Hepatitis C : host-virus interactions and their impact on treatment response

Hepatitis C is a major cause of chronic liver disease with over 120 million infected people worldwide. Untreated chronic hepatitis C (CHC) infection leads in 20-50% to liver cirrhosis culminating in cirrhosis-related complications and hepatocellular carcinoma eventually leading to death. For the last decade a combination therapy of pegylated Interferon-alpha (pegIFN-α) and Ribavirin was the standard of care resulting in sustained virologic response (SVR) rates of approximately 50%. The antiviral effect of this cytokine is achieved by creation of an antiviral state in infected host-cells and triggering of the host’s immune system through induction of hundreds of interferon stimulated genes (ISGs). Interestingly, some patients with CHC who show a strong induction of ISGs in the liver even before treatment do not respond to administered pegIFN-α and do not clear the virus. The lack of response is most likely due to refractoriness of the IFN signaling. The reason for this preactivation of the IFN system in the liver in a subset of patients and the following non-response remains unclear. It is the aim of this thesis to improve the understanding of host-virus interactions in hepatitis C infection with regard to the preactivation of the IFN system and its consequential failure to IFN-α based treatment regimens. For this purpose, the host-response in the liver of patients in the acute phase of hepatitis C (AHC) infection, i.e. the first six months after transmission, was investigated. To elucidate molecular mechanisms involved in non-response, we wanted to exploit the fact that in AHC SVR rates to therapy with pegIFN-α are substantially better than in CHC (>90% versus 50%). Six liver biopsies of AHC patients were analyzed for ISG expression and IFN signaling and compared to a set of patients with CHC and control liver samples. Additionally, IFN-α and -γ specific gene sets were defined in primary human hepatocytes. While both AHC and CHC non-responders (CHC-NR) showed a strong induction of ISGs, enrichment analysis revealed that in CHC-NR mainly IFN-α stimulated genes were induced, in contrast to IFN-γ stimulated gene expression in AHC. IFN-γ was increased in AHC and correlated with the amount of infiltrating CD8+ T cells that by immunostaining were found to be co-localized with activated hepatocytes. Analysis of negative regulators of IFN signaling in the IFN-α stimulated gene set revealed exclusively in CHC-NR an upregulation of USP18, a key molecule in establishing refractoriness to IFN-α signaling. These results provide an explanation for the preserved response to pegIFN-α in AHC and highlight USP18 as a potential therapeutical target. Additionally, a possible connection between genetic variants near the IL28B gene and ISG induction in livers of CHC patients was assessed. Four independent genome-wide association studies have revealed a highly significant association of single nucleotide polymorphisms (SNPs) near the IL28B gene with the outcome of therapy with pegIFN-α and ribavirin in CHC. We hypothesized that these genetic variants near the IL28B gene, which encodes for IFN-λ3, might be responsible for the preactivation of the IFN system in certain patients. 109 patients with CHC were genotyped for IL28B SNPs and the hepatic ISG expression was quantified. Interestingly, despite an association of the IL28B genotype with the expression of ISGs, stratification revealed that ISG expression is associated with response independent of its IL28B genotype making a direct link rather unlikely. A multivariate analysis using a random forest classifier analysis defined ISG expression, by the means of a 4-gene-classifier, as the strongest treatment predictor. Last, the pharmacodynamics of pegIFN-α in the livers of patients with CHC was explored. Due to higher efficacy, pegIFN-α has replaced conventional IFN-α as standard of care. It is generally assumed that the improved pharmacokinetic properties of the former with a longer half-life leads to better effectiveness through a continuous induction of ISGs. However, basic studies in vitro and in mouse models suggest a long-lasting refractoriness of the IFN-α signaling. We therefore addressed this issue directly in CHC patients receiving treatment. Each patient received a paired biopsy before and at a certain time point after the first injection with pegIFN-α. After transcriptome analyses, clusters of genes with distinctive temporal patterns were generated. The upregulation in the early ISG clusters was only transient and no prolonged upregulation or a second wave of induction could be noticed. Additionally, a direct comparison of the two commercially available pegIFNα, pegIFNα-2a versus -2b at 144h showed no significant difference in the amount or extension of upregulated ISGs, despite the longer serum half-life of pegIFNα-2a. This study indicates that the superior efficacy of pegIFN-α compared to conventional IFN-α cannot be explained by persistent signaling and ISG induction.