Of mice and men, calcineurin inhibitors and hepatitis C

Authors

  • Alleluiah Rutebemberwa,

    1. Division of Gastroenterology and Hepatology Hepatitis C Center, Department of Medicine, University of Colorado Health Sciences Center, Aurora, CO
    Search for more papers by this author
  • Hugo R. Rosen

    Corresponding author
    1. Division of Gastroenterology and Hepatology Hepatitis C Center, Department of Medicine, University of Colorado Health Sciences Center, Aurora, CO
    2. Liver Transplantation, Hepatitis C Center, Department of Medicine, University of Colorado Health Sciences Center, Aurora, CO
    3. Eastern Colorado VA, Denver, CO
    4. Integrated Program in Immunology, National Jewish Hospital, Denver, CO
    • Division of Gastroenterology and Hepatology (B-158), Department of Medicine, University of Colorado Health Sciences Center, 12631 East 17th Avenue, Academic Office Building 1, Room 7614, P.O. Box 6511, Aurora, CO 80045
    Search for more papers by this author
    • Telephone: 303-724-1858; FAX: 303-724-1891


  • See Article on Page 38

Abbreviations:

Alb, albumin; AFC8, FKBP-caspase 8 fusion gene; CNI, calcineurin inhibitor; Fah, fumarylacetoacetate hydrolase; HCV, hepatitis C virus; Hep, hepatocytes; HSC, human CD34+ stem cells, IFN, interferon; IL, interleukin; Rag2, recombination activation gene 2; SCID, severe combined immunodeficient; uPA, urokinase plasminogen activator.

Hepatitis C virus (HCV) continues to be a major challenge for global health care and affects approximately 200 million people worldwide.1, 2 Although antiviral therapy with pegylated interferon (IFN), ribavirin, and protease inhibitors in combination is effective in an increasing number of patients, HCV infection remains a common etiology of chronic liver disease, end-stage liver disease, and hepatocellular carcinoma and is the leading indication for liver transplantation throughout the world.3

As many excellent articles in Liver Transplantation have attested,4-6 reinfection after liver transplantation is universal, and its significance as a posttransplant clinical problem cannot be overstated.4-7 In simple terms, HCV recurrence remains the scourge of liver transplantation. Approximately 20% to 40% of HCV-positive patients who successfully undergo transplantation will develop allograft cirrhosis after only 5 years, whereas 3% to 20% will after 2 decades in a nontransplant setting.4 A number of donor, viral, and transplant factors have been shown to affect the severity of HCV recurrence as well as patient and graft survival. A consistent and longstanding finding is that the treatment of acute cellular rejection with corticosteroid boluses or OKT3 negatively affects HCV-infected transplant recipients.8, 9 However, one of the most hotly debated issues is the differential impact of the calcineurin inhibitors (CNIs) cyclosporine and tacrolimus, which remain the cornerstone of immunosuppression after liver transplantation, on the natural history of recurrent HCV.

Both CNIs interfere with signal 1 T cell signal transduction.10 Cyclosporine engages the immunophilin cyclophilin and forms a complex that then engages calcineurin.11 In turn, this complex inhibits the transcriptional activity of several genes, including interleukin-2 (IL-2), which is critical to T cell activation.10, 11 Engaging a different immunophilin (FK506-binding protein 12), tacrolimus leads to the creation of a complex that also inhibits calcineurin but with greater molecular potency.11 The use of tacrolimus has increased steadily, and being administered to more than 80% of patients, tacrolimus has become the dominant CNI. However, transplantation programs consider the strengths of both tacrolimus and cyclosporine for the individual patient.

To liver transplant recipients with recurrent HCV, cyclosporine offers 2 potential advantages in comparison with tacrolimus. First, cyclosporine directly suppresses HCV replication in vitro by binding to cyclophilin B and inhibiting HCV RNA polymerase.4 However, this antiviral effect is weak and appears limited to HCV genotype 1b in vitro.12 Second, unlike cyclosporine, tacrolimus can indirectly enhance HCV replication in vitro through the inhibition of the phosphorylation and nuclear translocation of signal transducer and activator of transcription 1 and thereby block IFN signaling pathways.4, 13 The latter observation, however, was recently challenged by Pan et al.,14 who found that tacrolimus did not interfere with the IFN-induced suppression of HCV in a subgenomic replicon model or with the de novo production of viral particles in the infectious HCV model.

The species restriction of HCV to humans and chimpanzees has limited the ability to study HCV immune responses in vivo. Unlike humans, HCV-infected chimpanzees typically develop minimal liver disease.15, 16 In an attempt to overcome these limitations, both transgenic mice and chimeric mice have been developed.17-19 Transgenic murine models are valuable for studying the consequences of HCV protein expression and overexpression.20 However, the expression of HCV proteins that is observed in these mice is not regulated in the same way as natural infections.20 In the past decade, many limitations of these models have been overcome through the development of chimeric mice whose livers are composed of human hepatocytes and that are capable of sustaining the entire HCV reproductive cycle in vivo. To date, several chimeric mouse models have been developed. The severe combined immunodeficient (SCID)/albumin (Alb)–urokinase plasminogen activator (uPA) mouse, which is homozygous for both the SCID trait and the Alb-uPA transgene,19, 21, 22 can undergo transplantation with human hepatocytes and is capable of supporting high levels of human chimerism within the liver.23, 24 Upon inoculation with infectious viral particles, this type of mouse is able to mimic HCV infections seen in the human host for extended periods of time (even for the life span of the animal).19 The SCID/Alb-uPA mouse model has been successfully used to evaluate anti-HCV drugs such as IFN-α, anti-nonstructural 3 (NS3) protease,25-27 and cyclophilin inhibitor candidates28 and even passive immunoprophylaxis. In the second chimeric murine model, fumarylacetoacetate hydrolase (Fah)–deficient mice, which have been proven to be a reliable model of liver repopulation with transplanted cells,29, 30 are crossed with recombination activation gene 2 (Rag2)/common γ-chain–knockout mice; this produces a triple-knockout mouse strain (Fah−/−/Rag2−/−/Il-2rg−/−) that is capable of supporting the growth of human hepatocyte grafts.31 In these immunodeficient mice, HCV propagation can be sustained for more than half a year.32 Both models represent significant steps forward in our ability to characterize HCV replication in vivo and evaluate novel antiviral approaches. However, neither model demonstrates a functional immune system, and this precludes the analysis of HCV pathogenesis. Very recently, a murine model supporting the efficient engraftment of both human immune and liver cells was developed; remarkably, in these AFC8-hu HSC/Hep mice, HCV infections induced HCV-specific human immune responses, liver infiltration, hepatitis, and even fibrosis.33

In the current issue of Liver Transplantation, Kneteman et al.,34 who pioneered the development of the SCID/Alb-uPA mouse model, set out to define the impact of CNIs (administered by oral gavage) on serum HCV titers in the presence or absence of subcutaneous IFN therapy in vivo. Chimeric mice were infected with HCV and subsequently treated with either cyclosporine or tacrolimus with or without IFN-α2b. The blood levels of the CNIs were comparable to the target levels in clinical practice. Graphing the changes in the log HCV titers, they were unable to demonstrate any beneficial effects of cyclosporine versus tacrolimus on HCV replication in vivo, and there was no additive effect when cyclosporine with IFN was compared to IFN alone.

Several limitations of the study are worth highlighting. The HCV RNA titers were statistically lower in the control group versus the non–IFN-treated arm, and there was a trend toward higher baseline HCV RNA levels in the cyclosporine arms versus the tacrolimus arms. On the other hand, in prior studies, the investigators have not seen a difference in response to therapies that was dependent on the initial viral titer when it was 104 to 107 IU/mL (the differences between the CNI groups were smaller than this). Clearly, the study was underpowered, particularly because of the inherent variability of the baseline HCV titers, which might be attributed to mouse-to-mouse differences in the levels of human hepatocyte chimerism. Each mouse was used as its own control, and the primary endpoint was a 1-log drop in the viral titer. The engrafted liver acted as an absorption column for the virus and was, of course, the source for new virus entering the circulation. The intrahepatic viral load was not determined in this study because liver sampling is typically a terminal event. Moreover, as explained previously, this type of mouse is immunodeficient and does not develop histological hepatitis. Emerging and provocative data from a number of laboratories around the world, including data about the roles of IL-28B polymorphisms and innate and adaptive immunity in the natural history of HCV and the IFN-based treatment response after liver transplantation, underscore why this murine model is not able to entirely address the ways in which CNIs, which have broad effects on many cell types, affect HCV after transplantation (other than replication).7, 35 A more appropriate model for sorting through the mechanisms by which cyclosporine or tacrolimus might offer relative benefits might be the AFC8-hu HSC/Hep mouse model with humanized immune and liver cells and histological hepatitis, as described previouly.33 On the other hand, the AFC8-hu HSC/Hep model has a limited level of engraftment (∼15%) that may be responsible for a complete lack of viremia; in the uPA and Fah models, greater than 50% human hepatocyte engraftment is typically necessary for detecting HCV in the blood. However, the point may be moot for several reasons. The initial published experience, which was derived largely from retrospective analyses, demonstrated a clinical benefit of cyclosporine for patients treated with IFN36, 37; however, the enthusiasm for cyclosporine has been tempered by more recent prospective data that have failed to show a clear-cut benefit of one CNI versus another.38 In the absence of such data, it will be challenging to strongly endorse cyclosporine versus tacrolimus. Furthermore, for the management of HCV infections, we have now entered a new age of triple therapy with significantly improved efficacy in both naive patients and previously treated patients.3 Although no data are yet available for liver transplant recipients, in a pharmacokinetic study of healthy volunteers, coadministration with steady-state telaprevir increased the dose-normalized area under the curve approximately 4- to 5-fold for cyclosporine and 70-fold for tacrolimus.39 The lower bioavailability of tacrolimus in healthy volunteers may make tacrolimus more susceptible to cytochrome P450 3A inhibition during first-pass metabolism.39 On the basis of these data, the first impulse might be to switch transplant recipients to cyclosporine if triple antiviral therapy is being considered, but it is important to recognize that the bioavailability of cyclosporine varies immensely with the patient population (eg, <10% in liver transplant recipients).39 Further studies are obviously warranted, but in the end, the choice of a CNI may be driven not so much by its effect on HCV replication but rather by drug-drug interactions with anti-HCV therapy.

Ancillary