CCL4 is the only predictor for non-responder in GT-1 CHC patients with favorable IL28B genotype when treated with PegIFN/RBV

Chronic hepatitis C is currently one of the leading causes of cirrhosis and hepatocellular carcinoma (HCC) in the whole-wide world [1, 2]. Eradication of HCV virus infection could reduce the risk of cirrhosis, hepatocellular carcinoma and hepatic decompensation [3, 4]. Though the direct antiviral agents (DAAs) are now the standard of care in Western countries [5], dual therapy of pegylated interferon-α/ribavirin (PegIFN/RBV) still is a popular and effective treatment in several countries where DAAs are not available or not affordable [6‐8]. In the treatment with PegIFN/RBV, the patients with non-response (NR) are a troublesome group of patients [9]. Even in the era of DAAs, NR group also highlight a special group of patient that needs special attention [10]. Recently, in the present newer generation of DAAs, this group of patients has finally achieved satisfactory SVR rate. However, in the next development of chronic hepatitis C treatment, shorter duration of interferon-free DAAs will be a hot issue to be investigated [11]. In this possible new trend of treatment development, this potential NR group is worthy of re-revaluation.

Host immune response strongly correlates to the success of antiviral treatment. According to the previous studies, chemokines/cytokines do play important roles in the pathogenesis of chronic hepatitis C. Chemokines and chemokine receptors are crucial in T cell recruitment into infected sites and are involved in inflammation, infection and tissue damage [12, 13]. Type I interferons upregulate either directly or indirectly the expression of CCL3–5, which were potent ligands of the chemokine receptors CCR5 and CCR1. Similarly, Type II interferons are recognized as the most potent inducers of CXCL9–10, which bind to the chemokine receptor-CXCR3 [13]. A previous study revealed that the predominant liver infiltration by majorly CCR5 high/ CXCR3 high phenotype CD8+ lymphocytes in GT-1 CHC patients correlates to intrahepatic chemokine expression level and the inflammatory activity of chronic hepatitis C [14, 15]. However, the clinical implications in treatment responses to pegylated interferon plus ribavirin (PegIFN/RBV) treatment have not yet been fully illustrated. In the era of PegIFN/RBV treatment, the treatment would be terminated if HCV RNA still detectable by 24 weeks (so-called NR). The host immune reaction between non-responder and responder under Peg-IFN/RBV remained unclear. Here, we examined the impact of cytokine and chemokine (CXCL9, CXCL10, CXCL11, CCL3, CCL4, IFN-γ and IL10) from peripheral blood mononuclear cells between NR and non-NR to elucidate why host immune failed to response toward PegIFN/RBV treatment.

In the present study, we focus on this group of patients with NR and found the rs12979860 non-CC genotype were strongly associated with treatment outcome of NR. Furthermore, in patients with advantageous rs12979860 CC genotype, higher baseline serum level of CCL4 was the only factor that is independently associated with non-response.

The role of IL28B genotype in predicting Peg/RBV treatment outcome like non-responder had been explored before like our previous studies [16, 17] and others [18‐20]. Interesting, in the rapid advance of DAAs treatment, the role of IL28B on the SVR had gradually dwindled when treatment regimen are non- pegylated-IFN based [21]. However, in consideration of minor group with possible treatment failure by DAAs, the IL28B might still have impacts on the outcome [21].

The finding about chemokines be influential to the treatment outcome was compatible with another report that serum CXCL10 and CCL4 levels decreased significantly in GT-1 CHC patients with virological response [22]. Furthermore, CCL3, CCL4, CCL5, CXCL9, CXCL10 and CXCL11 were found to increase in both liver and peripheral blood during chronic hepatitis C in several studies [14, 22, 23]. The intra-hepatic levels of CXCL11 and CXCL10 were reported to correlate with HCV disease severity [13]. Patients with high CXCL10 at baseline were much less likely to achieve SVR, and the CXCL10 level was observed to be decreased following successful antiviral therapy [24, 25]. In HCV-infected livers, inflammation and fibrosis are mainly located in the portal areas, which may explain the up-regulation of CCL3–5 in the portal tracts [13]. However, the relationship existed between CCL3, CCL4 levels and the therapeutic responses were still controversial. A study showed that a low pretreatment CCL4 concentration was not only an independent predictor of early but also sustained virological response in CHC patients, while another study didn’t found significant differences [26, 27]. Interestingly, patients of advantageous IL28B genotype predominated among all recruited patients in the former study. To the best of our knowledge, no study yet had analyzed baseline CCL4 level in patient groups of advantageous IL28B genotype.

There were some limitations for this study. First of all, it was a retrospective study. However, in the new era of DAAs treatment, it is difficult to conduct a large-scale study just focused on PegIFN/RBV treatment. In addition, it is a medium-size study with case number of 60. However, in this scale of study, the serum levels of CCL4 become the only predictor for NR in patients with advantageous IL28B genotype. Therefore, it has emphasized the importance of CCL4 among other serum chemokines, especially in considering the future shorter duration of treatment for chronic hepatitis C patients receiving shorter duration of interferon-free DAAs.

IL28B genotype is the predictor for non-responder in GT-1 CHC patients treated with PegIFN/RBV, while baseline serum level of CCL4 is the only predictor for non-responder in GT-1 CHC patients with favorable IL28B genotype.