditions, the guidance document will be updated to reflect best practices. 5PrEP. Fear of high levels of HIV drug resistance resulting from PrEP use is therefore unwarranted and should not impede its implementation to prevent HIV infection. n People with undiagnosed acute infections have the highest risk of developing drug resistance with PrEP, highlighting the need for routine HIV testing before initiating PrEP, for people taking PrEP and for improved diagnostics with shorter window periods. For this reason, PrEP roll-out should be accompanied by careful monitoring for HIV drug resistance. n The Global Evaluation of Microbicide Sensitivity project is working to pilot HIV drug resistance surveillance in programmes using PrEP. WHO guidance for countries rolling out PrEP will be a valuable addition to its global strategy for surveillance of HIV drug resistance. Clinical significance of low- abundance drug resistance mutations The preliminary findings of a systematic review on the clinical significance of low-abundance HIV drug resistance mutations were presented. n Evidence from published studies is not sufficiently consistent to define a threshold of clinical significance of low-abundance NNRTI-resistant variants. n Evidence regarding NRTI and PI low-abundance mutations (below 10–20%) suggests they are not clinically relevant. n The clinical relevance of integrase inhibitor mutations present at low abundance is unknown. n Until the clinical relevance of low- abundance mutations has been established for all drug classes, data derived from next-generation sequencing should be interpreted using a Sanger-like threshold of 15%. This threshold may be adjusted if the comparability between Sanger-based sequencing and next-generation sequencing, which may be position-specific, has been studied. n The low-abundance working group of the WHO HIVResNet will complete the literature review, including a pooled analysis with patient-level data, to the extent possible. Research priorities on HIV drug resistance Through an online survey before the meeting, WHO HIVResNet members reviewed and ranked key research priorities for HIV drug resistance as defined in the Global Action Plan on HIV drug resistance for 2017–2021. The meeting participants re-ranked the research priorities as defined by the online survey. The research topics considered as the highest priority within the five-year plan were ranked as tier 1. Topics deemed less critical over the next five years were ranked as tier 2 (see Section 10). n Increased training at clinics collecting DBS is warranted, and protocols should be more explicit regarding instructions on collecting, processing, storing and handling DBS. n Alternative and novel methods of extracting RNA from DBS may be explored within the HIVResNet, which may increase the amplification of HIV RNA when present at low copy number and may further decrease interference from proviral DNA. n Future approaches that may reduce the storage of DBS in suboptimal conditions include locally performed RNA extraction and polymerase chain reaction (PCR) amplification followed by subsequent shipping of PCR amplicons for sequencing to regionally designated sequencing laboratories. This approach would require that WHO develop designation criteria for national laboratories only performing PCR amplification. HIV drug resistance and pre-exposure prophylaxis (PrEP) n Drug resistance has been rare in PrEP trials, because high drug pressure from PrEP makes infection unlikely to occur. When drug pressure is low or absent, such as from non-adherence to PrEP, the risk of drug resistance emerging is low. n Resistance from treatment failure of ART containing TDF + emtricitabine (FTC) will be far more significant than that generated by TDF + FTC used as 61. BACKGROUND AND OBJECTIVES becoming more commonly used for individual management in some countries. In addition, powerful methods such as next-generation sequencing and low-cost approaches, which test for a small number of specific mutations in target protein coding regions associated with drug resistance, can now be implemented in low- and middle-income countries. Thus, standardization of next-generation sequencing methods, frameworks to assess the quality of laboratories performing next-generation sequencing and quality assurance, thresholds for clinically interpreting next-generation sequencing outputs and other novel assays and guidance on using individual HIV drug resistance testing in the era of DTG-based regimens are required. The 2017 WHO HIVResNet meeting took place on 9 and 10 November in Johannesburg, South Africa. The meeting was held after the XXVI International Workshop on HIV Drug Resistance and Treatment Strategies, thus capitalizing on the presence of HIV drug resistance experts. The meeting had the following objectives: n to review integrase inhibitor resistance and develop considerations for optimal treatment sequencing strategies; n to assess the need for routine surveillance of pretreatment and acquired HIV drug resistance in the DTG era; n to consider the role of drug resistance testing in low- and middle-income countries for clinical management; n to review point-mutation assays (technology landscape, priority mutations and prospects for point-of-care testing) and their role in surveillance and/or clinical management in low- and middle-income countries; n to review laboratory requirements for HIV drug resistance genotyping for drug resistance testing of integrase inhibitors; n to consider the clinical significance of low- abundance drug-resistant variants; n to review next-generation sequencing technologies (rationale, advantages and disadvantages) and its use in low- and middle-income countries; The WHO 2016 consolidated guidelines on the use of antiretroviral (ARV) drugs for treating and preventing HIV infection (1) promote early treatment initiation and well tolerated regimens for people living with HIV. The 2016 guidelines include dolutegravir (DTG)-based HIV treatment as an alternative first-line regimen. Clinical trials have demonstrated superior efficacy of DTG compared with efavirenz (EFV) at the standard dose (600 mg/day) and ritonavir-boosted protease inhibitors (PIs) (2). DTG-based regimens are better tolerated and thus protect against treatment discontinuation caused by adverse drug reactions. DTG is associated with fewer drug–drug interactions, has a higher genetic barrier to resistance than that of EFV and is being launched as a low-cost, once-daily generic formulation for use in low- and middle-income countries. In addition to its use in first-line antiretroviral therapy (ART), recent clinical trial data suggest that DTG could be used as an alternative to standard second-line ART in combination with two nucleoside reverse-transcriptase inhibitors (NRTIs) (3). According to WHO’s 2017 HIV drug resistance report (4), the prevalence of resistance to non-nucleoside reverse-transcriptase inhibitors (NNRTIs) among people initiating ART in several countries exceeds 10%, a level above which WHO recommends using a non-NNRTI- based first-line regimen, such as DTG-based ART (5). During the next 24 months, DTG is expected to be rolled out in several low- and middle- income countries as a component of first- line regimens in combination with tenofovir disoproxil fumarate (TDF) and lamivudine (3TC). Drug-naive treatment initiators will probably use DTG-based regimens as will people switching from regimens containing NNRTIs, 3TC + emtricitabine (FTC) and TDF, with or without viral load information. This has raised questions about the extent to which DTG resistance will emerge and to what extent mutations conferring resistance to TDF and/ or 3TC + FTC will impact the efficacy and durability of DTG-containing regimens. Although the availability of HIV drug resistance testing for individuals in low- and middle-income countries remains very limited, HIV drug resistance genotyping is 7n to review the collection and storage conditions of dried blood spot (DBS) specimens for HIV drug resistance testing; n to review HIV drug resistance emergence with pre-exposure prophylaxis (PrEP) and the need for surveillance as PrEP is rolled out; and n to define key research priorities on HIV drug resistance to be considered over the next five years. The group consensus on the expected outputs (summarized in this meeting report) was: n optimal sequencing strategies from a HIV drug resistance standpoint; n considerations on the role of HIV drug resistance surveillance in the DTG era; n the possible role of drug resistance testing for clinical management in low- and middle-income countries; n use of point-of-care testing and point-mutation assays for clinical and surveillance purposes; n coding region for integrase inhibitors required for resistance testing; n clinical significance of low- abundance drug-resistant variants; n use of next-generation sequencing technologies in low- and middle- income countries; n collection and storage conditions for DBS specimens for HIV drug resistance testing; n HIV drug resistance surveillance with PrEP roll-out; and n key research priorities on HIV drug resistance over the next five years. 2. LANDSCAPE: INTEGRASE INHIBITOR RESISTANCE AND CONSIDERATIONS FOR OPTIMAL TREATMENT SEQUENCING STRATEGIES previously ARV drug-naive individuals (Spring-1, SINGLE and FLAMINGO). However, in 45% of the people with pretreatment resistance to integrase inhibitors (a result of prior exposure to RAL or EVG), treatment-emergent mutations while receiving DTG were detected. To date, one case of DTG resistance has been reported for a treatment-naive individual (6). Most data on mutations associated with DTG drug resistance come from RAL-experienced people who received DTG salvage therapy in the VIKING trials (7,8). In these trials, people whose viruses had a baseline Q148 mutation in combination with an E138 and/or G140 mutation were at increased risk of viral failure. The Q148 pathway appears to be the main springboard for the emergence of high-level DTG resistance because the selection of 1 or 2 of several other drug resistance mutations associated with integrase inhibitors, including polymorphic drug resistance mutations such as L74I/M and T97A, may result in much greater reductions in DTG susceptibility. Despite its low propensity to select for drug resistance mutations, an increasing number of reports of viral failure have been associated with integrase inhibitor resistance among ARV-experienced integrase inhibitor–naive people receiving a DTG-containing regimen (9,10) and among individuals with viral 2.1 Landscaping of integrase inhibitor resistance Integrase inhibitors have dominated developments in HIV treatment in recent years, with the approval of raltegravir (RAL), elvitegravir (EVG) and DTG. Because of its potency, efficacy in clinical trials and safety profile, DTG is being introduced in low- and middle-income countries. DTG is considered a second-generation integrase inhibitor and exhibits a higher genetic barrier to selection of resistance than RAL and EVG. DTG has a prolonged dissociation half-life (70 hours) that may contribute to its lower tendency to develop resistance. DTG demonstrates activity against some RAL- and EVG-resistant HIV, with common resistance associated mutations including Y143CHR, N155H and Q148HKR in the absence of secondary mutations. Up to four-fold declines in DTG susceptibility have been observed during serial passage of wild-type HIV in cell culture, leading to emergence of E92Q, G118R, S153FY, G193E and/or R263K. Virus with Q148HR developed additional mutations (T97A, E138K, G140S and M154I) during serial passage and exhibited 92–98% loss of DTG activity. To date, no DTG resistance has been reported in clinical trials involving 8suppression who switch to DTG only as maintenance therapy (11,12). R263K has been reported in five people in the first scenario. Q148H/R (four people), N155H (six people), G118R (two people), S230R (two people), and R263K (one person) have been reported following viral failure among people receiving DTG monotherapy (12,13). 2.2 Experience from countries transitioning to DTG Botswana and Brazil are in the process of transitioning to DTG-containing regimens. Their experiences are summarized below. 2.2.1 Botswana All ART-naive people living with HIV are being initiated on DTG-containing first-line regimens. As of November 2017, about 40 000 people have been initiated on DTG: 30 000 treatment- naive people and 10 000 switched to a DTG- containing regimen. Despite lack of conclusive data for safety for pregnant women and use for people coinfected with tuberculosis (TB) because of interactions with rifampicin, DTG is being used in these populations. With about 17 months of follow-up data available, all previously treatment-naive people receiving DTG with viral load data available have viral suppression, and 86% of the people on DTG salvage therapy have viral suppression. The rates of toxicity have been extremely low, with 50 of 30 000 people reporting toxicity, primarily gastrointestinal disturbances. Rare cases of immune reconstitution inflammatory syndrome associated with occult TB infection have been reported; the pregnancy outcomes are similar to those observed for mothers treated with EFV in the second and third trimesters. Botswana plans to switch everyone currently receiving first-line NNRTI-based regimens, ritonavir-boosted PI-based regimens and third-line RAL-containing regimens to DTG. The country is adopting a cut-off of 400 copies/ ml rather than 1000 copies/ml to define viral failure; priority is given to switch those whose current treatment is failing. In some cases, twice-daily DTG is being used for people for whom ritonavir-boosted PI is failing as a safety measure, despite lack data demonstrating superiority over once-daily DTG. Finally, people taking EFV-based first-line ART are not switched to DTG without measuring viral load (must be within three months of the planned switch date). People with viral suppression are switched to DTG; however, people with viral non-suppression have HIV drug resistance genotyping before switching, with an optimized NRTI backbone chosen based on the test results. 2.2.2 Brazil Brazil has initiated the roll-out of DTG. The country’s motivation to expand DTG use included increasing transmitted NNRTI resistance (9.5% nationally and reaching 11.2% in south-eastern Brazil), lower prevalence of adverse effects compared with NNRTIs and no budget impact compared with NNRTI-based treatment. To date, about 45 000 treatment-naive people have initiated DTG. About 17 000 people are receiving RAL-based third-line ART, and there is a plan to switch them to DTG. DTG is not being used among pregnant women or among people coinfected with TB pending finalization of global clinical trials. No decision has been taken on whether to switch people with viral load suppression on EFV to DTG. Overall the drug has been well tolerated: of 3000 people receiving DTG- containing regimens at 10 clinics, 124 stopped it because of side-effects. The predominant side-effects include gastrointestinal intolerance, sleep disturbances and central nervous system toxicity. 2.3 Treatment sequencing options and considerations DTG is being rolled out differently in different settings: (1) as a component of first-line treatment among ART-naive people; (2) in place of a NNRTI for people currently receiving NNRTI-based ART (substitution) and (3) as a component of second-line ART in combination with two NRTIs for people for whom first-line ART has failed (switching). As the clinical trial data above suggest, DTG in combination with two NRTIs (TDF + 3TC) is likely to be highly potent for treatment-naive individuals, and little HIV drug resistance is expected to emerge if adequate adherence is maintained. To date, WHO surveys of pretreatment HIV drug resistance have found low TDF or 3TC resistance. Failure to achieve viral suppression in this population would most frequently be anticipated to be caused by suboptimal adherence. Programmatically, switching from the standard first-line TDF + 3TC + EFV to TDF + 3TC + DTG for individuals with viral suppression is likely not to lead to treatment failure or select for HIV drug resistance. However, for people with 9confirmed viral failure on EFV-based ART, the HIVResNet members present at the meeting generally agreed that using a ritonavir-boosted PI-based regimen is preferred over using TDF + 3TC because of concerns that TDF and/or 3TC resistance could be present and the impact of TDF + 3TC resistance in this population is unknown. The prevalence of TDF resistance among people for whom TDF + 3TC plus EFV is failing is significant: up to 57% of people for whom TDF + 3TC plus EFV is failing in Africa have TDF resistance (14). No studies have yet answered the question of the possible impact of K65NR or M184IV on the long-term effectiveness and durability of first-line DTG-containing ART regimens. However, it is widely accepted that the presence of M184V reduces viral replication capacity, does not completely eliminate the antiviral activity of 3TC + FTC and increases susceptibility to zidovudine (AZT), TDF and stavudine (d4T). Thus, the presence of M184IV alone does not contraindicate continued treatment with 3TC or FTC. K65R causes an approximate two-fold reduction in susceptibility to TDF; when present with 184V, the reduction in susceptibility to TDF is reversed. Although residual NRTI activity (activity of TDF + 3TC) or reduced viral fitness is likely in K65R- and M184V-containing variants, which could provide some protection against the selection of DTG resistance, no clinical trials or observational studies have assessed their impact among people receiving this regimen. Resistance concerns for people switching from TDF + 3TC plus EFV regimens to a DTG-containing regimen in the absence of viral suppression are further underscored by data that do not support using DTG monotherapy for individuals who have achieved viral suppression on other regimens. In the DOMONO study, at week 48, higher levels of viral failure with DTG resistance were observed among those receiving DTG monotherapy than among those receiving triple ART (15). In addition, similar results were obtained in the DoluMono study (16). Taken together, these studies suggest that the genetic barrier of DTG monotherapy is insufficient to enable maintenance monotherapy, and further studies should assess particular risk factors associated with the failure of DTG monotherapy. Although the denominators are low, the frequency of viral failure with emergent resistance may be about 5% over 48 weeks (although participants in the DOMONO study were selected very carefully and people with higher viral loads and/or lower CD4 cell counts may develop viral failure and HIV drug resistance at higher rates). Table 1 summarizes the considerations on sequencing strategies from the HIV drug resistance standpoint. For people for whom first-line ART (TDF + 3TC plus EFV) is failing with viral load >1000 copies/ml, PI-based ART (mostly based on darunavir (DRV)) is considered the preferred option. In countries planning to use DTG in second-line ART, there were three possible scenarios: (1) DTG plus AZT + 3TC; (2) DTG plus TDF + 3TC; or (3) DTG twice daily plus TDF + 3TC. As discussed above, lack of data on the impact of TDF and or 3TC resistance, which may be present among 60% of the people for whom EFV-based first-line ART is failing (based on the 2017 TenoRes study and other studies in low- and middle- income countries), using DTG plus TDF + 3TC was felt to be potentially unsafe from an HIV drug resistance standpoint. Although using AZT + 3TC raised concerns about toxicity and side-effects such as peripheral neuropathy and anaemia, from an HIV drug resistance standpoint, AZT + 3TC plus DTG was assumed to be a safer and a more conservative option than TDF + 3TC plus DTG although its use with DTG has not been studied in clinical trials. Some clinicians have proposed using twice-daily DTG plus TDF + 3TC; however, clinical trial data supporting this option for people with NRTI resistance (K65R/M184V) are lacking. An unanswered clinical question among integrase inhibitor– experienced people with viral failure on a DTG-containing regimen is whether to maintain them on DTG, switch them or use DTG twice daily super-boosting or in combination with a new drug. Table 1. Recommended approach to DTG transition for people receiving TDF + 3TC plus EFV Viral load Recommendation from HIVResNet <1000 copies/mla Stay on EFV-based ART or replace EFV with DTG (TDF + 3TC plus DTG) Unknown Stay on EFV until viral load is determined or use AZT + 3TC plus DTG Unknown but clinically failing Switch to a PI plus AZT + 3TC (or a PI plus TDF + 3TC) or DTG plus AZT + 3TC >1000 copies/ml Switch to a PI plus AZT + 3TC (or a PI plus TDF + 3TC) or DTG plus AZT + 3TC a The group noted a potential issue among people with viraemia with viral load <1000 copies/ml and with K65R. 10 In addition to discussion on DTG, other ARV sequencing issues were discussed. n There is a need to monitor and understand whether to continue with the current strategy of using TDF + 3TC in first-line ART and AZT + 3TC in second-line ART or to switch back to TDF + 3TC in combination with ritonavir-boosted PI for second-line ART because of safety concerns related to AZT + 3TC (anaemia and peripheral neuropathy) and the effectiveness of ritonavir-boosted PI functional monotherapy (EARNEST study). n Using DRV/ritonavir plus TDF + 3TC could also be considered because of superior efficacy and removing toxicity concerns regarding Combivir® (AZT + 3TC); however, costs were noted to be prohibitive, with minimal possibility for reducing the price because of manufacturing issues. The group consensus was as follows. n DTG is a potent integrase inhibitor. DTG is highly potent for treatment-naive individuals, and little HIV drug resistance is expected to emerge if adequate adherence is maintained. n Data are lacking on DTG use in settings with limited or no viral load monitoring. Programmatic issues such as drug stock- outs can result in poor adherence and can subsequently lead to DTG resistance emerging. n Substituting DTG for EFV for people already taking EFV-based first-line ART is ideally accompanied by viral load testing. The risk of viral failure and the subsequent emergence of HIV drug resistance among people with viral suppression are likely to be minimal. However, in people with viral non-suppression, the probability of dual resistance to both TDF + 3TC may be high (50% based on the TenoRes study). Because of lack of data on long-term viral suppression outcomes for people with TDF + 3TC resistance who are receiving DTG-based ART, a switch to ritonavir- boosted PIs is preferable. Alternatively, using DTG with an optimized backbone (based on HIV drug resistance testing results, if available) may be considered. n In settings in which viral load test results are unavailable at the time of a planned substitution from TDF + 3TC plus EFV to TDF + 3TC plus DTG, the most prudent choice is to wait for viral load test results to avoid changing to functional monotherapy or to use DTG with an optimized NRTI backbone (Table 1). n Based on the available evidence, the genetic barrier to resistance of DTG appears to be higher than that of NNRTIs but lower than that of ritonavir-boosted PIs. However, this genetic barrier may not necessarily correlate directly with the drug’s effectiveness, and more research is needed to answer this question. n As DTG is introduced in low- and middle- income countries, viral and resistance outcomes need to be closely monitored; thus, HIV drug resistance surveillance remains critical to inform care and treatment guidelines and programme functioning. n Surveillance of pretreatment HIV drug resistance is important to establish baseline polymorphic profiles of the integrase gene in HIV-1 non-subtype B, which may be associated with DTG clinical response or non-response in various subtypes and populations. Ongoing assessment of NRTI pretreatment drug resistance will remain important and relevant both for its possible and as yet unknown impact on ART based on DTG + two NRTIs and to measure the population burden of NRTI resistance caused by pre-exposure prophylaxis with TDF + FTC. — Surveillance of acquired HIV drug resistance and robust nationally representative measures of viral load suppression and retention are critical to assess programme functioning with respect to DTG roll-out and viral and HIV drug resistance outcomes. n The WHO HIVResNet will convene a small working group in 2018 to review existing data on integrase inhibitor resistance and update the WHO surveillance drug resistance mutations list to include integrase inhibitor resistance mutations. 11 3. ROLE OF HIV DRUG RESISTANCE TESTING FOR CLINICAL MANAGEMENT IN LOW- AND MIDDLE-INCOME COUNTRIES At present, individual-level HIV drug resistance testing in low- and middle- income countries is rare. The meeting considered the role of individual-level HIV drug resistance testing in low- and middle-income countries and drug-sequencing strategies based on DR testing during the coming DTG era. The potential future role (over the medium to long term) of individual- level HIV drug resistance testing was considered and, entering the era of DTG scale-up, the six scenarios listed below were discussed. Group consensus: items 1–3 (bold) were given the highest priority for consideration. 1. HIV drug resistance testing among people for whom DTG- based ART is failing to prevent unnecessary switching to a different line of ART (detection of integrase inhibitor resistance). a. In the absence of any drug resistance mutations, continuing TDF + 3TC plus DTG or abacavir (ABC) + 3TC plus DTG would be recommended. In the presence of M184V, continuing TDF + 3TC plus DTG would probably be acceptable because M184V increases TDF activity. In the presence of K65RN and M184IV, continuing TDF + 3TC plus DTG would not be advisable, since continuing would be functional monotherapy and may be associated with substantial risk of viral failure and the emergence of resistance. When ABC is used in combination with 3TC and DTG, its continued use in the presence of M184IV would be ill advised, since M184IV reduces ABC susceptibility about three-fold, making the regimen nearly equivalent to functional DTG monotherapy. b. The clinical impact of the DTG- associated mutation R263K, selected for by people on DTG, is largely unknown but may be clinically relevant; more data are needed. 2. Second-line PI failures, to prevent unnecessary switches to third-line ART (if no PI resistance mutations are detected, the second-line ritonavir-boosted PI regimen is continued). 3. People initiating EFV in countries with no access to DTG but high (>10%) levels of pretreatment resistance to NNRTI on national surveys. HIV drug resistance testing identifies individuals with NNRTI resistance who would initiate non-NNRTI-based ART. 4. For EFV failures, detect K65R and assess whether DTG-based ART is indicated in a second-line regimen (rather than ritonavir-boosted PI). If K65R is present, individuals would initiate a ritonavir-boosted PI-based regimen. 5. Second-line ART failures: determine dosage for DRV/r and/ or DTG and optimize third-line treatment. 6. People initiating DTG: to assess for NRTI resistance in countries with high pretreatment resistance to TDF. 12 4. POINT-MUTATION ASSAYS: TECHNOLOGY LANDSCAPE, PRIORITY MUTATIONS AND PROSPECTS FOR POINT-OF-CARE TESTING A point-mutation assay assesses one or more specific codons in which resistance-associated mutations are known to occur rather than generating a full-length sequence of a portion of the genome. Most commonly occurring drug resistance mutations (both transmitted and acquired) in protease and reverse transcriptase have been explored in datasets exceeding 20 000 and 3000 people, respectively (17). At least one of the six of the following mutations will be present in 60% with transmitted drug resistance: 184V, 65R, 103N, 181C, 190A and 106M. At least one for the following mutations will be present in 99% of people with acquired drug resistance: 184V, 65R, 103N, 181C, 106M and 190A. Although large datasets have been analysed for the PI, NRTI and NNRTI drug classes, data remain limited for the integrase inhibitor drug class. The potential advantages of point-mutation assays compared with standard HIV drug resistance genotyping include: n increased sensitivity for low-abundance drug-resistant variants; n lower cost; n simpler procedure (less operator training required); n fewer equipment requirements; n faster turnaround time; and n enabling point-of-care or near-point-of-care applications. The possible disadvantages of point-mutation assays compared with standard HIV drug resistance genotyping include: n no information about drug resistance positions not included in the point- mutation assay ; n no possibility to perform phylogenetic analysis as a quality assurance check or to assess transmission networks; n at present, limited data available to guide the selection of priority mutations in integrase to include in a point-mutation assay; n the clinically relevant threshold for low- abundance drug-resistant variants is not defined for all drug classes; n genetic variability in individual virus sequences in primer and probe binding sites can make amplification and detection strategies difficult to standardize and add variability to assay sensitivity thresholds between individuals; and n new equipment required in laboratories already performing standard HIV drug resistance genotyping. Point-mutation assays currently being developed include: n oligonucleotide ligation assay (University of Washington, Seattle, USA); n pan-degenerate amplification and adaptation (PANDAA; Aldatu Biosciences, Cambridge, MA, USA); n allele-specific primer extension (United States Centers for Disease Control and Prevention); n multiplex melt curve analysis (InSilixa, Sunnyvale, CA, USA); and n others (such as in-house allele-specific polymerase chain reaction (PCR) assays). A point-of-care assay can be performed at a health centre, a clinic or a HIV testing site or in a regional laboratory closely associated with a clinic. The rationale for a point-of-care assay is to provide test results as quickly as possible for clinical management, thereby enhancing linkage to care and the use of effective drugs, reducing travel costs and work absenteeism and enabling rapid informed adherence counselling. Currently, assays being considered for use at or near the point of care are also point-mutation assays. Thus, although point-mutation assays could be used either at the point of care or in a centralized laboratory, the point-of-care assays currently being considered are all point- mutation assays. 13 For some clinical scenarios, the subset of information provided by a point- mutation assay done at the point of care is likely to be sufficient. These scenarios may include: n before starting therapy; n viral failure on a first-line NNRTI- or DTG- containing regimen to exclude non-adherence and choose an optimized NRTI backbone of a subsequent ritonavir-boosted PI- or DTG-based regimen; and n viral failure on a second-line regimen containing lopinavir/ ritonavir to exclude non-adherence and support the choice of optimized third-line ART. For other clinical scenarios, more data are required before point-of-care assays can be used. Considerable advances have been made in point- of-care assays, but most still require laboratory equipment and trained technicians. Nevertheless, point- mutation assays performed at or near the point of care in low- and middle- income countries could potentially be combined with point-of-care viral load testing. The group consensus was as follows. n A point-mutation assay developed as a point-of-care assay would provide several theoretical advantages to people living with HIV, health-care providers and health care programmes. n Pilot studies and clinical outcome data are needed to establish whether the potential advantages of point-mutation assays, implemented following a centralized point-of-care or near- point-of-care model, outweigh the disadvantages for various applications. n Because point-mutation assays detect only a subset of possible resistance mutations, they are unlikely to be suitable for surveillance applications in which the full-length sequence is needed to capture data on all mutations and to detect transmission networks. n Further research areas identified for point-of-care assays and point- mutation assays include: — impact assessment studies and operational research to guide the placement of point-of-care resistance tests; — evaluating training needs, developing monitoring and implementation guidelines for using point-of-care point- mutation assays in low- and middle-income countries, developing global quality assurance and quality control parameters for point-mutation assays and developing external quality assurance for validating the assays; — considering sustainability, supply-chain management and market demands, especially in countries that have made large investments in centralized HIV drug resistance testing; and — integrating viral load and resistance tests into point-of- care point-mutation assays, in which viral load is measured first and a drug resistance test is reflexively performed in individuals with viral failure, with this testing algorithm potentially reducing costs, since most individuals are expected to have viral suppression. 14 5. ASSESSMENT OF RESISTANCE TESTING CAPACITY TO INTEGRASE INHIBITORS The meeting reviewed the practical aspects related to genotyping the integrase gene. To date, the sequencing of HIV-1 reverse transcriptase and protease has been facilitated by the fact that they are contiguous portions of the same gene. The integrase gene is separated from reverse transcriptase and protease by a large fragment, the RNAse domain, which would not need to be sequenced for assessing HIV drug resistance. At present, most integrase inhibitor gene testing is done using in-house assays; few commercial kits are available. Most commercial or in-house assays use separate primer sets to amplify protease and reverse transcriptase and integrase inhibitors because of the long fragment size. This leads to a larger number of required sequencing primers and hence overall increased cost. The meeting discussed the possibility of using newer enzymes to generate amplicons of the full length of HIV-1 pol. It was noted however, that multiple sequencing primers would still be required, suggesting only limited cost savings. Moreover, DBS specimens are unlikely to be a suitable specimen type for large amplicons, especially at low viral loads. If DBS remain a commonly used specimen type, the amplification of reverse transcriptase and protease and the integrase regions of HIV-1 would need to be performed separately to ensure optimal assay success. Next-generation sequencing, if volume is sufficiently high, may save costs and primer issues are of less concern; however, primary library preparation increases costs, which cannot be reduced even with multiplexing. The group consensus was as follows. n Increasing laboratory capacity within the WHO HIVResNet laboratory network to genotype the integrase region of HIV-1 for surveillance purposes is important, since integrase inhibitors are being rolled out widely in low- and middle-income countries. n The minimum region of HIV-1 integrase to be genotyped for surveillance purposes was defined as codons 51 to 263. n When surveillance is conducted, participants generally agreed that excluding the protease region during sequencing has no cost benefit, especially since data on PI resistance still need to be accumulated. 15 6. CLINICAL SIGNIFICANCE OF LOW-ABUNDANCE DRUG-RESISTANT VARIANTS Development of next-generation sequencing technologies has enabled drug-resistant virus present at low abundance (lower that the approximate Sanger sequencing threshold of detection of 10–20%). The clinical significance of these variants has been an open research question for several years. Before the meeting, WHO formed an ad hoc working group to review the literature on the clinical impact of low-abundance drug resistance mutations on clinical outcomes. A qualitative review of the evidence was presented at the meeting. For PI and NRTI, available data show that low-abundance variants have no clinical significance. The clinical significance of low-abundance integrase inhibitor mutations is unknown and is an important research question. For NNRTIs, published data are inconsistent on whether low-abundance variants predict viral failure. Most studies examining low- abundance NNRTI variants are small (<100 participants) and use different study designs (often substudies, with assessment of the impact of low- abundance drug resistance mutations not being the primary focus), which complicates the interpretation of results. In addition, studies have used different methods to detect low- abundance drug resistance mutations and report using different thresholds and definitions of viral failure. The novel concept of mutational load was defined. Mutational load is the burden of drug-resistant virus as a function of viral load. For example, if viral load is 1000 copies/ml and the frequency of drug-resistant variants is 5%, the mutational load is 50 resistant viruses/ml; if the viral load is 1 000 000 copies/ml and the frequency of drug resistance virus is 5%, the mutational load is 50 000 resistant viruses/ml. Only a few studies have investigated the concept of mutational load on outcome, and additional research assessing this outcome may be warranted. The group consensus was as follows. n The evidence from published studies is not sufficiently consistent to define a threshold of clinical significance of low-abundance NNRTI-resistant variants. n The evidence regarding NRTI and PI low-abundance mutations (less than 10–20%) suggests that they are not clinically relevant. n The clinical relevance of integrase inhibitor mutations present at low abundance is unknown. n Until the clinical relevance of low-abundance mutations has been established for all drug classes, data derived from next- generation sequencing should be interpreted using a Sanger-like threshold of 15%. This threshold may be adjusted if studies of the comparability between Sanger- based sequencing and next- generation sequencing, which may be position-specific, have been conducted. n WHO HIVResNet’s low-abundance working group will complete the literature review, including a pooled analysis with patient-level data, to the extent possible. 16 7. RATIONALE, ADVANTAGES AND DISADVANTAGES OF NEXT-GENERATION SEQUENCING IN LOW- AND MIDDLE-INCOME COUNTRIES: TECHNOLOGY LANDSCAPE AND BIOINFORMATICS n the need for standardized approaches for validating assays; n the need for standardized and agreed operational procedures; n the need to understand the clinical relevance of low-abundance drug-resistant variants; and n limited data on the applicability of DBS as a specimen type for next-generation sequencing methods. The implementation of HIV drug resistance sequencing approaches based on next- generation sequencing in low- and middle- income countries may not always be optimal, since such considerations as cost– effectiveness, specimen testing volumes, throughput, turnaround time, equipment and reagent availability and personnel training may mean that standard Sanger-based procedures should be used for surveillance of HIV drug resistance. The British Columbia Centre for Excellence in HIV/AIDS is already conducting training for and transferring technology to several WHO HIVResNet labs in low- and middle-income countries, including both wet-lab panels from the Virology Quality Assurance Program of the United States National Institute of Allergy and Infectious Diseases as well as dry panels for data analysis. The Virology Quality Assurance Program has also piloted external quality assurance panels for next-generation sequencing, which have been sent to some WHO HIVResNet laboratories. The group consensus is as follows. n Within the WHO HIVResNet laboratory network, any laboratory considering a transition to next-generation sequencing should first assess the need. No laboratory should rush to implement next-generation sequencing. Any assessment for readiness to transition to next-generation sequencing Next-generation sequencing has emerged as an important tool in molecular genetics since it was developed and commercialized. Next-generation sequencing enables a parallel approach capable of producing large numbers of reads at exceptionally high coverage throughout a genome. The advantages of next-generation sequencing compared with standard sequencing that may prove useful for surveillance of HIV drug resistance in low- and middle-income countries include: n improved sensitivity and resolution for low- abundance variants; n improved scalability for large-scale testing; n improved time efficiency; and n decreased cost per specimen. Despite these advantages, the disadvantages include: n personnel training and infrastructure requirements: the upfront costs often exceed those of Sanger sequencing despite lower overall cost per specimen; n the need to develop laboratory PCR workflow for library preparation and challenges in designing molecular assays associated with high viral diversity; and n large data volume, which can be difficult to manage, a lack of uniformity in next- generation sequencing data mining and fully validated comparable pipelines and no standardized approach to validating assays. Several challenges are associated with introducing next-generation sequencing within the WHO network: n lack of uniformity of current next- generation sequencing platforms and bioinformatics pipelines; n large data volumes; 17 should include: testing volume, laboratory capacity and the development by WHO HIVResNet of standard operating procedures and methods for HIVResNet laboratories. n The group agreed that no laboratory should report variants present at less than 15% in a specimen using next-generation sequencing for the purpose of surveillance. Using this threshold will facilitate data comparability between laboratories using standard Sanger sequencing and laboratories using next-generation sequencing. As additional data become available, this threshold may be revised. n Designation criteria and external quality assurance need to be developed for WHO HIVResNet laboratories using next-generation sequencing to generate data for surveillance of HIV drug resistance. WHO will lead the development of these designation criteria. n A guidance document elaborating standardized next-generation sequencing methods (wet laboratory and data analysis) to enable the comparability of results between laboratories should be developed; WHO is well positioned to lead this activity, and this document should build upon the strengths and limitations of next-generation sequencing outlined above. A subgroup of HIVResNet laboratory members and researchers have begun this work. n Opportunity exists to engage in dialogue with next-generation sequencing manufacturers to reduce costs. 8. DRIED BLOOD SPOTS AND HIV DRUG RESISTANCE DBS have been shown to be a suitable alternative to plasma samples for HIV genotyping; however, HIV RNA in DBS is not stable if the DBS specimens are not collected, stored and transported properly. Standard operating procedures must be followed strictly, especially minimizing humidity and freeze-thaw cycles. At present, sample size estimates in most HIV drug resistance survey protocols account for 80% of successful PCR amplification. This may be an overestimation if DBS specimens are not treated carefully. Poor amplification success may lead to a smaller than desirable sample size and insufficient statistical power to robustly respond to major survey outcomes. Poor DBS specimen quality may lead to wasting reagents and time, increasing the costs and length of surveys, delaying the release of data and producing final results with less statistical precision than desired. DBS specimens may be less sensitive than plasma and, depending on the extraction techniques, few of which have been validated, there may be interference from proviral DNA. Nevertheless, despite these limitations, using DBS has major advantages that generally outweigh any disadvantages. These include lower cost of transport compared with plasma and enabling surveys of pretreatment drug resistance and acquired drug resistance to be nationally representative, something not possible with plasma, since not all ART sites can collect process and store plasma specimens. The group consensus is as follows. n Plasma can be promoted as the gold standard; however, when using plasma is not feasible, DBS may be used as specimens for an HIV drug resistance survey. However, when DBS are used for surveys, the survey sample size should be increased to account for an amplification failure rate that is larger than anticipated. n DBS are an important alternative to plasma for genotyping. WHO has developed guidance on collecting, processing and storing DBS. Although the WHO standard operating procedures for DBS collection, processing, storage and handling are largely adequate, proper implementation in the field must be more strongly emphasized. Current guidance recommends that DBS be stored at room temperature with desiccant and humidity indicator for a maximum of 14 days from the day of collection to storage at –20 °C or –80 °C. Based on survey data, it was suggested that the guidance document be strengthened by clarifying that DBS should be stored at the lowest temperature possible (–80 °C is preferred, but –20 °C is acceptable if a –80 °C freezer is unavailable) for the shortest amount of time with minimal (and ideally no) additional freeze-thaw cycles. As new evidence become available on long-term DBS stability under various temperature and humidity storage conditions, time durations and transport conditions, the guidance document will be updated to reflect best practices. 18 n Increased training at clinics collecting DBS is warranted, and protocols should be more explicit regarding instructions on collecting, processing, storing and handling DBS. n Alternative and novel methods of extracting RNA from DBS may be explored within the HIVResNet, which may increase amplification of HIV RNA when present at low copy number and which may further decrease interference from proviral DNA. n Future approaches that may reduce the storage of DBS in suboptimal conditions include locally performed RNA extraction and PCR amplification followed by subsequent shipping of PCR amplicons for sequencing to regionally designated sequencing laboratories. This approach would require WHO to develop designation criteria for national laboratories only performing PCR amplification. 9. HIV DRUG RESISTANCE AND PRE-EXPOSURE PROPHYLAXIS Prevention strategies that include ARV drugs have the potential for reducing HIV incidence. In 2012, a fixed-dosed combination of TDF + FTC became the first PrEP regimen approved for use by the United States Food and Drug Administration. The approval was based on data from Phase III HIV prevention trials that showed that seronegative individuals had a 44–75% reduction in HIV acquisition by using daily, oral TDF + FTC (18,19). Concurrently, WHO has recommended TDF with FTC or 3TC plus an NNRTI as the preferred regimen for first- line ART, creating concern that using the same drugs for treatment and prevention could increase HIV drug resistance. Breakthrough infection and subsequent selection of resistance with continued use of TDF + FTC PrEP during acute infection could compromise the effectiveness of first-line ART containing TDF + FTC. Conversely, the efficacy of TDF + FTC for PrEP could be reduced if the transmitted variant is from a partner failing a TDF + FTC–based ART regimen. FTC and TDF resistance selection is infrequent for HIV-1 seroconverters in active product arms from TDF + FTC PrEP trials. To date, five placebo- controlled, Phase III trials have assessed the effectiveness of daily oral TDF + FTC PrEP in preventing HIV infection in various populations, including men who have sex with men and transgender women in iPrEx (20), at-risk men and women in TDF2 (21), discordant couples in Partners PrEP (22) and women of reproductive age in FEM-PrEP (23) and VOICE (24). All studies included an active arm in which participants were assigned a once- daily regimen of oral TDF + FTC, and all participants underwent monthly rapid testing for HIV seroconversion. Only five cases of resistance with M184I/V or K65R were identified by standard genotype analysis in a combined total of 160 seroconverters assigned daily oral TDF + FTC in the five trials listed above. Four of these cases were in FEM-PrEP, but only one of the four is likely to have had M184V selection from PrEP failure despite adherence to TDF + FTC (intracellular concentrations of the active drug, tenofovir diphosphate, were equivalent to taking four or more tablets per week). Two of the three other cases of M184I and V were detected within four and eight weeks of study enrolment (one with detectable tenofovir diphosphate and one without, respectively) among individuals for whom acute infection at enrolment could not be ruled out. The fourth case was a participant who seroconverted 48 weeks after discontinuing TDF + FTC, and thus FTC resistance was likely transmitted from the participant’s partner. In all cases, the resistance mutation became undetectable after stopping the study drug: M184I by four weeks and M184V by 24–36 weeks (23,25), reflecting the negative effect of the 184 mutations on viral replication fitness. The fifth case of resistance occurred in VOICE: 1 of 61 participants who became infected on the TDF + FTC arm developed FTC resistance, with a mixed population of M184M/V. This participant had been received TDF + FTC 309 days from enrolment (68 days since the last negative plasma collection) and had detectable plasma tenofovir diphosphate within six weeks of the first quarterly plasma specimen collection (24,26). FTC resistance is frequently selected among participants with undetected acute infection at enrolment in active product arms from TDF + FTC PrEP trials. In contrast to the low frequency of TDF or FTC resistance among participants who became infected during the five PrEP trials, resistance selection was more frequent among those enrolled in the trials during the 19 seronegative window of acute HIV infection. Specifically, seven of 17 participants in the TDF+ FTC arms of the PrEP trials who had undetected acute infection at enrolment (seronegative with detectable HIV-1 RNA) had FTC resistance, with M184V or I, and one of the seven also had TDF resistance with K65R. For the two participants with FTC resistance in iPrEx, the M184V/I mutations waned to undetectable levels (<0.5%) by week 4 in one case and by week 14 in the second case and did not reappear through 52 weeks of follow- up. One participant with acute HIV-1 infection at enrolment into TDF2 had resistance to both TDF and FTC after continuing to take TDF + FTC for seven months after enrolment. M184V developed one month after study entry and A62V and K65R developed between 4–7 months, with all mutations present at high frequency. In Partners PrEP, two of the four participants on the TDF + FTC arm who were retrospectively found to be HIV infected at enrolment had >1% M184V. Finally, two of nine participants in VOICE who were enrolled during seronegative acute HIV infection developed FTC resistance, with M184V or M184I/V, after being on TDF + FTC for 26 and 29 days, respectively. Resistance to TDF is rarely selected among seroconverters from tenofovir PrEP trials. Tenofovir resistance has been infrequently detected among seroconverters in four relevant Phase III trials of products containing tenofovir only – CAPRISA 004 and VOICE for tenofovir gel and Bangkok Tenofovir, Partners PrEP and VOICE for oral TDF. No cases of K65R or K70E occurred in studies of 1% tenofovir gel, including 0 of 35 seroconverters in CAPRISA 004 and 0 of 60 in VOICE (23). In CAPRISA 004, low-frequency K65R was not detected in plasma collected within 30 days of estimated seroconversion or in vaginal swabs collected a median of 19 days from the time of infection (27). In VOICE, one seroconverter from the 1% tenofovir gel arm had 1.2% K65R detected by allele-specific PCR. The resistance results from FACTS 001 have not yet been reported (28). Resistance from oral TDF PrEP has also been rarely detected. In the Bangkok Tenofovir study, no cases of K65R or K70E were found in 17 seroconverters in the TDF arm, and in Partners PrEP, only one of 30 seroconverters on TDF and one of eight participants who enrolled with unrecognized acute infection had low-frequency K65R or K70E. In VOICE, no instances of K65R or K70E were detected among the 58 seroconverters on the TDF arm. Surveillance of HIV drug resistance during PrEP roll-out: the Global Evaluation of Microbicide Sensitivity project is monitoring resistance among PrEP seroconverters at selected sites in Kenya, South Africa and Zimbabwe. WHO plans to develop guidance, informed by this experience, for routine HIV drug resistance surveillance in all countries rolling out PrEP. The group consensus is as follows. n Resistance has been rare in PrEP trials, because when drug pressure from PrEP is high, infection is unlikely to occur. When drug pressure is low or absent, such as from non-adherence to PrEP, the risk of resistance emerging is low. n Resistance from treatment failure of TDF + FTC–containing ART will generate far more resistance than that generated by TDF + FTC used as PrEP. Fear of high levels of HIV drug resistance resulting from PrEP use is therefore unwarranted and should not impede its implementation to prevent HIV infection. n People with undiagnosed acute infection have the highest risk of developing resistance with PrEP, highlighting the need for routine HIV testing before initiating PrEP and for improving diagnostics with shorter window periods. For this reason, PrEP roll-out should be accompanied by careful monitoring for HIV drug resistance. n The Global Evaluation of Microbicide Sensitivity project is working to pilot HIV drug resistance surveillance in programmes using PrEP. WHO guidance for countries rolling out PrEP will be a valuable addition to its global strategy for surveillance of HIV drug resistance. 20 10. RESEARCH PRIORITIES FOR HIV DRUG RESISTANCE In July 2017, WHO launched a five-year Global Action Plan on HIV drug resistance. The Global Action Plan comprises five pillars: prevention and response, monitoring and surveillance, research and innovation, laboratory capacity and governance and enabling mechanisms. The consultation process for the Global Action Plan yielded a list of 45 research questions. The research priorities defined in the Global Action Plan were further set through discussion and voting by the WHO HIVResNet group during the meeting. The output of this priority-setting is provided below. Tier 1 research topics are considered to be of highest importance and should be given priority for study within the five years. Tier 2 topics are also important but were considered lower priority in the short and medium term. HIV drug resistance research gaps: epidemiology and clinical impact Tier 1 n Effect of pre-existing resistance to the NRTI backbone on the efficacy of DTG-based ART n Rates of viral suppression and prevalence and pattern of HIV drug resistance mutations among people for whom DTG- based ART has failed in low- and middle- income countries n Cost–effectiveness of individualized HIV drug resistance testing for people for whom a boosted PI + DTG is failing to minimize unnecessary switches to subsequent lines of ART n HIV drug resistance emerging in programmes scaling up PrEP n Impact of K65R/M184V mutations on the efficacy of TDF-based PrEP n Validated local, inexpensive and sustainable corrective actions to minimize the emergence and transmission of preventable drug resistant virus n Clinical impact of RAL-based ART among children harbouring resistance to the NRTIs n Clinical impact of DTG twice daily among children for whom RAL-based ART is failing n Optimal viral load switching algorithm to minimize the emergence of resistance n A simple HIV drug resistance interpretation algorithm for caregivers for individual clinical management Tier 2 n Efficacy of DTG twice daily as a strategy to overcome resistance in the co-administered NRTI backbone n Rates of viral suppression and acquired HIV drug resistance in populations on second- line boosted PIs in low- and middle-income countries, with particular focus on ATV/r n Response of TDF + 3TC in subpopulations at risk of treatment failure, including adolescents and people coinfected with TB and HIV n Clinically significant thresholds of low- abundance NNRTI-resistant variants n Cost–effectiveness analysis tools for use in countries for financing and advocating for optimized treatment HIV drug resistance research gaps: optimal technologies 1. Virological aspects of HIV drug resistance Tier 1 n Clearer correlation of genotype and phenotype and clinical response for all mutations n List of transmitted integrase inhibitor mutations n Minimum set of mutations for PI, reverse- transcriptase inhibitors and integrase inhibitors for clinical purposes for point- mutation technology 21 Tier 2 n Impact of novel drug delivery methods (such as long-acting drug formulations) on selection of HIV drug resistance 2. Innovative technologies Tier 1 n Simple and affordable point-of- care HIV drug resistance assays n Inexpensive, simple, easy-to- interpret tests that combine viral load and HIV drug resistance testing that can minimize unnecessary switches to subsequent regimens n Simple and affordable next- generation sequencing bioinformatics algorithms n Newer collection matrices for HIV drug resistance testing should be performed n Affordable, simple and easy to use point-of-care tests to measure drug levels to distinguish people for whom treatment is failing because of poor adherence versus drug resistance REFERENCES 1. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: recommendations for a public health approach – second edition. Geneva: World Health Organization; 2016 (http://www. who.int/hiv/pub/arv/arv-2016/en, accessed 18 June 2018). 2. Kanters S, Vitoria M, Doherty M, Socias ME, Ford N, Forrest JI et al. Comparative efficacy and safety of first-line antiretroviral therapy for the treatment of HIV infection: a systematic review and network metaanalysis. Lancet HIV. 2016;3:e510–20. 3. DAWNING Study: dolutegravir versus lopinavir/ritonavir as second- line treatment. Bethesda (MD): clinicaltrials.gov, United States National Institutes of Health; 2017 (https://clinicaltrials.gov/ct2/show/ NCT02227238?term=dolutegravir&rank=110, accessed 18 June 2018). 4. HIV drug resistance report 2017. Geneva: World Health Organization; 2017 (http://www.who.int/hiv/pub/drugresistance/hivdr-report-2017/en, accessed 18 June 2018). 5. Guidelines on the public health response to pretreatment HIV drug resistance. Supplement to the 2016 consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: second edition June 2016. Geneva: World Health Organization; 2017 (http://www. who.int/hiv/pub/guidelines/hivdr-guidelines-2017/en, accessed 18 June 2018). 6. Fulcher JA, Du Y, Sun R, Landovitz RJ. Emergence of integrase resistance mutations during initial therapy with TDF/FTC/DTG. 2017 Conference on Retroviruses and Opportunistic Infections, Seattle, WA, USA, 13–16 February 2017 (Abstract 500 LB; http://www.croiconference.org/sessions/ emergence-integrase-resistance-mutations-during-initial-therapy-tdfftcdtg, accessed 18 June 2018). 7. Eron JJ, Clotet B, Durant J, Katlama C, Kumar P, Lazzarin A et al. Safety and efficacy of dolutegravir in treatment-experienced subjects with raltegravir- resistant HIV type 1 infection: 24-week results of the VIKING Study. J Infect Dis. 2013;207:740–8. 8. Naeger LK, Harrington P, Komatsu T, Deming D. Effect of dolutegravir functional monotherapy on HIV-1 virological response in integrase strand transfer inhibitor resistant patients. Antivir Ther. 2016;21:481–8. 9. Lepik KJ, Harrigan PR, Yip B, Wang L, Robbins MA, Zhang WW et al. Emergent drug resistance with integrase strand transfer inhibitor-based regimens. AIDS. 2017;31:1425–34. 10. Cahn P, Pozniak AL, Mingrone H, Shuldyakov A, Brites C, Andrade- Villanueva J et al. Dolutegravir versus raltegravir in antiretroviral- experienced, integrase-inhibitor-naive adults with HIV: week 48 results from the randomised, double-blind, non-inferiority SAILING study. Lancet. 2013;382:700–8. 22 11. Blanco JL, Oldenbuettel C, Thomas R, Mallolas J, Wolf E, Brenner B et al. Pathways of resistance in subjects failing dolutegravir monotherapy. 2017 Conference on Retroviruses and Opportunistic Infections, Seattle, WA, USA, 13–16 February 2017 (Abstract 42; http://www.croiconference.org/sessions/pathways-resistance-subjects-failing- dolutegravir-monotherapy, accessed 18 June 2018). 12. Brenner BG, Thomas R, Blanco JL, Ibanescu RI, Oliveira M, Mesplède T et al. Development of a G118R mutation in HIV-1 integrase following a switch to dolutegravir monotherapy leading to cross-resistance to integrase inhibitors. J Antimicrob Chemother. 2016;71:1948–53. 13. Yoshinaga T, Kobayashi M, Seki T, Miki S, Wakasa-Morimoto C, Suyama-Kagitani A et al. Antiviral characteristics of GSK1265744, an HIV integrase inhibitor dosed orally or by long-acting injection. Antimicrob Agents Chemother. 2015;59:397–406. 14. TenoRes Study Group. The epidemiology of drug resistance after failure of WHO recommended first-line regimens for adult HIV-1 infection: a multicentre retrospective cohort study. Lancet Infect Dis. 2016; 16:565–75. 15. Rojas J, Blanco JL, Marcos MA, Lonca M, Tricas A, Moreno L et al. Dolutegravir monotherapy in HIV-infected patients with sustained viral suppression. J Antimicrob Chemother. 2016;71:1975–81. 16. Wijting I, Rokx C, Boucher C, van Kampen J, Pas S, de Vries-Sluijs T et al. Dolutegravir as maintenance monotherapy for HIV (DOMONO): a phase 2, randomised non-inferiority trial. Lancet HIV. 2017;4:e547–54. 17. Rhee SY, Jordan MR, Raizes E, Chua A, Parkin N, Kantor R et al. HIV-1 drug resistance mutations: potential applications for point-of-care genotypic resistance testing. PLOS One. 2015;10:e0145772. 18. Roehr B. FDA approves first drug to prevent HIV infection. BMJ. 2012;345:e4879. 19. Plosker GL. Emtricitabine/tenofovir disoproxil fumarate: a review of its use in HIV-1 pre-exposure prophylaxis. Drugs. 2013;73:279–91. 20. Grant RM, Lama JR, Anderson PL, McMahan V, Liu AY, Vargas L et al. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med. 2010;363:2587–99. 21. Thigpen MC, Kebaabetswe PM, Paxton LA, Smith DK, Rose CE, Segolodi TM et al. Antiretroviral preexposure prophylaxis for heterosexual HIV transmission in Botswana. N Engl J Med. 2012;367:423–34. 22. Baeten JM, Donnell D, Ndase P, Mugo NR, Campbell JD, Wangisi J et al. Antiretroviral prophylaxis for HIV prevention in heterosexual men and women. N Engl J Med. 2012;367:399–410. 23. Van Damme L, Corneli A, Ahmed K, Agot K, Lombaard J, Kapiga S et al. Preexposure prophylaxis for HIV infection among African women. N Engl J Med. 2012;367:411–22. 24. Marrazzo JM, Ramjee G, Richardson BA, Gomez K, Mgodi N, Nair G et al. Tenofovir-based preexposure prophylaxis for HIV infection among African women. N Engl J Med. 2015;372:509–18. 25. Grant RM, Liegler T, Defechereux P, Kashuba AD, Taylor D, Abdel-Mohsen M, et al. Drug resistance and plasma viral RNA level after ineffective use of oral pre-exposure prophylaxis in women. AIDS. 2015;29:331–7. 26. Parikh UM, Eskay K, Hardesty R, Kelly C, Magaret C, Molitor C et al. HIV-1 resistance outcomes in seroconverters from the MTN 003 (VOICE) Study. 2014 Conference on Retroviruses and Opportunistic Infections; Boston, MA, USA, 3–6 March 2014. 27. Wei X, Hunt G, Abdool Karim SS, Naranbhai V, Sibeko S, Abdool Karim Q et al. Sensitive tenofovir resistance screening of HIV-1 from the genital and blood compartments of women with breakthrough infections in the CAPRISA 004 tenofovir gel trial. J Infect Dis. 2014;209:1916–20. 28. Rees H, Delany-Moretlwe S, Lombard C, Baron D, Panchia R, Myer L et al. FACTS 001 Phase III trial of pericoital tenofovir 1% gel for HIV prevention in women. 2015 Conference on Retroviruses and Opportunistic Infections, Seattle,WA, USA, 23–26 February 2015 (http://www.croiconference.org/sessions/facts-001-phase-iii-trial-pericoital- tenofovir-1-gel-hiv-prevention-women, accessed 18 June 2018). 23 ANNEX 1. LIST OF PARTICIPANTS Name Country Name Country John Mellors USA Ava Avalos Botswana Karin Metzner Switzerland Wayne Barnes South Africa Jonathan Schapiro Israel Roger Paredes Spain Dan Kuritzkes USA Walter Campos Sweden Michelle Moorhouse South Africa Brian Chirombo South Africa Robert Shafer USA Keith Crawford USA Sergio Carmona South Africa Kim Steegen South Africa Andrea De Luca Italy Joshua DeVos USA Gillian Hunt South Africa Diana Dickinson Botswana Busisiwe Msimanga-Radebe South Africa Joe Fitzgibbon USA Lloyd Mulenga Zambia Pontiano Kaleebu Uganda Mary Natoli USA Lisa Frenkel USA Urvi Parikh USA Tendani Gaolathe Botswana Neil Parkin USA Katy Godfrey USA Andrew Phillips UK Marije Hofstra Netherlands Ana Flávia Nacif P. Coelho Pires Brazil Navin Horthongkham Thailand Romina Quercia USA Seth Inzaule Kenya Annette Sohn Thailand Michael Jordan USA David Raiser USA Cissy Kityo Uganda Deogratius Ssemwanga Uganda Rami Kantor USA Artur Ramos USA Johanna Ledwaba South Africa Jilian Sacks USA Barry Lutz USA Paul Sandstrom Canada Iain Macleod USA Kevin McCormick USA David Van de Vijver Netherlands Khanh Thu Viet Nam Frank Maldarelli USA Laura Broyles USA Gelareh Mazarei USA Carole Wallis South Africa Tamyo Mbisa United Kingdom Chunfu Yang Haiti Duping Zheng USA Mukhlid Yousif South Africa Siriphan Saeng-Aroon Thailand Michael Baffi USA Elliot Raizes USA Silvia Bertagnolio Switzerland 24 ANNEX 2. MEETING AGENDA Day 1: Thursday, 11 November 2017 Time Topic Presenter(s) 9:00 Welcome and introductions Silvia Bertagnolio 9:20 WHO HIV drug resistance report, guidelines and Global Action Plan on HIV drug resistance Silvia Bertagnolio 9:50 Dolutegravir use in low- and middle-income countries: approaches used in Botswana and Brazil Tendani Gaolathe (Botswana) and Ana Flavia Pires (Brazil) 10:20 Role of resistance testing in low- and middle-income countries for patient management and treatment sequencing strategies Jonathan Schapiro 10:50 Discussion (role of HIV drug resistance testing) All (moderator: Silvia Bertagnolio) 11:20 Break 11:35 Clinical relevance of genotypic data on dolutegravir Bob Shafer 12:05 Genotyping integrase: consensus of technical considerations Gillian Hunt 12:15 Discussion (integrase) All (moderator: Jonathan Schapiro) 12:45 Lunch break 13:45 Clinical significance of low-abundance drug-resistant variants Karin Metzner 14:15 Discussion (clinical significance) All (moderator: Dan Kuritzkes) 14:45 Point-mutation assays: technology landscape, priority mutations and prospects for point-of-care testing Bob Shafer 15:15 Point-mutation assays: practical considerations, assay validation, quality assurance and external quality assurance programmes Joe Fitzgibbon 15:35 Discussion (point-mutation assays) All (moderator: Michael Jordan) 15:55 Break 16:10 Next-generation sequencing: rationale, advantages and disadvantages of next-generation sequencing in low- and middle-income countries: technology landscape and bioinformatics Roger Paredes 16:30 Next-generation sequencing: practical considerations for low- and middle-income countries, assay validation, quality assurance and external quality assurance programs Paul Sandstrom 17:00 Discussion (next-generation sequencing) All (moderator: John Mellors) 17:30 Summary of the discussion: implications of using innovative tools for surveillance of HIV drug resistance All (moderator: Michael Jordan) 17:45 Adjourn 25 Day 2: Friday, 12 November 2017 Time Topic Presenter(s) 9:00 Day 1 recap and day 2 agenda Silvia Bertagnolio 9:15 WHO Global HIV Drug Resistance Laboratory Network status, integrase sequence assay validation, quality assurance and external quality assurance programmes Neil Parkin 9:45 Dried blood spots and HIV drug resistance testing: experience from surveys Artur Ramos 10:00 Discussion (Laboratory Network and DBS) All (moderator: Neil Parkin) 10:30 HIV drug resistance and PrEP: clinical considerations Urvi Parikh 10:50 PrEP implementation in low- and middle-income countries and considerations for surveillance of HIV drug resistance WHO South Africa 11:05 Discussion (PrEP) All (moderator: David Van de Vijver) 11:25 Break 11:40 HIV drug resistance research gaps: report from Global Action Plan consultations and United States National Institutes of Health meeting Silvia Bertagnolio, Katy Godfrey 12:00 HIV drug resistance research gaps: epidemiology and clinical impact All (moderators: Michelle Moorhouse and Roger Paredes) 12:45 Lunch break 13:45 HIV drug resistance research gaps: virological aspects and optimal technologies All (moderators: Ana Flavia Pires and Katy Godfrey) 14:15 Adjourn For more information, contact: World Health Organization Department of HIV/AIDS 20, avenue Appia 1211 Geneva 27 Switzerland E-mail: hiv-aids@who.int www.who.int/hiv