Modeling the Ongoing Hepatitis C Outbreak Among Injection Drug Users in West Virginia: Can Early Initiation of Sofosbuvir and/or Suboxone Reduce Hepatitis C Disease Burden for People Who Inject Drugs?

Abstract

Background Hepatitis C Virus (HCV) infection is a pressing public health issue worldwide, as nearly 180 million people are currently infected. Similarly, in the United States, people who inject drugs (PWID) are disproportionately a ected by HCV and they generally constitute a large portion of the HCV disease burden. An ongoing outbreak of hepatitis C virus (HCV) among people who inject drugs (PWID) in West Virginia (WV) has been recently reported by the Centers for Disease Control and Prevention (CDC). It is interesting to note that WV has also maintained the highest rate of drug overdose deaths in the U.S. for the past few years. The demographic makeup of this current outbreak is mainly non-Hispanic whites aged less than 30 years. Recently, physicians in WV have begun prescribing buprenorphine/naloxone (suboxone) to aid PWID in recovering from their opiate addiction. Under the Medicaid program in WV, treatment for HCV with direct-acting-antivirals (DAAs) such as sofosbuvir in infected patients is delayed until F3 METAVIR (Meta-analysis of Histologic Data in Viral Hepatitis) stage of liver brosis|when signi cant scarring of the liver tissue may have already occurred. In WV, the policy is to only treat ex-IDUs or those individuals on opiate substitution therapy (OST) for HCV infection. In this thesis, it is hypothesized that early initiation of sofosbuvir and/or suboxone (OST) may be able to reduce hepatitis C disease burden for PWID. Methods A deterministic compartmental transmission model of HCV with layers incorporated for PWID demography was constructed in MATLAB R2015b (v. 8.6). The HCV component of the model was subdivided into eight compartments, each representing a speci c stage of HCV infection. The PWID demographic trajectory was strati ed among four classes, representing the transition from current IDU to ex-IDU or into opiate substitution therapy (i.e. suboxone). The model created was calibrated to HCV prevalence data between 2001 and 2010 from the West Virginia Department of Health and Human Resources (WV DHHR). Interventions were simulated for three di erent trajectories of increasing treatment coverage: modest, moderate, and optimal. Results The mathematical model constructed accurately re ected the trends in HCV among current and ex-IDUs in WV. Moderate interventions of both suboxone and sofosbuvir scale-up are able to reach a -25% change from the baseline scenario. When scaling up treatment coverage for just ex-IDUs in WV (current policy in WV), there was no change in the prevalence for current IDUs, which suggests that a large contributor of the HCV disease burden is left out. Similarly, there was no change in the incidence among current IDUs following ex-IDU scale-up alone. 7 Conclusion The results of this thesis suggest that if the WV government continues with just treating ex-IDUs, a large portion the HCV disease burden caused by current IDUs will be neglected. This thesis shows that a potential \treatment as prevention" intervention (even with modest coverage proportions) can be bene cial in ameliorating the HCV outbreak among PWID in WV.