Testing to Sustain Hepatitis C Elimination Targets in People Who Inject Drugs

A Network-Based Model

Chloë Siegele-Brown; Martin Siegele-Brown; Charlotte Cook; Salim I. Khakoo; Julie Parkes; Mark Wright; Ryan M. Buchanan

J Viral Hepat. 2023;30(3):242-249.

Abstract and Introduction

Abstract

Little is known about the level of testing required to sustain elimination of hepatitis C (HCV), once achieved. In this study, we model the testing coverage required to maintain HCV elimination in an injecting network of people who inject drugs (PWID). We test the hypothesis that network-based strategies are a superior approach to deliver testing. We created a dynamic injecting network structure connecting 689 PWID based on empirical data. The primary outcome was the testing coverage required per month to maintain prevalence at the elimination threshold over 5 years. We compared four testing strategies. Without any testing or treatment provision, the prevalence of HCV increased from the elimination threshold (11.68%) to a mean of 25.4% (SD 2.96%) over the 5-year period. To maintain elimination with random testing, on average, 4.96% (SD 0.83%) of the injecting network needs to be tested per month. However, with a 'bring your friends' strategy, this was reduced to 3.79% (SD 0.64%) of the network (p< .001). The addition of contact tracing improved the efficiency of both strategies. In conclusion, we report that network-based approaches to testing such as 'bring a friend' initiatives and contact tracing lower the level of testing coverage required to maintain elimination.

Abstract and Introduction
Methods
Results
Discussion
References

Table 1. Key parameters used in the model
Parameter	Value	Reference
Network size	689	[31]
Network structure	UK-net	[13]
Proportion of nodes with chronic HCV at baseline	0.417 × (1–0.72) = 0.1168	[31]
New nodes rate/year (as proportion of network size)	0.176	[32]
Proportion of new nodes with HCV	min(0.008, 0.116 × current network prevalence)	[33]
Node turnover/year as proportion of network size	0.176	[32]
Frequency of injection with each neighbour—proportion of nodes	6×/week—0.12, 3×/week—0.22 1×/week—0.43 0.5×/week—0.23	[15]
Probability of equipment sharing	0.4 (AES) 0.33 (RNS) (0.16–0.51)	[15]
Transmission probability when sharing	0.0023 (AES) 0.0073 (RNS)	[34]
Spontaneous clearance rate of acute infection	0.25	[34]
Minimum testing interval	3 monthly
Oral mouth swab antibody test sensitivity	95% (95% CI 90–99)	[35]
Confirmatory point of care RNA test sensitivity	Abbott Xpert 100% (96–100) Cephiad gene drive 98.6 (95% CI 96.9–99.5)	[18, 19]
Probability a partner node will be tested in CT strategy	42% (95% CI 0.19–0.86)	[20]
Positive test to commence treatment rate	0.5 (0.25–0.75)	Local data and [21]
Treatment duration (weeks)	2 months	[36, 37]
Treated persons coverage per month at baseline	Unlimited	a
Rate of sustained virological response	0.95 (95% CI 94–100)	[36, 37]

Abbreviations: AES, ancillary equipment sharing; CI, confidence interval; CT, contact tracing; HCV, hepatitis C virus; RNS, receptive needle sharing.
^aThe local treatment capacity in Southampton (UK) (the modelled PWID network) currently comfortable exceeds the number of patients with HCV RNA; this is likely to be the case in most areas that achieve HCV micro-elimination.