Insulin resistance, telaprevir, and virological response in hepatitis C: The debate must go on

Authors

  • Manuel Romero-Gómez M.D., Ph.D.,

    1. UCM Digestive Diseases and CIBERehd, Valme University Hospital, University of Seville, Sevilla, Spain
    Search for more papers by this author
  • José A. Del Campo Ph.D.

    Corresponding author
    1. UCM Digestive Diseases and CIBERehd, Valme University Hospital, University of Seville, Sevilla, Spain
    • Address reprint requests to: Manuel Romero-Gómez, M.D., Ph.D., Unit for Clinical Management of Digestive Diseases and CIBERehd, Valme University Hospital, University of Seville, Avda de Bellavista s/n, 41014 Sevilla, Spain. E-mail: mromerogomez@us.es; fax: +34 955 01 58 99.

    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

  • See Article on Page 1897

Abbreviations
Akt

protein kinase B

DM

diabetes mellitus

HCV

hepatitis C virus

HOMA-IR

homeostasis model assessment for insulin resistance

IFN-α

interferon-alpha

IL28B

interleukin-28B

IR

insulin resistance

LDL

low-density lipoprotein

LVPs

lipoviroparticles

mTOR

mammalian target of rapamycin

Peg-IFN

pegylated IFN

PTP1B

protein-tyrosine phosphatise

RBV

ribavirin

SOC-3

suppressor of cytokines-3

SVR

sustained virological response

TNF-α

tumor necrosis factor alpha.

The effect of metabolic abnormalities on the possibility of achieving sustained virological response (SVR) with antiviral therapy has been receiving considerable interest over the past decade. Insulin resistance (IR)-impaired SVR rate in patients with hepatitis C treated with pegylated interferon (Peg-IFN) and ribavirin (RBV)[1, 2] and the clearance of the hepatitis C virus (HCV) was associated with decreased risk of metabolic abnormalities and diabetes mellitus (DM) during follow-up.[3, 4] Furthermore, HCV interacts with the insulin-signaling pathway promoting IR subsequent to interaction with protein kinase B (Akt), with the mTOR (mammalian target of rapamycin)/FKBP12-rapamycin-associated protein pathway, and with peroxisome proliferator-activator receptor gamma. These interactions may be modulated by suppressor of cytokines-3 (SOC-3), tumor necrosis factor alpha (TNF-α), and protein-tyrosine phosphatase (PTP1B).[5, 6] PTP1B was down-regulated in infected cells treated with metformin,[7] and, when inhibited with pervanadate or knocking down PTP-1B, IFN-α response in vitro was restored.[8] Despite the wealth of molecular, clinical, and epidemiological knowledge, IR modulation did not become a priority in the management of hepatitis C in standard clinical practice. This could be a result, at least in part, of the lack of effect of adding metformin[9] or pioglitazone[10] to Peg-IFN and RBV in patients infected by genotype 1. Insulin sensitizers did not increase SVR significantly, despite improvement in insulin sensitivity. In the hepatitis C setting, IR appears to result from the sum of viral-induced and host-related IR. Danoprevir in monotherapy (excluding bias related to Peg-IFN use) has been found to improve insulin sensitivity in parallel with HCV RNA decline after 2 weeks of therapy.[11] Last, a recent meta-analysis showed several key aspects influencing the relationship between homeostasis model assessment for IR (HOMA-IR) and SVR: (1) selection and interaction between covariables. HOMA-IR showed a strong colinearity with variables such as steatosis or obesity, but also with gamma-glutamyl transpeptidase levels, triglyceride levels, age, and fibrosis[12]; (2) baseline biochemical characteristics of the cohort studied. The higher the baseline HOMA-IR, the better the HOMA-IR prediction; and (3) SVR rate; in easy-to-cure patients, such as those with mild fibrosis or those infected by genotype 2, or in patients with higher expected SVR such as those receiving potent antiviral agents, HOMA-IR would show less or null influence on SVR.

In the current issue of HEPATOLOGY, Younossi et al.[13] report on a negative post-hoc study of the effect of IR (measured by HOMA-IR) on SVR in genotype 1 patients who, previously, had failed with Peg-IFN and RBV, including relapsers, partial responders, and null responders. Patients were allocated to receive Peg-IFN/RBV as well as placebo or telaprevir-based triple therapy with or without a lead-in phase (4 weeks of Peg-IFN/RBV therapy). Patients receiving double therapy showed a strong association between baseline HOMA-IR and SVR. However, in patients receiving triple therapy, HOMA-IR level was found to be related to SVR in the univariate analysis, but not in the multivariate analysis. The selection of variables becomes crucial when addressing this type of multivariate analysis. Unfortunately, interleukin-28B (IL28B) genotype was not available in this study. Recently, HOMA-IR has been found to be independently associated with SVR, together with IL28B polymorphisms, fibrosis, and viral genotype in patients treated with dual Peg-IFN/RBV therapy.[14] In the study by Younossi et al., variable selection for inclusion into the study could not be done because of the post-hoc nature of the analysis. However, they demonstrated a strong colinearity between metabolic variables and HOMA-IR. The mean baseline HOMA-IR in this study was <3 (threshold to define the possibility of HOMA-IR influencing SVR) in all groups of patients, including relapsers (2.4), partial responders (2.7), and null responders (2.9). Furthermore, in a cohort of 859 veterans with genotype 1 with chronic hepatitis C (a third of whom had cirrhosis and nearly half with failure of previous treatment), treated with boceprevir- (n = 661) or telaprevir-based (n = 198) triple therapy, DM, and type of previous response to Peg-IFN/RBV were variables independently associated with end-of-treatment virological response.[15] This result supports conclusions from meta-analyses highlighting the influence of metabolic abnormalities on the possibility of achieving virological response in very difficult-to-cure patients. More data on the effect of diabetes on SVR are needed. Diabetes seems to be a barrier to triple therapy, but the effect of the correct management of diabetes in these patients needs to be demonstrated in future studies.

The interaction between the virus, lipid metabolism, and IR imply a complex network surrounding these factors having influence on SVR. HCV particles produced in primary hepatocytes had lower average buoyant density and higher specific infectivity, compared to HCV particles produced in cell cultures.[16] The infectivity of hepatitis C viral particles is inversely related to their density, and it has been established that lipoviroparticles (LVPs) are low-density HCV particles that have high infectivity, because LVPs may mask neutralizing epitopes.[17] IR was related to maximum LVP ratio and LVP density associated with SVR[18] in patients receiving IFN-based therapy, so that LVP ratio was found to be higher in null responders. All these interactions should be taken into account when explaining the association between high levels of circulating low-density lipoprotein (LDL) cholesterol and the raised SVR rate in treatment-experienced patients receiving telaprevir-based triple therapy.[19] Thus, metabolic alterations seem to play a role in the possibility of cure in patients with hepatitis C. However, demonstration of this fact remains elusive, probably as a result of adding specific virus-induced alterations and colinearity with variables associated with very difficult-to-cure patients. Understanding the complex relationship between HCV, host lipid, and glucose metabolism will lead to wider and new therapeutic options being introduced in patients without response to antiviral therapy. Nevertheless, the use of new direct antiviral agents reaching SVR in most of the patients will overcome the effect of metabolic factors on virological response. The debate remains open, and further studies are warranted to illuminate the last sprint of this controversial and enthralling topic.

Figure 1.

HCV-core protein promotes IRS1 degradation through a genotype-dependent mechanism (Akt/mTOR or SOCS-3) inducing IR and up-regulating TNF-α expression. NS5A plays a key role on host lipid metabolism interaction. Changes in the insulin-signaling pathway lead to IR, fatty liver, obesity, and type 2 DM, facilitating fibrosis progression to cirrhosis (left panel). Elevated LDL cholesterol levels, together with low-density circulating LVPs (HCV RNA density: ≤1.07 g/mL) and normal insulin sensitivity, were associated with raised success of protease inhibitor-based therapy (right panel). IRS1, insulin receptor substrate 1; NS5A, nonstructural protein 5A; APO-proteins, apolipoproteins; miRNA, microRNA.

  • Manuel Romero-Gómez, M.D., Ph.D.

  • José A. del Campo, Ph.D.

  • UCM Digestive Diseases and CIBERehd

  • Valme University Hospital

  • University of Seville

  • Sevilla, Spain

Ancillary