Real‐world outcomes of rapid regional hepatitis C virus treatment scale‐up among people who inject drugs in Tayside, Scotland

Summary Background In 2017, Tayside, a region in the East of Scotland, rapidly scaled‐up Hepatitis C Virus (HCV) outreach and treatment among People Who Inject Drugs (PWID) using novel community care pathways. Aims We aimed to determine treatment outcomes for PWID during the scale‐up against pre‐determined targets; and assess re‐infection, mortality, and post‐treatment follow up. Methods HCV treatment was delivered in community pharmacies, drug treatment centres, nurse‐led outreach clinics, prisons, and needle exchanges, alongside conventional hospital care. We retrospectively analysed clinical outcomes and compared pathways using logistic regression models. Results Of 800 estimated HCV‐infected PWID, 718 (90%) were diagnosed. 713 treatments commenced among 662 (92%) PWID, delivering 577 (81%) Sustained Virologic Responses (SVR). SVR was 91% among those who attended for testing. Forty‐six individuals were treated more than once. Needle exchanges and community pharmacies initiated 49% of all treatments. Regression analyses implied pharmacies had superior follow‐up, but there was no difference in likelihood of achieving SVR in community pathways relative to hospital care. Re‐infection occurred 39 times over 256.57 person years (PY), yielding a rate of 15.20 per 100 PY (95% CI 10.81‐20.78). 54 deaths occurred (29 drug related) over 1,553.04 PY, yielding a mortality rate of 3.48 per 100 PY (95% CI 2.61‐4.54). Drug‐related mortality was 1.87 per 100 PY (95% CI 1.25‐2.68). Conclusions Rapid HCV treatment scale‐up to PWID in community settings, whilst maintaining high SVR, is achievable. However, other interventions are required to minimise re‐infection; reduce drug‐related deaths; and improve post‐SVR follow‐up testing regionally.


| INTRODUC TI ON
Hepatitis C Virus (HCV) is a blood-borne virus (BBV) that is transmitted mainly via percutaneous exposure to infected blood. Therapeutic advances led the World Health Organization, in 2016, to release its Global Health Sector Strategy, providing an implementation plan for HCV elimination. 1 It specified targets to diagnose 90% of infected individuals and initiate treatment for 80% of those diagnosed, to facilitate HCV elimination by 2030.
In 2017, concurrent to the WHO strategy response, National Health Service (NHS) Tayside-a health board in the East of Scotland serving approximately 416,000 people-commenced a trial of Treatment as Prevention (TasP) in HCV infection, by rapidly scaling up treatment among PWID. 2,3 The first step was a regional programme of intensified HCV testing and treatment to reduce chronic HCV to <10% among PWID over the long term and achieve WHO elimination criteria. 3 The concept of TasP is using treatment to lower HCV prevalence among PWID to prevent new infections and reinfections, maintaining elimination or reducing the amount of treatment needed to maintain it. 3,4 The programme was underpinned by novel community care pathways, and highly effective Direct Acting Antiviral (DAA) treatments. 5,6 The pathways were implemented by a multidisciplinary team. Clinical initiatives focused on PWID, as injection drug use (IDU) was the primary driver of regional HCV transmission, and over 90% of HCV infections in Scotland occur consequent to IDU. 7,8 This study aims to determine if rapid regional scaleup of HCV treatment was achieved in the PWID population in line with local and WHO targets. We report critical real-world outcomes of the scale-up including cure rates (Sustained Virologic Response, SVR) by treatment pathway, risks of HCV re-infection, mortality rates, and effectiveness of post-treatment follow-up, both for SVR and reinfection testing.

| Testing and treatment targets
Prior to scaleup, there were an estimated 2800 PWID resident in the Tayside heath board region, of whom 800 (29%) were estimated to have chronic (6+ months infected) HCV. 3,9 In line with these estimates, local targets were set to achieve the prevalence reduction required to demonstrate TasP over the long term. Targets were to diagnose and engage approximately 680 (85%) HCV infections among PWID, initiate approximately 592 (87% of diagnosed) treatments, and obtain approximately 533 (≥90% of treated cases) undetected SVRs.

| Study setup
The study is described on clinicaltrials.gov (NCT03356405) and ISCRCTN (ISRCTN72038467) and co-sponsored by NHS Tayside and University of Dundee (2016GA08). It received NHS ethical favourable opinion (17/ES/0136). Data was obtained from HCV clinical databases; virology/microbiology records; and patient records.
Caldicott Guardian approval for this was granted (IGTCAL7005). 10 Tracking individuals was by Community Health Index (CHI) number, an identifier allocated to every registered patient in Scotland. All adults who initiated HCV treatment with DAAs from January 2017 to mid-April 2020, immediately prior to Tayside's elimination declaration, who acquired HCV via IDU, were eligible. 11  RNA samples were tested using an in-house assay 15 or the Abbott m2000 system. 16

| Patient and public involvement
Patient or public involvement groups were not involved in the design of this study. Study design was developed from prior experience of the investigators working with the stakeholders involved.

| Population and outcome definitions
PWID was defined as ever injected drugs (self-reported). Cirrhosis was defined as any of the following: Fibroscan reading >18kPa; Fib4 >3.46, or confirmation via: liver biopsy; ascites with evidence of liver disease; hepatic encephalopathy plus chronic liver disease; oesophagogastric varices with patent portal vein. SVR was undetectable HCV RNA at least 12 weeks post-treatment. Relapse was undetectable RNA at end of treatment, but detectable prior to or at SVR; or treatment initiation and detectable RNA prior to or at SVR, if end of treatment test not conducted. Non-response was detectable RNA above the lower limit of detection (

| Treatment pathways
There are six HCV care pathways in Tayside; five in community settings. They are named by location: conventional hospital care; drug treatment centres; needle exchanges; community pharmacies; nurse-led outreach clinics and prisons. All community pathways are in environments accessible to PWID. In the nurse-led outreach pathway, patients are seen in their local neighbourhood, typically in remote areas. Drug treatment clinics are in central locations.
Testing, pre-treatment assessment, and treatment are available in each (ultrasound requires central travel). The needle exchange and community pharmacy pathways were initially experimental, implemented through pragmatic pathway trials, which became embedded to provide routine testing and treatment. [18][19][20][21] All pathways are multidisciplinary, involving specialist nursing staff, gastroenterologists/ hepatologists, infectious disease physicians, specialist and community pharmacists, and psychologists. Nurses and community pharmacists can prescribe DAAs for simple cases of HCV infection (eg treatment naïve and non-cirrhotic), either as independent prescribers or through Patient Group Direction. 22

| Statistical analysis
Primary outcomes (treatments and SVR) were summarised using descriptive statistics, for comparison against targets. Comparison of SVR by patient and treatment characteristics were conducted in the same manner in intention-to-treat (ITT) and per-protocol (PP) populations. The ITT group includes all who initiated treatment, and the PP group includes treatment completers. We hypothesised that variation in SVR in community pathways compared to specialist hospital care would be due to differences in follow-up. Logistic regression modelling was used to assess this. Models were adjusted for patient characteristics and explored steps in the treatment journey for the ITT and PP groups, with and without SVR tests. Models were adjusted in a stepwise approach for age, gender, cirrhosis, and genotype. Where applicable, participants missing SVR were assumed treatment failures. P of ≤0.05 was assumed to demonstrate statistical significance. Analyses were undertaken using IBM SPSS Statistics 25.

| Re-infection and mortality
Those who achieve SVR are offered annual follow-up RNA testing through all pathways. We measured the efficacy of re-infection follow-up by calculating the proportion of cases with follow-up RNA tests received in line with local policy (annual re-testing post-SVR). We estimated re-infection incidence per 100 person-years  Tayside Table 1 outlines descriptive cohort data (tabulated by pathway in Data S1). All patients were treated with DAAs. All patients had an injection history, 77.3% received OST and 51.9% reported injecting in the 12 months prior to treatment. Most were male.

Annual treatments initiations among PWID in
Median age at first treatment was just under 40. Most individuals had a registered postcode in the most deprived areas. HCV genotypes 1 and 3 were most common. Co-infection with other BBVs was uncommon, and few were diagnosed with cirrhosis.
Approximately one third were treated through a pragmatic trial of a novel pathway.
The majority were treatment naïve prior to the scaleup and, among those who were treatment experienced, 9 in 10 treatments were based on pegylated interferon (Peg-IFN). Most (88.4%, Table 2) completed their course of DAAs. Figure 1 shows outcomes by treatment initiation, and cumulative treatment and post-treatment outcomes.
Eighty-two cases did not have an SVR test (lost to follow-up, deceased or moved). Excluding those missing an SVR test in the ITT group, SVR was 91.4% (Table 2). In the PP population (n = 630), 536 SVRs were obtained (85.1%). Excluding 62 PP cases that did not have an SVR test, this rose to 95.7%. Loss to follow up (LTFU) varied substantially across pathways (Table 2).
Of treatment completers, eight died prior to SVR testing. Of

| Treatment pathways
Needle exchanges initiated the highest proportion of treatments (  Figure 2).
Results for pathway-denominated logistic regressions are presented in Table 4, with full tabulation in Data S1 (Sections 2 and 3).  for females may require attention across all pathways, as the effect fell away in the PP group with an SVR test (Data S1, Tables 2.8 and 2.9). Models combining all community pathways into one comparator were non-significant across all groups (Data S1, Section 3).  Our findings suggest a high level of attrition at the post-treatment follow-up point among our cohort of PWID. This contrasts with data from, for example, the Icelandic national scaleup among PWID, which reported minimal attrition at each step of the HCV cascade of care. 26 The statistical analyses initially implied that receiving treatment through a community pharmacy was associated with higher odds of SVR. However, in the ITT population who attended for SVR, there were no significant differences between community pathways and hospital care. This implies differences in SVR were caused by variation in follow-up and that, irrespective of treatment completion, pharmacies provided superior follow-up for SVR testing. The models further suggested that SVR did not vary by treatment pathway or patient characteristics among those who completed treatment.

| D ISCUSS I ON
One part-adjusted model was significant which implied that when accounting for treatment pathway, age and gender, female gender was weakly associated with reduced odds of SVR. Consequently, targeted follow-up initiatives may be required to improve follow-up for SVR testing among females across all pathways.  The data indicate high deprivation, in line with national estimates, and reinforces evidence that HCV disproportionately impacts the most disadvantaged. [27][28][29] This Tayside cohort is relatively young, with low incidence of BBV co-morbidity. However, mortality risk in the population was notable, and over half of deaths were drug related. This aligns with reports suggesting Tayside has the second-highest rate of DRDs in Scotland and implies that service improvements are required to reduce drug-related harms. 30,31 Compared to a recent cohort study of drug-related mortality PY-our observed mortality rate is concerningly high. 32 Ongoing re-infection and mortality risks in our cohort suggest substance misuse and harm reduction services for PWID in Tayside must improve.
The high proportion of cure adds valuable real-world evidence to existing literature demonstrating the efficacy of DAAs for treating HCV in PWID. 5,6 The highest proportion of SVR (excluding unknowns) in ITT analysis was in the nurse-led community pathway.
In this pathway, diagnosis and DAA prescribing are undertaken by nurses. Similarly, in the pharmacy pathway, diagnosis and treatment can be wholly facilitated by the pharmacist through the use of simplified DAA prescribing algorithms and the use of dried blood spot sampling. The outcomes in our cohort strengthen the case for taskshifting elements of care which can be completed in the community into the community for PWID. [33][34][35] Our data augments the literature suggesting re-treatment following re-infection should be offered without stigma or discrimination. 36

| Limitations
This study is limited by its observational nature, its reliance on routinely collected data, and the lack of a comparator cohort, which is the basis of the next phase of the study. During data collection, some tion on a regional level using real-world data.

| CON CLUS IONS
Our analyses indicate that Tayside achieved pre-specified local diagnosis, treatment, and SVR targets set for the TasP scale-up programme. Further, Tayside has met WHO testing and treatment targets having diagnosed 90% of the estimated PWID population with HCV, and treated over 80% of cases. 1 Our data implies that rapid regional scale-up of HCV treatment for PWID can be achieved through novel multi-stakeholder pathways focussed on shifting testing, DAA prescribing, and treatment into community settings. To our knowledge, is the first real-world study from the United Kingdom to demonstrate the effects of rapid regional HCV treatment scaleup among PWID, highlight the benefits of utilising non-specialist prescribers regionally, and quantitatively compare novel community pathway contributions to scaling up HCV treatment.
The re-infection rate, and proportion of DRDs, suggests we reached the highest-risk individuals for transmission of HCV and implies that harm reduction and substance misuse services in the region need strengthening. Other areas targeting PWID to achieve elimination should provide substantial-related care in these respects alongside comprehensive testing and easy access to treatment.

ACK N OWLED G EM ENTS
The authors acknowledge the work of nurses, clinicians, pharmacists, virology laboratory staff, clinical trial staff, and third-sector workers who helped to implement the HCV service. We thank Shirley Cleary, Maya Prakash and Jan Tait for support with data gathering. We also acknowledge Dr Amy Malaguti and Prof Fabio Sani for statistical discussions, and Dr Malaguti for providing manuscript feedback.
Declaration of personal interests: CB has nothing to disclose.

AUTH O R S H I P
Guarantor of the article: Christopher J Byrne.
Author Contributions: CJB co-ordinated the study, collected data, performed statistical analyses and wrote the initial manuscript draft.
LB co-ordinated the study and collected data. SKI managed the study. ER advised on testing data. AR, DJG, MH, SH and JFD collaboratively conceived, designed, planned and executed the trial. All authors critically revised the manuscript for intellectual content and approved the final version.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data underpinning this study were obtained from routinely updated NHS health records in line with approval granted by the NHS Caldicott Guardian. The individuals to whom the data pertains did not explicitly consent to its use for research purposes. Therefore, it is not possible for the authors to share this data. However, interested parties can make specific requests to NHS Tayside Information Governance by email on: informationgovernance.tayside@nhs.scot.