The multifaceted functions of ribavirin: Antiviral, immunomodulator, or both?

Authors

  • Mario U. Mondelli M.D., Ph.D., FRCP

    Corresponding author
    1. Department of Infectious Diseases, Fondazione IRCCS Policlinico San Matteo, Pavia, Italy
    2. Department of Internal Medicine and Therapeutics, University of Pavia, Pavia, Italy
    • Address reprint requests to: Mario U. Mondelli, M.D., Ph.D., F.R.C.P., Department of Infectious Diseases, Fondazione IRCCS Policlinico San Matteo, Via Taramelli 5, 27100 Pavia, Italy. E-mail: mario.mondelli@unipv.it; fax: +39 0382 526450.

    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

  • See Article on Page 1160

Abbreviations
HBV

hepatitis B virus

HCV

hepatitis C virus

IFN

interferon

IL

interleukin

ISGs

interferon-stimulated genes

NK

natural killer

NS

nonstructural protein

Peg-IFN-α

pegylated interferon-alpha

pSTAT

phosphorylated signal transducer and activator of transcription

RBV

ribavirin

STAT

signal transducer and activator of transcription

SVR

sustained virological response

Th

T helper

TRAIL

tumor necrosis factor–related apoptosis-inducing ligand

Treg

T-regulatory cell

Ribavirin (1-β-d-ribofuranosyl-1,2,4-triazole-3-carboxamide; RBV) is a guanosine analog with broad-spectrum virustatic activity originally developed for the treatment of severe respiratory syncytial virus infection.[1] The drug has since been empirically used, with varying degrees of proven clinical efficacy, for the treatment of several RNA virus infections, including Lassa fever and other arenaviruses, Crimean-Congo hemorrhagic fever, La Crosse encephalitis, influenza, adenovirus, and hepatitis E virus (reviewed in earlier studies[2, 3]). However, beneficial effects in these clinical settings are often controversial, and clinical trials are, for obvious reasons, largely missing.Various mechanisms of action of RBV have been suggested, including direct inhibition of viral RNA-dependent RNA polymerases, inhibition of host inosine monophosphate dehydrogenase, inhibition of viral capping enzymes, and, last but not least, induction of lethal mutagenesis (i.e., the point at which the number of mutations per genome is too large to allow viability of newly assembled virions).[4] However, the overwhelming success of RBV largely derives from its excellent performance in synergy with standard or pegylated interferon-alpha (Peg-IFN-α)-based therapies leading to dramatic improvement in sustained virological response (SVR) rates among patients with chronic hepatitis C virus (HCV) infection,[2, 5] which has been extended to combination treatment with first-[6, 7] and second-generation[8] direct-acting antivirals, and, very recently, to interferon (IFN)-free regimens.[9] Although the predominant mode of action of RBV in HCV infection was shown to consist in a modest and transient, even though significant, reduction of viral replication,[10] several lines of experimental evidence pointed to concomitant immunomodulatory effects, particularly on adaptive immune responses, by tipping the balance toward a T-helper (Th)1 cytokine profile, increasing IFN-γ production in vitro,[11] and reversing T-regulatory (Treg)-mediated suppression of CD4+ effector T cells through inhibition of interleukin (IL)−10 secretion by Treg cells.[12] However, these findings in vitro were not replicated in vivo, because IFN-γ-producing T cells were instead reduced in patients treated with IFN-α/RBV combination.[13] Interestingly, hepatic expression of interferon-stimulated genes (ISGs) is up-regulated in nonresponders and it has been shown that RBV decreases expression of ISGs, possibly restoring their responsiveness to exogenous IFN.[14] This interpretation is supported by a modest, but significant, increase in ISG up-regulation in patients treated with PEG-IFN-α/RBV, compared with IFN-α alone,[14] suggesting a role for RBV as a modulator of innate immunity. These findings indicate numerous potential clinical applications of RBV as an immunomodulator. However, a randomized, controlled clinical trial of Peg-IFN-α with or without RBV in a chronic DNA viral infection, such as hepatitis B virus (HBV), failed to show a beneficial effect of combination versus Peg-IFN-α monotherapy,[15] implying that RBV-mediated immunomodulation alone would be insufficient as adjuvant treatment of any viral infections. Whether this is a result of the fact that many ISGs are already down-regulated in chronic HBV infection so that RBV would provide no additional benefit remains to be determined.

Natural killer (NK) cells are a major component of the innate immune system that are instrumental in establishing a coordinated and efficient adaptive immune response in viral infections. Most studies suggest that they are functional and activated in the peripheral blood of patients with chronic HCV infection, but are polarized toward cytotoxicity as a consequence of chronic exposure to endogenous IFN-α, resulting in deficient IFN-γ secretion.[16, 17] This is caused by a prevalent phosphorylation of signal transducer and activation of transcription (STAT)1 over STAT4, which instead controls IFN-γ production.[18] The significance of this finding with respect to the natural history and pathogenesis of chronic hepatitis C is unclear, but it has been hypothesized that failure to produce adequate amounts of IFN-γ, a potent noncytolytic mechanism of virus control, would contribute to inability to eradicate HCV, whereas enhanced cytotoxic function would be responsible for persistent inflammation.[17] Along these lines, deficient IFN-γ production may also impinge on response to antiviral therapy. This view is certainly attractive, but current evidence suggests that NK cytotoxic function is involved in treatment outcome. Indeed, treatment of patients with chronic HCV infection with IFN-α/RBV treatment causes an almost sudden NK cell activation and cytotoxic function, as shown by increased tumor necrosis factor–related apoptosis-inducing ligand (TRAIL) and degranulation in the face of reduced IFN-γ production, which is in line with preferential STAT1 phosphorylation resulting from “binge” exposure to IFN-α.[19] Induction of a strong NK cytotoxic function correlates with early[19] and/or sustained[20] virological response and it is of interest that an increase in NK perforin content[20] and in the proportion of CD56dim/CD16+ NK cells[20, 21] before in vivo exposure to IFN-α/RBV correlated with subsequent SVR. The importance of creating a robust cytotoxic response in the liver to achieve SVR is also illustrated by the disappointing behavior of intrahepatic NK cells in the setting of HCV infection, showing reduced expression of TRAIL and degranulation ability, most likely caused by exhaustion or by the liver immunosuppressive environment with apparently normal intrahepatic NK IFN-γ production.[22]

It is clear from the aforementioned that studies designed to analyze baseline and on treatment NK cell phenotypic and functional changes identified features associated with SVR in patients submitted to conventional IFN-α/RBV treatment. However, a distinct role for RBV could not be ascertained, being masked by an exuberant IFN-α effect. In this elegant study, Werner, Serti, et al.[23] address, for the first time, the hypothesis that NK cell function may be influenced by exposure to RBV in vitro and in vivo before Peg-IFN-α/RBV treatment and report an interesting, and hitherto unrecognized, effect of RBV, which reduces phosphorylated signal transducer and activator of transcription (pSTAT)4 levels in NK cells with no effect on STAT1 phosporylation both in vitro and in vivo, resulting in increased inducibility of pSTAT4 and of pSTAT4-dependent IFN-γ production in response to IFN-α. This is also illustrated by the increased proportions of IFN-γ-producing NK cells in patients pretreated with RBV and in rapid responders during the second-phase virological response, which is known to be faster in patients treated with Peg-IFN-α/RBV, compared with Peg-IFN-α alone. Changes induced by RBV were independent from reduction of viremia as they were observed after in vitro incubation of peripheral blood mononuclear cells with RBV. The present study was not designed as a pilot trial to test whether RBV pretreatment would achieve higher SVR rates, but, rather, as a proof-of-concept study aimed at clarifying possible innate immune regulatory mechanisms responsible for RBV contribution to response to IFN-α-based therapy. In view of the compelling evidence provided in the Werner, Serti, et al. study,[23] it is tempting to consider the coincidence of pSTAT4 inducibility with improved second-phase virological decline as a proof of a predominant immunomodulatory effect of RBV over the purely antiviral mechanisms cited above. Despite this attractive hypothesis supported by data presented in this article, we are still left with the conundrum of the yet incompletely explained mode of action of RBV, at least in HCV infection. The versatility of RBV and its undeniable, albeit modest, antiviral effect on HCV strongly suggest that a combination of lethal mutagenesis and a supporting role in IFN-α-induced modulation of innate immunity are key mechanisms of a successful therapy outcome, whereas for other RNA virus infections, an indirect role as immune modulator would be obscured by a more potent direct antiviral effect. The current experience with first-generation nonstructural protein (NS)3 protease inhibitors clearly confirms the role of polymorphisms of IFN-λ3 in responses to triple therapies, suggesting that the efficiency of IFN-α/RBV-containing regimens are still dependent upon an efficient innate immune response. In view of Werner, Serti, et al.'s data,[23] it may be inferred that RBV's modulation of the NK cell response to IFN-α remains of interest also in the context of new IFN-free, RBV-containing regimens, and it is fascinating to think that RBV would be able to stimulate NK cells to mount an improved IFN-γ response to endogenous type I IFNs, induced by increased HCV replication during a viral breakthrough. However, the sophisticated countermeasures developed by HCV to interfere with type I IFN-signaling pathways (reviewed in a previous work[24]) cast doubts on the hypothesis that this is the mechanism through which RBV would improve immunosurveillance. Moreover, the evidence emerging from recently published clinical trials suggests that RBV may eventually be redundant in second- and third-generation, all-oral regimens, which will include combinations of sofosbuvir with NS5A and non-nucleosidic RNA polymerase inhibitors.[25, 26] Whether RBV will survive the imminent burial of IFN-α is difficult to say, but the promising results achieved by an all-oral sofosbuvir/RBV combination regimen,[27] particularly in treatment-experienced patients with severe fibrosis, imply that RBV will stay with us for quite a while and the additional mechanistic immunological insights provided by Werner, Serti, et al. (included in Fig. 1[23]) will certainly contribute to optimize treatment protocols.

Figure 1.

Effects of RBV on innate and adaptive immune responses. RBV lowers baseline ISG expression and resets IFN responsiveness in the liver, making the liver more permissive to the action of exogenous IFN. The drug also improves inducibility of pSTAT4 in NK cells during conventional IFN-α/RBV combination treatment, resulting in increased proportions of IFN-γ-producing NK cells. RBV effects on adaptive immunity mainly consist in switching CD4 T-cell responses toward a Th1 profile favored by RBV-mediated suppression of IL-10 secretion by Tregs, which results in improved efficiency of CD4 T-cell effectors secreting IFN-γ.

  • Mario U. Mondelli, M.D., Ph.D., FRCP1,2

  • 1Department of Infectious Diseases

  • Fondazione IRCCS Policlinico San Matteo

  • Pavia, Italy

  • 2Department of Internal Medicine and Therapeutics

  • University of Pavia

  • Pavia, Italy

Ancillary