Quantitation of pretreatment serum interferon‐γ–inducible protein‐10 improves the predictive value of an IL28B gene polymorphism for hepatitis C treatment response

Hepatitis C (HCV) is a single stranded RNA virus that usually establishes persistent infection in its host. Of patients exposed to HCV, about 80% will develop chronic viral infection characterized by liver infiltration of HCV-specific and nonspecific T cells accompanied by “proinflammatory” cytokines resulting in damage to virus-infected, as well as bystander, hepatocytes with resultant fibrosis formation. About 30–35% will develop cirrhosis and once a patient is cirrhotic, there is a 1–4% annual rate of hepatocellular carcinoma development (1). Combined treatment with peginterferon (pegIFN) and ribavirin achieves sustained virological response (SVR) in 42–52% of genotype 1 patients (2–4). Unfortunately, the remainder either fail to respond, or must discontinue treatment prematurely due to adverse events. Response rates to pegIFN and ribavirin are associated with both viral and host factors. Pretreatment predictors of non-response include genotype 1 infection, high viral load (>800,000 IU/mL), advanced fibrosis or cirrhosis, high body mass index (BMI), age older than 40, and African American race (2–4). Currently, on-treatment predictors of response to pegIFN and ribavirin include viral kinetics at weeks 4 and 12. Patients who do not attain an early virologic response have only a 1–3% chance of viral clearance and therapy is usually halted (2, 5). Conversely, 87% of those who achieve a rapid virological response (defined as HCV RNA undetectable at week 4 of therapy) achieve SVR (6). While viral kinetics have proven useful, better predictors of SVR and non-response would be helpful to identify patients with the best chance of response before the initiation of combination antiviral therapy. The US population has proven to be a more difficult group to treat than many others with lower SVR rates, perhaps due in part to higher BMI’s and a greater racial variation. African-Americans (AA) harbor predominantly genotype 1 virus and have notably lower overall response rates to pegIFN and ribavirin (approximately 26–28%) compared to Caucasian Americans (CA) (7–9). Determining why AA patients respond less well to antiviral therapy with pegIFN and ribavirin compared to CA was the focus of The Study of Viral Resistance to Antiviral Therapy of Chronic Hepatitis C (VIRAHEP-C). This NIH-sponsored study examined a variety of clinical, immunologic, virologic, and host genetic causes for the lack of response to treatment, although no single factor was identified that could account for the diminished rate of response (8, 10–12). Chemokines and cytokines are attractive as potential markers for treatment outcome as they are regulators of immunity and inflammation in HCV infection. Many are modulated by exogenous interferon and play critical roles in viral clearance. Responders tend to have a lower baseline activation of the immune system prior to treatment that is more markedly induced in response to IFN treatment (13–15). Several studies have shown that the non-ELR CXC chemokine interferon-γ (IFNγ) inducible protein-10 (IP-10 or CXCL10) may be a prognostic marker for HCV treatment outcome in genotype 1 infection (15–20). Elevated pretreatment IP-10 levels correlate with nonresponse to pegIFN and ribavirin therapy although this relationship has not been validated in AA patients. More recently, data has been published on a gene polymorphism (rs12979860) upstream of IL28B that is favorably associated with treatment response to pegIFN and ribavirin in both AA and Caucasian patients (21). Regardless of race, carriage of the C allele increases treatment response rates with those who have the CC genotype having the highest SVR rates, CT intermediate and TT the lowest (21). This favorable genotype is seen more frequently in Caucasian patients and likely explains about half of the difference in response between AA and CA of European ancestry. The IL28B gene encodes interferon-λ-3 which is a type III IFN induced by viral infections (22, 23). While the mechanism underlying the association of IL28B genotype and HCV clearance has not been elucidated, modulation of the innate immune response is likely playing a role in control of this viral infection. IL28B genotyping may provide useful pretreatment stratification of patients for HCV treatment in the future but it does not completely explain response discrepancies between AA and CA patients. In this study, we measured pre-treatment IP-10 levels in serum samples from 272 patients in the VIRAHEP-C cohort (115 non-responders and 157 SVR). This analysis demonstrated IP-10 to be equally predictive of SVR in both CA and AA patients. We then assessed the combination of pretreatment serum IP-10 levels with IL28B genotype as predictors of response to pegIFN and ribavirin in this cohort.