Article Text

Hepatitis C elimination by 2030 through treatment and prevention: think global, act in local networks
1. M Hellard1,2,3,
2. R Sacks-Davis1,4,
3. J Doyle1,2,4
1. 1Centre for Population Health, Burnet Institute, Melbourne, Victoria, Australia
2. 2Department of Infectious Diseases, The Alfred, Melbourne, Victoria, Australia
3. 3Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Victoria, Australia
4. 4Department of Medicine, University of Melbourne, Melbourne, Victoria, Australia
1. Correspondence to Professor M Hellard, Centre for Population Health, Burnet Institute, 85 Commercial Road, Melbourne, VIC 3004, Australia; hellard{at}burnet.edu.au

## Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Hepatitis C virus (HCV) elimination is now an achievable goal;1–3 the WHO is likely to set ambitious but achievable 2030 elimination targets later this year.4 In high-income countries, people who inject drugs (PWID) are the group at greatest risk of HCV infection,5–7 with HCV prevalence among PWID varying from 10% to 97% in different regions of the globe.7 ,8 In low income and middle income countries, HCV transmission can be due to iatrogenic transmission (in the formal and informal health systems), injecting drug use or a combination of the two.9

Highly effective direct-acting antiviral (DAA) treatment with improved tolerability is the cornerstone of HCV elimination. Models that combine HCV treatment as prevention and harm reduction services—including opioid substitution therapy (OST) and needle and syringe programmes (NSPs)—suggest that HCV prevalence and incidence can be significantly reduced and the proposed WHO elimination targets can be met.10 ,11 In countries with significant levels of iatrogenic transmission, health system strengthening is also required.12

The concept of disease eradication and elimination is a daunting prospect: only one infectious disease, smallpox, has been successfully eradicated from the world.13 As well as HCV, the 2030 elimination targets include malaria, tuberculosis and HIV, all of which involve considerable challenges. In part, this is due to how they are transmitted: malaria is vector-borne, making it unlikely that the disease reservoir can be totally eliminated. Tuberculosis is airborne, making transmission difficult to control, and the treatment duration is long, making multidrug resistance possible if therapy is not taken correctly. Although HIV can be well controlled using antiretroviral therapy, it cannot be cured and there is no effective vaccine. Also, HIV prevention programmes—such as the use of condoms to stop sexual transmission—require behaviour changes to which many people are resistant14–16 and can be difficult to implement culturally in others.14 The long-term prevention benefits of pre-exposure prophylaxis, post-exposure prophylaxis and treatment as prevention are yet to be confirmed in real-world settings.

In this paper, we examine current approaches to treatment and harm reduction and the tools necessary to eliminate HCV. We describe the role of networks in HCV transmission and the potential for intervention to eliminate transmission. While we focus on PWID networks, this approach could be applied to networks of HIV/HCV co-infected gay and bisexual men, or HCV networks or clusters in towns or villages where HCV transmission is occurring, or within families. Finally, we consider barriers to treatment at the local level and how they could be overcome.

## Treatment and harm reduction

HCV, a blood borne virus, is potentially easier to eliminate than other infections. First, there are now all-oral treatments that can cure infection in over 90% of people17–19 with few side effects and a comparably short duration of therapy (up to 12 weeks in most patients).17–19 Treatment of HCV will reduce prevalent cases and prevent new infections. Second, there are simple ways to interrupt transmission, such as the provision of a new needle and syringe for each injecting episode.20 Unlike the situation with condoms and sex, the vast majority of PWID would prefer to use a new needle and syringe each time they inject.21 Also, there is growing evidence that OST, another form of harm reduction, reduces HCV incidence.22–24 The barriers to these highly cost-effective harm reduction programmes in many jurisdictions are political rather than evidence-based, and it is hoped that as HCV elimination is recognised as an achievable target, more jurisdictions implement harm reduction programmes.

### The role of network structure in hepatitis C elimination

Another important factor increasing our ability to eliminate HCV is the structure of the injecting network through which disease transmission occurs. While treating a large homogeneous group of PWID appears daunting, if instead we think about treating small local networks of PWID, it becomes easier to conceptualise making inroads into the epidemic.

PWID do not have an equivalent injecting relationship with every person who injects drugs within their community, city or country;25 instead, most PWID inject with only a handful of other injectors, some of whom may have links with other small groups of PWID. These interconnected small groups, or subnetworks, comprise the injecting network more broadly within a community or city. Importantly, these subnetworks of PWID are often situated in very particular geographic and social contexts within the broader network. Work by our group, involving collection of data on the injecting networks of over 380 Australian PWIDs, shows that while connections can be extensive, large clusters of network members are located in separate geographic areas with evidence of assortative mixing by location, with fewer links between distinct geographic areas and ethnic groups.26 As a result, the structures of injecting networks can reduce the onward transmission of HCV. Modelling confirms this assertion, with HCV transmission being slower through an empiric network than if all PWID are homogeneously linked.27 Also, phylogenetic data suggest that injecting networks are often quite distinct:27 a large international collaboration28 examining incident HCV infection in PWID reported distinct genetic clusters of acute HCV infection in North America, Europe and Australia.29

Treatment of individuals and their injecting network, analogous to methods for treating sexually transmitted infections, has been proposed. A concern about embarking on a locally based strategy of treating the immediate network (known as a ‘treat your friends’ approach—see below)10 is that HCV infection might be introduced from outside the network. However, our models showed that HCV transmission through the network can be disrupted if an infected individual is treated and becomes HCV negative, even when some HCV is introduced (or imported) from outside the network.27 This suggests that the structure of the injecting network makes HCV elimination plausible at a local level, particularly when treatment is combined with effective harm reduction. Treatment of PWID can reduce the prevalence of HCV within 10 years;11 the use of effective harm reduction, combined with timely treatment of any new infections and clusters, will stop the reintroduction of HCV into the broader injecting network.

## Barriers to HCV elimination

HCV elimination is plausible, but local barriers must be addressed to make it achievable.

### The cost of treatment

Cost is a key barrier to achieving the WHO's HCV elimination targets. In many countries, the cost of DAAs is prohibitively expensive for health systems and individuals, reducing the likelihood of significant numbers of people accessing treatment at this time.30 However, the cost of treatment is already falling—in several resource-limited countries—to as low as $1000 per course.31 Similarly, in some high-income countries, the cost of treatment has fallen well below the US$80 000 dollars per course initially proposed in the USA.30 While it is unlikely that cost will disappear as a barrier to treatment access, over the next few years it is likely to fall significantly in many jurisdictions due to competition between drug companies, off-label patents, individual country agreements and the low manufacturing costs of these drugs.32 Nevertheless, the cost to an individual living with HCV in jurisdictions without subsidised healthcare or insurance will continue to be a barrier in many settings.

A second key barrier to achieving HCV elimination is that many jurisdictions restrict access to HCV treatment for PWID.33 In the USA, many states and health insurers have restricted HCV treatment access based on past or current injecting drug use. While the rationale for this is not explicit, it is probably driven by concerns about HCV reinfection following treatment. While HCV reinfection does occur among active PWID at 1–6 per 100 py,34 it should not be a justification for restricting healthcare. First, it is difficult to think of another disease with significant morbidity and mortality for which an individual is excluded from receiving a proven cure due to concerns about future infections (which can also be treated successfully). Additionally, in jurisdictions where ongoing HCV transmission occurs predominately through injecting drug use, it is imperative that PWID have access to therapy to stop ongoing transmission. Put simply, elimination of HCV will not occur without engagement with and treatment of current PWID. Such an approach will require some jurisdictions to reconsider their current policies in regard to treatment access. From the perspectives of both fairness and economics, an approach which engages PWID in care and treatment is ultimately in the best interests of individuals living with HCV and payers who have traditionally been hesitant to direct care towards PWID.

### Community-based treatment

The elimination of HCV requires a significant increase in the number of people accessing treatment. Currently, the majority of HCV treatment occurs in tertiary hospital settings. There is increasing evidence that moving treatment out of tertiary hospital settings and into the community improves treatment capacity and access for PWID, and decreases the cost of care.35–38

Traditionally, HCV care and treatment has been the province of specialists and tertiary care centres due to the complexity of care, liver complications when treatment was initiated in patients with advanced fibrosis and funding models that prioritise hospital-led care. In order to treat everyone living with HCV, including PWID, services will need to be expanded and shifted to primary care or other healthcare workers.38 This shift in service provision should be accompanied by appropriate workforce training and development. This shift from the tertiary care setting to the community should have the added benefit of accommodating patient preference, thereby facilitating treatment uptake, adherence and quality of life. Peer support workers should play an important role in designing appropriate models of care and providing education and support if treatment scale-up is to be achieved and simplified.

While the delivery of oral HCV therapies has not yet been evaluated outside of tertiary care settings, given the safe use of pegylated interferon based treatment in community settings including primary care,35 nurse-led models of care36 and prisons,39 it is reasonable to assume that oral therapies can also be delivered in these enviroments. Health services and programme managers now have the opportunity—and challenge—to redesign their models of care to accommodate those with HCV, including key populations like PWID and the increase in patient throughput that treatment scale-up will represent.

## Opportunities to enhance elimination

### Treating PWID using a networks-based approach

The structure of the network has the potential to slow disease transmission through the network; also, modelling suggests that taking the network structure and characteristics into account can make treatment programmes more effective. Recent modelling by our group that accounted for the social network of PWID found that treating an individual and their immediate network members has a greater effect on reducing HCV incidence and prevalence across an entire network of PWID than treating people individually.10 ,40 The ‘treat your friends’ strategy10 directly benefits PWID—their HCV is cured; also, it has the secondary benefits of the PWID not transmitting their infection to others, and if their injecting partners are also treated, it substantially reduces the risk of reinfection.

The benefit of a network-based ‘treat your friends’ strategy is it reminds us that HCV transmission is not a random event, and provides us with a road map to the future. It is not necessary to know a PWID's entire network; simply offering treatment to a PWID and his/her immediate injecting partners impacts on HCV transmission, incidence and prevalence.10

### Elimination –think global, start local

Taking a considered, strategic approach to HCV treatment is vital to eliminating hepatitis among PWID. For example, in a population of 100 000 people, if we assumed that 1%i of the population currently inject drugs (1000 PWID) and the prevalence of hepatitis in PWID is 50–80%,ii then between 500 and 800 current PWID in the local area will have chronic HCV infection.41 Treating 500–800 people would require 5 primary care (or nurse) practitioners to each treat 100–160 HCV-infected PWID. In the context of DAAs, treating that number of patients per year is feasible, although identifying and recruiting all of the HCV-infected PWID in a local area rapidly would be challenging and probably unrealistic. Importantly, modelling suggests that not every PWID must be treated immediately to reduce HCV incidence and prevalence.11 ,10 Treating much smaller proportions of HCV-infected PWID in a local area can still lead to reductions in HCV incidence, leading to further long-term reductions in prevalence in the area. Using a ‘treat your friends’ strategy, where PWID are encouraged to bring their injecting partners to the clinic for treatment, may lead to even greater reductions in HCV incidence through reducing the risk of reinfection of treated individuals.10 Even if HCV is reintroduced into a local area from an external source, the use of NSP and OST to reduce reinfection risk, combined with treatment of those infected or reinfected, means HCV prevalence could be contained at a very low level.

### Reducing HCV in local areas using treatment as prevention

Currently, there are no empirical data on the effectiveness of any treatment-as-prevention strategies, but (as outlined above) we and others have developed projections for these using mathematical models.10 ,11 ,40 While an area-by-area treatment strategy has not been modelled explicitly, we modelled an analogous scenario. The empirical social network on which we based our model includes participants from three distinct areas in Melbourne, Australia. Each of these areas is home to an illicit drug market; although there were links between areas, most social-injecting partnerships that we observed were within-area links. Our algorithms for selection of participants for treatment did not take neighbourhood into account, so effectively we modelled treating three areas simultaneously.

Importantly, these are not the only neighbourhoods with illicit drug markets in Melbourne. One of the features of our model is that it assumes that only part of the burden of incident infection came from within the social network itself, and the remainder was imported.10 ,27 ,40 The portion of imported infection includes sources from the neighbourhoods that we sampled who were not recruited into our study, as well as sources from outside these neighbourhoods. Importantly, even though we allowed importation of infection from outside the networks and neighbourhoods that we sampled, our treatment strategies led to reductions in HCV incidence in the modelled social networks. This implies that it is possible to reduce HCV prevalence and incidence in a particular area even if HCV infection is also being imported from elsewhere. As long as the bulk of HCV transmission occurs within rather than between local areas, HCV can be reduced area by area in an effort to achieve elimination in the long term.

In larger cities, such an exercise could be repeated in multiple local sites. Although there may be some connections between localities, the non-homogeneous structure of the injecting network and the likelihood of reduced connections across geographically distinct areas mean that a similar impact will be observed: multiple local areas of elimination can combine to eliminate HCV from a large city.

## Conclusion

The WHO's proposed global targets for HCV for 2030 include an 80% reduction in new cases of HCV infection from the 2010 level.4 This is not an impossible task if we take a ‘think global, act local’ approach, in which clinics are structured to support PWID accessing HCV treatment with DAAs, combined with NSP and OST programmes.

View Abstract

## Footnotes

• Funding National Health and Medical Research Council (Early Career Fellowship, Postdoctoral Fellowship, Senior Research Fellowship).

• Competing interests None declared.

• Provenance and peer review Commissioned; externally peer reviewed.

• i Global estimates of the prevalence of hepatitis C virus in the population vary greatly. Mathers et al reported midpoint country estimates ranging from 0.02% (India and Cambodia) to 5.21% (Azerbaijan). In Western Europe (with the exception of the Netherlands, where there is very little reported injecting drug use) the prevalence ranged from 0.13% (Greece) to 0.83% (Italy). In countries like the USA, Canada, Australia and New Zealand, estimated prevalences are close to 1%.41

• ii As outlined earlier, estimated HCV prevalences in the national populations of PWID vary greatly, ranging from 10% to 97%.6 ,7