Skip to main content

Collaborative HIV and Anti-HIV Drug Resistance Network

Final Report Summary - CHAIN (Collaborative HIV and Anti-HIV Drug Resistance Network)

Executive Summary:
Major advances in HIV management and care have significantly improved quality of life for HIV-infected individuals. However, a number of important challenges remain, such as the continued emergence of resistance against new and existing classes of antiretroviral drugs. In addition, the rate of new HIV infections in Europe continues to rise, demanding effective surveillance and monitoring to ensure appropriate allocation of resources. Moreover, due to the close relationship of worldwide HIV epidemics, it is important to improve the conditions of antiretroviral rollout in resource-poor regions such as Africa.
The Collaborative HIV and Anti-HIV Drug Resistance Network (CHAIN) was established in 2009 to provide an integrated response to these challenges via pooling of knowledge, resources and tools on a pan-European scale and in partnership with African institutions in Senegal and Cameroon. The CHAIN consortium comprises 26 partners, each with a high level of relevant scientific expertise and access to an extended peer network. Over the past five years, a collaborative research effort has focused on delivering the overall mission of the CHAIN programme, to ‘effectively and durably combat new and existing anti-HIV drug resistance in clinical settings, with special emphasis on Eastern Europe and heavily affected resource-poor regions in Africa.’
In pursuit of this mission, seven work streams were formed to address seven interrelated challenges identified by the CHAIN consortium. The aims of the work streams were to:

1. Identify novel mechanisms of resistance to existing drug classes (reverse transcriptase inhibitors & protease inhibitors) and to new classes of antiretrovirals (integrase inhibitors & entry inhibitors e.g.CCR5 antagonists).
2. Develop and validate optimal laboratory methods for predicting response to antiretroviral therapies, particularly in the presence of drug-resistant viruses.
3. Coordinate a clinical/virological data management system enabling assessment of the clinical implications of resistance.
4. Maximize the utility of HIV gene sequences and follow-up data through development of a bioinformatics HIV Data Sharing System.
5. Develop paradigms for linking resistance data to existing national HIV reporting structures.
6. Assess and describe the potential risks for resistance development with rollout of antiretroviral therapy in resource-limited settings.
7. Define and implement minimally required standards of scientific and clinical expertise in HIV drug resistance in Europe (mainly Eastern) and Africa.

A number of valuable advances have been made towards preventing further development and spread of anti-HIV drug resistance. The CHAIN consortium has developed new laboratory tools to predict and measure anti-HIV drug resistance, and implemented management strategies to reduce its impact and incidence. HIV gene sequence data have been incorporated into national HIV reporting structures and evidence-based recommendations have been made for public health and regulatory authorities, to help limit the emergence and transmission of HIV drug resistance. Some of the most significant achievements of the seven CHAIN work packages are discussed herein.
Project Context and Objectives:
Despite the huge advances made with HIV management and care, leading to a dramatic improvement in the quality of life for HIV infected individuals, there remain a number of significant challenges. Anti-HIV drug resistance represents a foremost threat to the effective use of HIV therapy in the developed and resource-poor regions of the world. Although the availability of new classes of antiretroviral drugs has dramatically reduced HIV-related mortality and morbidity, the emergence of HIV resistance in patients receiving Highly Active Anti-Retroviral Therapy (HAART), is now associated with increasing rates in disease progression.

As new classes of antiretroviral drugs are introduced, so the virus evolves new mechanisms to escape from inhibition. The implications of these genetic changes, both for the individual and the population, remain unknown. Insights into the susceptibility of viruses to antiretrovirals, and mechanisms underlying resistance development to both existing and novel drugs are essential to better clinically manage resistant viruses as these emerge, and to prevent further resistance development and spread. Antiviral therapy paradigms have been standardized during the last years, and remarkable results have been obtained. However the circulation of drug-resistant viral strains continues to be a major limitation of antiviral therapy. There are also major gaps in knowledge regarding drug resistance to existing drug classes, such as protease inhibitors. As continual virus escape from therapy occurs, there is a requirement to translate new knowledge of basic mechanisms into clinical practice. This requires the development of laboratory monitoring tools to best predict new resistance mechanisms, monitor resistance, and also ensure that clinical management of infected individuals remains optimal. The tools developed to investigate HIV resistance have historically been based on the B HIV-1 subtype (by far the most common HIV-1 strain circulating in North America). By contrast, non-B subtype viruses are responsible for more than 90% of worldwide HIV-1 infections and their prevalence is rising in Europe. Therefore, it is necessary to refine methodology, interpretation and the role of cross-resistance in B HIV-1 as well as in non-B HIV-1 and in HIV-2.

In addition to conducting studies to improve the treatment and care of infected individuals, it must be realised that the European epidemic is ever changing. The rate of new infections continues to rise, fuelled not only by within-Europe transmissions, but also through migration of infected individuals into Europe. Surveillance and monitoring of these shifts is essential in order that appropriate resources are allocated at country and European level to mitigate the adverse impact of the health of the population. New techniques of molecular epidemiology can be linked to established HIV surveillance structures, to optimise the real time monitoring of how the epidemic is changing. Of particular interest in this respect is the explosion of HIV amongst intravenous drug users and heterosexuals in Eastern Europe, and importation of infections from the endemic areas of sub-Saharan Africa. It requires the development of bioinformatics tools such as phylogenetic analysis to develop a global epidemiological model able to predict future trends. On a micro scale, the development of resistance within the host needs to be investigated, using tools such as data mining to predict the evolution of the virus under drug selective pressure. The most important information lacking is large pan-European datasets of temporal viral sequence and associated epidemiological and clinical data. Assimilation of real-time molecular epidemiology in the context of such national and international data provides a unique opportunity to map the growing epidemic, thus allowing for evidence-based interventions.

In view of the increasingly close relationship of worldwide HIV epidemics, it is important also to improve the conditions of antiretroviral rollout in the resource-poor regions of Africa. In particular the level of expensive laboratory monitoring required to optimise the World Health Organisation (WHO) rollout programme remains unclear. Modelling experiments, to advance understanding of the long-term impact of antiretroviral therapy rollout on treatment failure and drug-resistance in resource-poor regions, have the potential to produce highly informative data. Evidence-based recommendations can contribute to more effective strategies for public health and regulatory authorities. In the overall context of shifting epidemics, new interventions, and a European integrated approach to HIV therapy and drug resistance, there is an opportunity to guide new educational and drug licensing structures, in order to provide the optimal European response to the new HIV challenge. The degree of real-time monitoring itself will, in part, determine the likelihood of spread of drug resistance in the population. Further, the creation of a level playing field in the European and African scientific landscape in terms of scientific and clinical expertise in management of HIV drug resistance has been the ambitious focus of CHAIN.
The objectives of CHAIN:
• Identification of new mechanisms by which HIV escapes drug inhibition. This relates to new classes of drugs, such as integrase inhibitors, as well as existing classes of drugs.
• The development and validation of new and common laboratory tools for prediction and measurement of HIV drug resistance, particularly based on the mechanistic studies described above.
• Improved understanding of the clinical implications of drug resistance, with the development of optimal management strategies for transmitted and acquired HIV drug resistance, reducing the incidence and impact at population and patient level.
• A sustainable evidence base for better control of the epidemic and management of infected individuals. This requires new monitoring tools, and the development of systems for linking existing national HIV surveillance systems, together with the new science of molecular epidemiology, to better monitor HIV spread across Europe.
• Develop an evidence base to support antiretroviral rollout in Africa, and WHO initiated surveillance of HIV drug resistance in these settings.
• Provide an integrated educational framework for disseminating the findings from CHAIN research in an appropriate format. In addition, direct input into European Medicines Agency (EMA) post licensing surveillance guidance.
Project Results:
In pursuit of the CHAIN mission and in line with the above-mentioned challenges a range of main objectives and associated specific scientific, training and dissemination objectives were formulated and organised into seven work packages (WP1-7). The main objectives and key achievements and outputs of each work package are further detailed below.
WP1- NOVEL RESISTANCE MECHANISMS
Main objective: to complete gaps in our knowledge for resistance to existing drugs and identify novel resistance mechanisms to new drugs, preventing further development and spread.

Associated specific objectives:
• To identify novel mechanisms of resistance to existing drug classes (reverse transcriptase inhibitors (RTI) and protease inhibitors (PI)) and to new classes of antiretrovirals (integrase Inhibitors and CCR5 antagonists).
• To develop innovative and cost effective diagnostic approaches for measuring and monitoring susceptibility and resistance to CCR5 and integrase antagonists.
• To guide study on clinical relevance of newly identified drug resistance mechanisms.

The key achievements and outputs from WP1:
Drug-protein interactions
Due to its essential role in reverse transcription of HIV-1 RNA into DNA, the reverse transcriptase (RT) enzyme is a major target for antiretroviral therapy (ART). WP1 partners have reported novel findings on drug–protein interactions between HIV-1 RT and non-nucleoside RTI (NNRTI). The changes in protein conformation that occur following binding of the drug offer insights into the mechanisms underlying the development of drug resistance. Wright DW et al., J. Am. Chem. Soc., 2012;134:12885–12888; Wright DW et al., Biology 2012;1:222-244.
Genotypic and phenotypic assays for entry- RNAse H- and integrase inhibitors
A number of novel assays have been developed, including a drug susceptibility assay for HIV-2, and a genotypic assay to analyze the C-terminal domain of HIV-1 RT. A two-round phenotypic assay for protease inhibitor susceptibility testing of recombinant and primary HIV-1 isolates was developed (Puertas et al., 2012. J Clin Microbiol 3909–16), and a non-infectious cell-based phenotypic assay for the assessment of HIV-1 susceptibility to protease inhibitors (Buzon et al., J Antimicrob Chemother 2012; 67: 32–8). In addition, a quantitative Polymerase Chain Reaction (qPCR) method was developed to measure the viral load and replication capacity of mutant HIV-2 viruses, and a novel reverse-transcriptase (RT-PCR) strategy was established to sequence the connection and RNAse H domains of HIV-1 in plasma samples.
Novel HIV-1 resistance mutations
The majority of mutations conferring resistance to HIV-1 RTI have been identified within the polymerase domain of RT. Both the connection and RNAse H domains of RT contain key protein residues that are essential for RT structure and function, yet these are not routinely analysed in clinical samples. Partners involved in WP1 have sequenced the entire RT gene from a cohort of HIV-1 patients who were failing therapy with zidovudine (AZT), lamivudine or duo therapy regimens incorporating AZT. This allowed identification of novel resistance mutations outside of the polymerase region of RT, which have subsequently been introduced into reference strains for analysis of their effects on drug susceptibility and replication. As a result, the mechanisms involved in the selection of HIV-1 reverse transcriptase thumb subdomain polymorphisms associated with nucleoside analogue therapy failure were described (Betancor et al., 2010. Antimicrob Agents Chemother 54: 4799-811). Further, the clinical, virological and biochemical evidence was shown supporting the association of HIV-1 reverse transcriptase polymorphism R284K and thymidine analogue resistance mutations M41L, L210W and T215Y in patients failing tenofovir/emtricitabine therapy (Betancor et al., 2009. Retrovirol 9:68).
The group also demonstrated the role of specific substitutions in RT in association with patterns of mutations to some nucleoside reverse transcriptase inhibitors. In this regard, specific substitutions have been described as having a regulatory role for Leu-214 in the emergence and phenotypic resistance of the TAM1 complex, which includes Leu-41, Trp-210, and Tyr-215 (Puertas et al., 2009. J Virol 83: 7434–9).
New mutations to non-nucleoside reverse transcriptase inhibitors have been identified. Thus, the A376S substitution in the connection subdomain of HIV-1 reverse transcriptase has been shown to confer increased risk of virological failure to nevirapine therapy (Paredes et al., 2011, J Infect Dis 204:741–52).
A threonine for alanine substitution at position 400 in the connection domain of RT was found to occur with high frequency in patients failing therapy with RTI compared with treatment-naïve patients (Wright DW et al, PLOS One 2013; 8:e74078). The T400 mutation was shown to confer resistance to the NNRTIs nevirapine and efavirenz (Figure). Furthermore, T400 was shown to reduce RNAse H activity both on its own and when combined with other mutations, through conformational changes to the primer grip domain. It is suggested that the slower degradation of the viral RNA genome provides more time for dissociation of the NNRTI from the stalled RT template/primer, allowing reverse transcription to resume.

Within HIV-1 integrase, the group assessed raltegarvir susceptibility and fitness to progression of HIV-1 integrase in patients on long-term antiretroviral therapy (Buzon et al. 2008. Antivir Ther 13:881-93), and found that the HIV-1 integrase genotype strongly predicts raltegravir susceptibility but not viral fitness of primary virus isolates (Buzon et al. 2010. AIDS 24:17–25).

Novel HIV-2 resistance mutations
Using gene-sequencing techniques, a novel mutation – V111I – was identified in the polymerase region of RT in a cohort of patients with HIV-2 who were failing therapy with nucleoside RTI. This mutation was associated with mutations K65R and Q151M. Introduction of these mutations into reference strains showed that V111I both increases polymerase activity and increases the replication capacity of the K65R and Q151M viruses. Modelling experiments further revealed that V111I results in a catalytically competent RT enzyme that supports dNTP incorporation (Deuzing IP et al., manuscript in development).

Cell-based mechanisms of HIV-1 drug resistance
It has become increasingly apparent that the balance between drug influx and efflux transporter activity plays a critical role in the overall disposition of anti-HIV drugs in both cells and tissues. Thus, drug transporters directly influence the appearance of drug resistance and toxicity, and could also be related to persistence of HIV-1 (Minuesa et al., 2011. Pharmacol & Therap 132:268–79; Minuesa et al., 2008. J Pharmacol Exp Ther 324:558–67; Minuesa et al., 2009. J Pharmacol Exp Ther 329:252–61).

WP2- LABORATORY MONITORING TOOLS

Main objective: To develop and validate new and common laboratory tools for the prediction and measurement of HIV resistance.

Associated specific objectives:

• Develop and validate optimal laboratory methods for predicting response to antiretroviral therapies, particularly in the presence of drug resistant viruses.
• Identify appropriate uses for antiretroviral drug pharmacokinetics and resistance testing.
• Identify the clinical utility of HIV tropism and CCR5 antagonist resistance testing.
• Optimise interpretation algorithms of resistance data to guide therapy.
• Ensure optimal synergy of currently available quality assurance/quality control (QA/QC) systems for drug resistance testing.
Work package 2 was divided into four research tracks and involved the collaborative efforts of thirteen European laboratories. Some of the important research arising from these collaborations is discussed below.

Track 1: Standardization of laboratory techniques, QA/QC and new technologies
Two novel and highly sensitive assays were developed and validated for amplification and genotyping of the protease and reverse transcriptase (RT) genes from dried blood spots, which will be of particular value in resource limited setting (RLS) (Aitken SC et al., J Clin Microbiol 2013; 51:1757-61; Aitken SC et al., submitted for review). Further, a novel strategy for protease inhibitor susceptibility testing, based on the ability of the HIV protease to cleave cellular translation factors such as eIF4GI and PABP, was evaluated and validated using HIV-1 proteases from 46 clinical isolates. The assay showed good correlation with the Virco TYPE HIV-1 assay and the Antivirogram assay (Puertas MC et al., J Clin Microbiol. 2012; 50:3909-16).
Track 2: Minority species
Studies were conducted to assess the clinical value of ultrasensitive genotyping. In partnership with EU and US groups, the EU-FP7-funded CHAIN consortium showed that low frequency drug-resistant HIV-1 more than doubles the risk of virological failure to first-line NNRTI-based ART (Figure). This is the first time this question has been addressed using state of the art Next Generation Sequencing (NGS) technology in a well designed, multi-cohort, case-controlled study (Li JZ et al, JAMA 2011; 305:1327-1335; Li JZ et al, J Infect Dis 2013; 207:893-897; Cozzi-Lepri A et al, Lancet Infect Dis; submitted for peer-review). Given the rapid developments in NGS technologies and their expected future clinical application, the CHAIN NGS Working Group was formed, which will serve as a platform for scientific discussions around NGS issues in Europe and work towards standardization of NGS analyses.

Minority variants (MVs) detected and outcomes. Distribution of minority drug-resistant HIV-1 variants (left), prevalence of minority variants in cases and controls (upper right), and association with risk of virological failure (VF) (right). NRTI, nucleoside reverse transcriptase inhibitor; NNRTI, non-nucleoside reverse transcriptase inhibitor. Unpublished data.

Track 3: Tropism
The EMA and FDA mandate testing for viral co-receptor usage, or viral tropism, prior to administration of new antiretrovirals such as the CCR5 antagonist maraviroc. Track 3 standardized and improved tropism prediction using sequence and phenotypic analysis of the HIV env-V3 region (V3). This novel approach to tropism prediction was shown to be highly effective and safe for diagnostic use. Further research demonstrated that proviral DNA can be sequenced and used to reliably predict tropism. This can guide a switch to CCR5 antagonist-based therapy in patients with suppressed plasma HIV-1 RNA who need to change their current regimen (Vandekerckhove L et al, Lancet 2011; Bellecave P et al, CROI 2012; Abstract 716).

Track 4: Algorithms, subtypes and HIV
For the first time, mutations associated with resistance to HIV-2 were identified based on criteria established by a panel of European experts, and are now freely available at Charpentier, C et al, Clin Infect Dis. 2013; 56(11: 1654-8). In addition, a large cooperative study including CHAIN, COHERE and EUROCORD is underway to determine the contribution of different mutations to HIV-1 resistance to darunavir. Such studies are important to improve antiretroviral drug resistance-associated mutation interpretation and optimize combination therapy for HIV-infected individuals (Zazzi M et al, HIV Med 2011; Vercauteren, PLOS1 2013; e61436).



WP3- CLINICAL MANAGEMENT OF RESISTANCE

Main objective: to improve understanding of the clinical implications of drug resistance and develop optimal management strategies for transmitted and acquired HIV drug resistance, reducing the incidence and impact of resistance at a population and patient level.

Associated specific objectives:

• Coordinate a clinical/virological data management system enabling assessment of the clinical implications of resistance.
• Determine the impact of HIV drug resistance on virological and immunological response in treatment-experienced patients.
• Identify optimal strategies for treatment of patients with multi drug-resistant HIV.
• Identify the impact of transmitted drug-resistance (TDR) on response to therapy.


Through a collaborative research effort involving numerous research centres in Europe, WP3 has incorporated several different data sources into multiple studies, resulting in numerous publications in high impact journals. Results from just a handful of these important studies are discussed below.

Transmitted drug resistance
The effect of TDR on outcomes in the first year of combination antiretroviral therapy (ART) was assessed in a multi-cohort study including 10,056 patients. TDR was detected in approximately 5% of subjects. Those with TDR and resistance to at least one drug in the ART regimen were shown to be three times more likely to develop virological failure than those without TDR (Figure, Wittkop et al., Lancet Infect Dis 2011; 11:363–371). These findings support current treatment guidelines for HIV, which recommend selection of initial treatment based on resistance testing in treatment-naive patients.


Kaplan-Meier estimates of the proportion of patients with virological failure according to patient group. TDR, transmitted drug resistance; cART, combination antiretroviral therapy. Reprinted from The Lancet Infectious Diseases, 11, Wittkop et al., Effect of transmitted drug resistance on virological and immunological response to initial combination antiretroviral therapy for HIV (EuroCoord – CHAIN joint project): a European multi-cohort study, 363-371, Copyright (2011), with permission from Elsevier.

Minority variants
A systematic review conducted by a multi-national group of researchers (including WP2 and WP3 members) led by Harvard University reported low-frequency drug resistance mutations in 14% of participants (Li et al., JAMA 2011; 305:1327–1335). The presence of minority variants was associated with 2.5–3 times the risk of virological failure. Importantly, a dose-dependent increase in the risk of virological failure was observed in participants with a higher proportion or quantity of drug-resistant variants. A further study conducted in collaboration with WP2, WP4 and WP5, involving 6 European cohorts, found that the presence of at least one pre-existing minority variant in reverse transcriptase almost doubled the risk of virological failure to non-nucleoside reverse transcriptase inhibitor (NNRTI)-based ART. These data were presented at the ‘International Workshop on HIV and Hepatitis Virus Drug Resistance and Curative Strategies’ (Cozzi-Lepri et al., Antiviral Ther 2013; 18(Suppl1): A41) and have now been submitted for publication.

Viral subtype
Utilizing a large CHAIN cohort, the impact of viral subtype on responses to first-line therapy was assessed. Overall, HIV-1 subtype had no effect on virological and immunological responses. When restricting the analysis to adults, those without TDR had very similar responses to all patients combined (Wittkop et al., CROI 2013; poster presentation). In a collaborative project with EuroCoord and EPPICC, the prevalence of TDR in children was shown to be comparable to that found in adults. Younger age was associated with a higher risk of virological failure but not TDR (Wittkop et al. IAS 2013; poster TUPE306). Authors suggested that the effect of age may be related to fading of NNRTI resistance mutations in children exposed to single-dose nevirapine, Manuscripts discussing this data are currently being drafted.



WP4 - MOLECULAR EPIDEMIOLOGY AND BIOINFORMATICS OF RESISTANCE

Main objective: provide a sustainable knowledge base on HIV and HIV drug resistance for generation of data on a macro and a micro epidemiological scale to improve patient care, control the epidemic and prevent further spread of HIV drug resistance.

Associated specific objectives
• Maximize the utility of HIV gene sequences and follow-up data through development of a bioinformatics HIV Data Sharing System (HDSS).
• Improve clinical decision making for individuals with drug resistant viruses through web-based bioinformatics tools.
• Explain and predict the current and future spread of HIV variants throughout Europe and high prevalence areas such as Africa, with special attention to resistant virus variants.

Through expert collaboration on an international scale, partners involved in WP4 have developed a framework for merging large clinical HIV resistance databases such as EuResist, Eurosida and other national databases. These web-based 'super' datasets provide a tool to guide therapeutic decisions regarding optimal combination therapy for individual patients. These quality-checked datasets also provide a valuable resource for conducting molecular epidemiology and resistance studies.

Epidemiology
Large datasets have been generated for the most prevalent subtypes and recombinant HIV strains (subtypes A, B, C, D, G, and CRF01_AE, CRF02_AG) circulating in Europe and worldwide, including resource-limited settings (Abecasis et al, Retrovirology 2013; 10:7). Studies have shown that HIV epidemics are highly compartmentalized, and have also identified significant routes of HIV migration both within Europe and between Europe and other continents (Abecasis et al, Retrovirology 2013; 10:7; Lai et al, PLoS One 2012; 7:e42223; Lawyer et al, Med Microbiol Immunol 2012; 201:239-269, and several manuscripts in preparation). Another significant achievement of WP4 was the combined effort to analyse HIV subtypes in remote Russian regions and former Soviet Union (FSU) countries for the first time (Kazennova et al, Vopr Virusol 2013; 58:28-35; Laga et al, Vopr Virusol 2012; 57:26-32; Kazennova et al, Vopr Virusol 2011; 56:30-34). Several training programmes on HIV drug resistance have been developed and implemented in these regions.

Resistance
Studies conducted as part of WP4 have significantly advanced viral genotype technology, allowing for accurate estimation of viral resistance in individual patients (Thielen et al, Intervirology 2012; 55:113-117; Bogojeska et al, Stat Appl Genet Mol Biol 2012; 11:Article 11). Importantly, treatment response prediction engines that work without genotypic information have also been developed, to facilitate therapy optimization in RLS (Prosperi et al, PLoS One 2010; 5:e13753). Moreover, an in-house HIV-1 resistance test has proven reliable also at low levels of viraemia (Santoro et al, Clin Infect Dis 2014). Our understanding of the distribution and circulation of drug-resistant viral strains among B and non-B subtypes across Europe has vastly improved (Lai et al, PLoS One 2012; 7:e42223; Santoro M et al, AIDS Res Hum Retroviruses 2012;28:1285-1293), as has our awareness of the clinical implications of transmitted drug resistance (TDR). A study carried out in treatment-naïve patients indicated that TDR might increase, rather than decrease, viral fitness, leading to faster progression of disease in newly infected individuals (Theys et al, Retrovirology 2012; 9:81).


WP5 - EPIDEMIOLOGY AND SURVEILLANCE OF RESISTANCE

Main objective: incorporate HIV resistance data into national HIV reporting structures, creating a common surveillance system for monitoring HIV resistance and the dynamics of HIV spread in resource poor regions and Europe, enabling accurate defining and reporting of the emergence and transmission of HIV resistance.

Associated specific objectives:

• Develop paradigms for linking resistance data to existing national HIV reporting structures.
• Develop cost effective mechanisms for pooling HIV drug resistance data at national level.
• Demonstrate an ability to apply real time phylogenetics to assist in mapping local spread of resistant viruses.
• Determine the optimal laboratory methods for cost effective surveillance of HIV resistance in resource limited settings and algorithms for management of resistance.
• Develop implementation strategies for surveillance of HIV resistance in RLS.
• Estimate the burden of resistance within the population as a whole.

Surveillance of resistance
A key achievement of WP5 has been the development of structures and protocols for pooling clinical, HIV drug-resistance and socio-demographic data at a national level. Merging these datasets has allowed application of real-time phylogenetics to map the dynamics of HIV spread. In a Belgian cohort, cluster transmission analyses are currently being performed to evaluate the factors involved in the spread of resistance in local settings. In Russia, a national database of HIV drug resistance, including transmitted drug resistance mutations, has been constructed for the first time.



Resistance in treated populations
Progress has been made towards understanding the factors associated with virological and immunological failure following treatment with combination antiretroviral therapy (cART) in RLS. Adherence of <95%, as determined by pharmacy refill, was shown to predict both virological and immunological failure. This may help to identify patients for viral load monitoring and drug resistance testing in RLS.
Gupta RK et al., Lancet Infectious Diseases 2009; 9(7): 409-17; Ndembi N at al., J Infect Dis. 2010 Jan 1; 201(1): 106-13.

Several members of WP5 contributed to a report on retrospective prevalence and time-trends of HIV drug resistance in ART-exposed individuals in Western Europe between 1997 and 2008 (De Luca et al. J Infect Dis. 2013; 207:1216-1220). A significant reduction in HIV drug resistance was observed in more recent calendar years, despite increasing drug exposure. Whole ART sequencing strategies were thought to contribute significantly to these improvements.

A further key activity of WP5 was coordination a policy forum to determine how knowledge on HIV drug resistance can guide WHO HIV treatment guidelines for RLS. An expert workshop was organized to identify evidence-based strategies for minimizing the selection and transmission of resistance in RLS. These discussions have been summarized and published (Pillay D et al, Antiviral Therapy 2013; 18:831-6).

Transmission of resistance
As more people join antiretroviral therapy (ART) programmes in RLS, it is important to improve understanding of transmitted drug resistance (TDR) in these settings. With support from the Bill and Melinda Gates Foundation, a meta-analysis was published describing global trends in ART resistance in drug naïve individuals following ART rollout in RLS (Gupta et al, Lancet. 2012; 380:125–1258).

Estimates of TDR are based upon a consensus list developed by the WHO that includes mutations identified before ART became available. A higher prevalence of polymorphisms has been shown to reduce the likelihood of correctly estimating the prevalence of TDR (Frentz et al., J Acquir Immune Def Syndrome 2011; 58:e135-e137), which may lead to unnecessary adaptation of treatment regimens or implementation of costly steps to limit drug resistance. As such, an updated list of TDR mutations across viral subtypes is in development. A report detailing our current understanding of TDR mutations across subtypes is available: www.who.int/hiv/pub/drugresistance/report2012/en/

Through generation of a mathematical model, WP5 predicted the potential future impact of TDR on mortality and treatment outcomes in RLS. When TDR prevalence is low, the impact of its elimination was predicted to be small. However, when prevalence is high, the impact became substantial with increasing time of follow-up (Figure) (Cambiano V et al., Infect Dis 2013; 207:S57-S62). As such, it will be important to remain vigilant over transmission of drug-resistant HIV.


WP6: EVIDENCE-BASED PUBLIC HEALTH

Main objective: Translate clinical evidence into effective strategies and evidence-based recommendations for public health and regulatory authorities.

Associated specific objectives
• Assess and describe the potential risks for resistance development with rollout of antiretroviral therapy (ART) in resource-limited settings (RLS).
• Develop public health approaches to minimize development of resistance and its clinical consequences.
• Make recommendations and support EMA efforts to optimize guidance and strategies for use of ART to minimize drug resistance.
• Utilize contemporary epidemiological information on drug resistance to identify the potential utility of new antiretroviral drugs.
• Develop guidelines and strategies for post marketing surveillance to evaluate the resistance consequences of newly approved therapies.

Assessing risk of resistance
Work package 6 has significantly advanced understanding of the long-term impact of ART rollout in RLS on treatment failure and drug resistance. Overall, the data have been promising. A study in India reported good responses to first-line therapy for nearly four years (Neogi Uet al., PLoS One. 2013; 8:e55421) while in Senegal, the risk of virological failure following first-line ART for >5 years was low compared with high-resource settings (De Beaudrap P et al., J Acquir Immune Defic Syndr 2013; 62:381–387). In both studies, lower adherence with therapy increased the risk for treatment failure and drug resistance. Understanding these issues is important, as the risk of virological failure at 24 months was shown to be high in the Senegal cohort after switching to second-line therapy.

Modelling experiments have also produced highly informative data. A collaborative project with the WHO demonstrated that using tenofovir, rather than zidovudine, as a component of first-line therapy in RLS is cost-effective and more likely to preserve future treatment options in the absence of virological monitoring (von Wyl et al., PLoS One 2012; 7:e42834). Mathematical modelling further indicated that pre-exposure prophylaxis (PrEP) could be used as a cost-effective approach for preventing new HIV infections in RLS. Moreover, good adherence with therapy was predicted to reduce the risk of developing drug resistance if individuals became infected despite use of PrEP (Nichols et al, PloS One 2013; 8:e59549).

Public health approaches, guidelines and recommendations to minimize risk of resistance
A data and literature analysis carried out by WP6 concluded that detection of virologic failure within 12 months of initiating ART prevents the development of complex drug-resistance patterns and carries a low risk of compromising second-line treatment options in RLS. Since improved access to HIV drug resistance tests is clearly important in these settings, a simple and affordable assay was developed to detect phenotypic resistance to etravirine (ETR) and cross-resistance to other NNRTIs (Agneskog et al, J Med Virol 2013; 85:703-708).

Research carried out as part of WP6 contributed to a report on the WHO’s HIV Drug Resistance Prevention and Assessment Strategy. As of June 2011, 52 countries had implemented at least 1 element of the strategy, and 27 laboratories had been accredited for HIV drug resistance genotyping to support assessment of acquired and transmitted HIV drug resistance in RLS (Figure) (Jordan et al, Clin Infect Dis 2012; 54:S245–249).

Based on data collated during the work programme, recommendations have been submitted to the EMA concerning the clinical development of medicinal products for the treatment of HIV, as well as the development of oral and topical HIV PrEP. In addition, numerous training events on HIV drug resistance have been developed and delivered to clinicians and epidemiologists in RLS.


WP7- TRAINING AND CAPACITY BUILDING

Main objective: create a level playing field in the European and African scientific landscape in terms of scientific and clinical expertise in management of HIV drug resistance.

Associated specific objectives

• Define and implement minimally required standards of scientific and clinical expertise in HIV drug resistance in Europe (mainly Eastern) and Africa.
• Improve the scientific and clinical expertise of clinical practitioners, diagnosticians, epidemiologists and policy makers in HIV drug resistance, in order to reduce the adverse impact of HIV drug resistance.


Over the past 5 years, CHAIN partners have organized and implemented almost 40 workshops on HIV drug resistance throughout Europe and Africa, greatly exceeding the expectations of the work package. During this time, partners involved in WP7 continually updated the training material for each of the training categories, to ensure inclusion of the latest developments and clinical data.

Highly regarded experts in the field of HIV drug resistance delivered courses in a variety of different formats, including face-to-face workshops, online presentations and discussion of clinical cases. Attendees included clinicians, epidemiologists, diagnosticians and policy makers.
Examples of workshops organized and successfully implemented as part of WP7

• Workshops in several African countries on ‘Early Infant Diagnosis of HIV’, including courses on ‘Clinical Management’, ‘Epidemiology and Surveillance’ and ‘Monitoring Tools’.
• Courses on ‘Advanced Diagnostics’ and ‘Antiviral Resistance in Europe’.
• Two- and four-day training courses on HIV resistance for clinicians, epidemiologists and medical officers in several Eastern European countries.
• Workshops on Drug Resistance in Africa, Asia and South America.
• A course on implementation of strategies for HIV RNA monitoring in resource-limited settings.
Potential Impact:
WP1- NOVEL RESISTANCE MECHANISMS
The focus of WP1 has been on expanding current knowledge on the mechanisms underlying resistance to existing and new antiviral drugs. The insight gained through WP1 research is essential to better clinically manage resistant viruses as they emerge, and to prevent further resistance development and spread. Understanding the genetic basis of resistance and cross-resistance is crucial for optimizing the use of existing drugs and for designing new antiviral agents and new therapeutic approaches.

The overall impact of the work package:
- Improved clinical management of drug resistant strains through improved understanding of drug-protein interactions.
- Detection of new resistance mutations using novel genotypic and phenotypic assays.
- Optimization of therapy and management of drug resistant strains via improved understanding of resistance mutations.
Remaining challenges and future research
• Continue to assess the effects of newly identified HIV-1 resistance mutations in combination with other NNRTI resistance mutations on drug susceptibility and replication.
• Determine the extent to which connection and RNase H domain mutations impact patient outcomes in HIV-1.
• Elucidate the impact of the polymerase V111I mutation in HIV-2.


WP2- LABORATORY MONITORING TOOLS
The general challenge of WP2 was to improve and validate new monitoring tools for the prediction and measurement of HIV resistance. Activities carried out as part of WP2 have helped to counter HIV drug resistance and impact the fight against HIV infection in RLS by:

- Developing simplified, cost-effective genotyping protocols and methods for handling clinical specimens, thereby improving access to HIV drug resistance genotyping information capable of optimizing therapeutic decisions.
- Extending knowledge on resistant minority species, providing the scientific basis for improved genotypic testing during the next decades.
- Standardizing genotypic tropism prediction and improving analysis of clinical samples with very low or undetectable viral load.
- Establishing rules for interpretation of genotypic HIV-2 resistance and providing free access to the list of identified HIV-2 resistance mutations.
- Refining the methods for establishing algorithms for genotype resistance interpretation.

Studies conducted to assess the clinical value of ultrasensitive genotyping showed that low frequency drug-resistant HIV-1 more than doubles the risk of virological failure to first-line NNRTI-based ART. This is the first time this question has been addressed using state of the art Next Generation Sequencing (NGS) technology in a well designed, multi-cohort; case-controlled study of several hundred HIV-1 infected individuals throughout Europe. It provides the scientific basis for improved genotypic testing during the decades to come. Further, work towards the standardization of NGS analyses (through establishing minimum technical and bioinformatics standards) should enable the development of tools to perform inter-laboratory QA/QC tests.

For the first time a list of mutations associated with HIV-2 resistance has been published and an automated tool for HIV-2 drug resistance analyses made freely available on the web: http://www.hiv-grade.de A panel of European experts voted on a rule set for interpretation of mutations in HIV-2 protease, reverse transcriptase, and integrase. The mutations listed here have been identified by 1 or more of the following criteria: (i) in vitro passage experiments or validation of contribution to resistance by using site-directed mutagenesis; (ii) susceptibility testing of laboratory or clinical isolates; (iii) nucleotide sequencing of viruses from patients in whom the drug is failing. This list of HIV-2 resistance mutations will be updated each year and a meeting in 2015 is planned.

In a broader context, the achievements of WP2 are many. They contribute to the possibility of generating and collecting HIV DR genotyping information from HIV-1 infected individuals in RLS, as has already been demonstrated in Uganda and South Africa. Having established a novel strategy for protease inhibitor susceptibility testing (using a cost-effective molecular-based assay) the potential is to optimise therapeutic decisions – particularly when the establishment of salvage regimens with protease inhibitors is required.

Future challenges for work related to WP2 are listed below:

• Update the list of HIV-2 resistance mutations on a yearly basis.
• Continue quality control assessment of the tropism prediction assay with a specific focus on non-B HIV subtypes.
• Continue the highly valuable activities of the NGS working group.

WP3- CLINICAL MANAGEMENT OF RESISTANCE
After five years of work the contribution of WP3 has improved our understanding of HIV drug resistance, particularly the clinical impact of transmitted drug resistant variants as well as the role of minority drug resistant variants on the virological outcomes of first-line and salvage ART. Tasks undertaken have set the clinical standard for drug resistance detection methodologies that have a direct impact on treatment strategies in high- and low-income settings. Also, studies undertaken in collaboration with WP2 and WP4 as well as with US research centres have allowed us to finally establish the relevance of the minority variants, which was an issue completely unknown when this grant started.

Work conducted as part of WP3 has facilitated the implementation of optimal therapeutic strategies that limit the spread of drug resistant variants in RLS and elsewhere by:

- Establishing a clinical standard for drug resistance detection methodologies in high- and low-income settings.
- Understanding of the clinical impact of TDR variants on the virological outcomes of first-line and salvage ART.
- Understanding, for the first time, the relevance of minority variants on the virological outcomes of first-line and salvage ART.
- Assessing the impact of viral subtype on responses to first-line therapy in adults and children.
A key future challenge in the fight against HIV and HIV drug resistance is to transfer expensive analytical systems to low-income settings. Improving or modifying technologies so that they can be implemented in a cost-effective manner may achieve this.

WP4 - MOLECULAR EPIDEMIOLOGY AND BIOINFORMATICS OF RESISTANCE
The main aim of WP4 is to provide a sustainable evidence base for better control of the HIV epidemic and management of infected individuals through development of comprehensive macro and micro-epidemiological bioinformatics tools that can be used to predict epidemiological trends and prevent any further expansion of HIV drug resistance.

Through the collaborative efforts of the work package a framework for merging large clinical databases into "super" datasets has been established. EuResist is already a very large resistance database, but can now be further expanded by clinical dataset like Eurosida and other national databases. These web-based "super" datasets provides Europe with a clear competitive advantage in large-scale research on several aspects of HIV infection, including resistance and epidemiology. They have paved the way forward for studies on HIV spread within Europe. The work is on going but has already provided new and deeper insights in to HIV epidemiology, which will ultimately lead to greater HIV prevention.

The overall impact of the work package:

- Provision of a platform for large-scale research on HIV drug resistance and epidemiology.
- New epidemiological information to guide design of HIV infection prevention programmes, in Europe as well as in RLS where these data were previously very limited.
- Improved understanding of how HIV resistance tests should be interpreted and availability of free, online treatment response prediction engines for everyday use in HIV clinics worldwide.
- Improved selection of the most effective drug combination therapies for individual patients using these web-based resources.
- Improved education and awareness of HIV drug resistance among doctors and epidemiologists through development and implementation of training programmes, particularly in former Soviet Union countries.

The impact and practical implications of WP4 findings can be summarized under the following subheadings:

Epidemiology
Large datasets have been generated for the most prevalent subtypes and recombinant HIV strains circulating in Europe and elsewhere in the world, including RLS. The results showed the global pattern of the spread of the epidemic e.g. areas acting as sources or sinks for viral infections, and in RLS. The expectation is that the data can be used to design interventions to prevent new infections. Other studies have shown that HIV epidemics are highly compartmentalized, and have also identified significant routes of HIV migration both within Europe and between Europe and other continents. Another significant achievement of WP4 was the combined effort to analyse HIV subtypes in remote Russian regions and former Soviet Union (FSU) countries for the first time. Several training programmes on HIV drug resistance have been developed and implemented in these regions.

In the broader context WP4 has aided in the development of improved tools for interpretation of HIV resistance that are freely accessible and thereby used all over the world. The development of a prototype system for predicting response to treatment in the absence of HIV genotype information is of significance in RLS where resistance testing might be difficult to incorporate in routine patient care. There is now better understanding of the predictors of adherence in RLS and the opportunity to build treatment response prediction engines that can work with or without genotypic information. The work programme was also first to combine the effort and expertise in HIV genotyping between all Russian-speaking FSU countries most of which are RLS.

Resistance
It has been seen that transmitted drug resistance of single mutations (especially NRTIs) originate predominantly from other TDR patients, rather than from treated patients. This could have a big influence on choices of first line regimen in the face of TDR. Evidence that transmission of drug resistance has the potential to increase viral fitness (rather than reduce it), due to preferential reversal of drug resistance mutations that reduce viral fitness, while compensatory mutations remain stable, is a valuable insight in to the mechanism of HIV resistance.


Future challenges

• Evaluate how progress in antiretroviral treatment and continuous human migration affect development and transmission of the different drug-resistant HIV strains worldwide.
• Use new data generating technology to track HIV spread and epidemiology to inform HIV education and prevention programmes.
• Use human genetic data to select candidates for HIV eradication pilot studies and to improve efficacy, safety and cost-effectiveness of antiretroviral treatment.
• Improve understanding of behaviour determinants of HIV transmission chains, to help guide prevention measures.
• Study the contribution of host factors on HIV infection treatment.
• Disseminate new knowledge effectively, including development of continuous medical education programmes for specialists in HIV/AIDS.


WP5 - EPIDEMIOLOGY AND SURVEILLANCE OF RESISTANCE
The focus of WP5 has been on assimilating the empiric data produced previously within the consortium to contribute to modelling. These approaches have been used to study the potential impact of transmitted drug resistance in the developing world, particularly regarding the trigger for change of antiretroviral guidance. Further, such approaches have been extended to consider the issue of treatment as prevention- a prevention paradigm that has rapidly gained credence during the existence of CHAIN. The work has focused not only on adults but also children. Despite major success in reducing mother to child transmission, the problem of treatment options for infected children remains high.

Finally, CHAIN has facilitated a number of country-wide studies on prevalence of resistance in treated and untreated individuals, which is reassurance that transmitted drug resistance is remaining low in a European context. This is not the case in the generalised epidemics in southern Africa, however.

Overall impact of the work package
- Improved selection of patients for viral load monitoring and drug resistance testing.
- Improved access to drug resistance testing in RLS.
- Identification of strategies to reduce drug resistance in RLS based on findings in developed countries.
- Potential to optimize treatment regimens using updated list of TDR mutations.
- Greater understanding of the potential impact of TDR in RLS.
Future challenges
• Continue to update the list of TDR mutations across viral subtypes.
• Introduce whole ART sequencing to RLS to help reduce prevalence of resistance to ART.
• Continue efforts to understand the dynamics of TDR spread, in an effort to reduce its prevalence and the associated mortality risk.

WP6: EVIDENCE-BASED PUBLIC HEALTH
The final year of CHAIN witnessed the increasing linkage of WP5 and WP6 activities, as the surveillance and epidemiology components of WP5 are utilised to develop epidemiological models that can inform public health. In particular, work with the World Health Organisation partner has taken CHAIN into the heart of international policy. 2013 witnessed the publication of the updated WHO guidelines for antiretroviral therapy. A number of questions remain, and we have directly addressed these. For instance, what level of transmitted drug resistance is needed to justify a change in policy from first line NRTI to first line PI therapy? What are the most cost effective means of monitoring antiviral rollout? CHAIN has led the world in such key analyses; the overall impact of the work package is outlined below:

- Optimization of initial therapy to help prevent further transmission of drug resistance.
- Improved strategies for the prevention and spread of HIV.
- Knowledge of the clinical value of PrEP in preventing new HIV infections in RLS, which can be used to inform prevention programmes.
- Improved access to drug resistance testing in RLS.
- Implementation of drug-resistance prevention strategies by numerous countries.
- New evidence-based recommendations to inform clinical practice guidelines.
Future challenges
• As access to ART expands, strengthen HIV drug resistance surveillance efforts in the face of increasing concern about HIV drug resistance emergence and transmission.
• Close gaps in knowledge on the impact of first-line resistance patterns on second-line outcomes through initiation of multi-centre trials.
• Improve knowledge regarding the impact of protease inhibitors and other ARTs on drug resistance.
• Continue surveillance of resistance at the time of treatment failure to allow optimal choice of ART and preservation of treatment options for life long therapy.

WP7- TRAINING AND CAPACITY BUILDING

Overall impact of the work package
- Improved management of HIV drug resistance among clinical practitioners and diagnosticians.
- Increased awareness among policy makers on the issue of HIV drug resistance.
- Development of a framework for implementing policies at the level of the National Health System.
Future challenges
The major challenge of the work programme was establishing a large network of courses in Eastern Europe and Africa and encouraging wide participation. Challenges moving forward include implementation of the remaining workshops and collection of pre- and post-test results of all courses organized in the framework of the CHAIN project.


Project's Coordinator (University College London)
- Coordinator (scientific representative): Prof. Deenan Pillay (d.pillay@ucl.ac.uk)
- Manager (administrative): Ilaria Marsili (i.marsili@ucl.ac.uk)