Is there such a thing as protocol immunosuppression in liver transplantation?

Abstract

Protocol immunosuppression in liver transplantation is largely an outdated concept. Immunosuppression is now personalized to the individual patient on the basis of several factors including underlying etiology of original liver disease (e.g., HCV, hepatocellular carcinoma), renal function, metabolic co-morbidities and the patient’s immunological state. These include omission of corticosteroids in HCV infection and those with major metabolic risk factors, the minimization of calcineurin inhibitors in the presence of renal dysfunction and the use of mTOR inhibitors in patients with malignancy. The basis for such decision-making is discussed in this editorial.

Keywords:

• hepatitis C
• immunosuppression
• liver transplantation
• rejection
• renal failure

If it existed, what would protocol immunosuppression look like?

A historical view of immunosuppression (IS) suggests that a triple-drug regime of corticosteroids (CS)/calcineurin inhibitor (CNI) combined with an anti-metabolite, form the basis of modern IS in liver transplantation. Randomized controlled trials (RCTs) in the 1990s and early 2000s suggested that tacrolimus was a superior CNI to cyclosporine with less allograft rejection and possibly improved survival while mycophenolate mofetil (MMF) was superior to azathioprine in preventing allograft rejection but did not improve survival [1,2]. By 2004, the most common initial immunosuppressive regimen in the USA was either dual therapy with CS plus tacrolimus or triple therapy with the addition of MMF. The latter regimen was associated with fewer late allograft rejection episodes. Some groups, however, claimed similar outcomes with dual IS, induction with antithymocyte globulin and even CNI monotherapy [3]. Thus, protocol IS is often in the eye of the beholder rather than based on objective outcomes data. However, it is clear that currently few patients are in fact treated according to a single protocol and different aspects of the individual patient are now considered to determine the most appropriate immunosuppressive therapy. These distinct considerations are reviewed in this article.

Exception 1: renal dysfunction

It is not uncommon to adjust CNI doses in the presence of early and or late renal dysfunction. The approach, in general, has been to limit CNI exposure and add renal sparing agents such as induction antibody antithymocyte globulin or an anti-IL-2 receptor monoclonal antibody, anti-metabolites such as azathioprine or MMF (if the patient was not already on them) or a mammalian target of rapamycin inhibitor (mTORi) such as sirolimus or everolimus. So, what is the evidence that such approaches actually work?

Acute kidney injury post-liver transplant is associated with worse outcomes including progression to end-stage kidney disease and increased mortality and can often be anticipated in patients with either pretransplant acute kidney injury or chronic kidney disease [4]. The use of CNI at standard doses is an important factor in early renal dysfunction. In patients with pre-existing or early post-transplant renal dysfunction, and even in patients with normal renal function, RCTs investigating the use of IL-2 receptor monoclonal antibody induction (e.g., basiliximab on days 0 and 4), with delayed introduction of low-dose CNIs, have demonstrated an improvement in early renal function and reduction in the number of patients undergoing hemodialysis in the short term with no increase in acute rejection (a systematic review and meta-analysis has, however, questioned the need to introduce renal sparing agents versus simply aiming for reduced tacrolimus trough levels in the early post transplant period) [5]. In that review, trough levels between 6 and 10 ng/ml versus >10 ng/ml in the first post-operative month did not lead to an increase in acute allograft rejection but were associated with improved renal function at 1 year. Perhaps, if a significant delay in the introduction of CNIs post-transplant was required, then the use of an additional agent may be prudent: if not, then simply reducing CNI exposure may be enough.

Strategies to preserve renal function in the long term have included the use of mTORi agents. A recent RCT and meta-analysis of the available literature did not support conversion from CNI-based to sirolimus-based IS to improve renal function [6,7]. However, a recent RCT comparing sirolimus plus MMF to MMF plus CNI at 4–12 weeks post-transplant did lead to significantly improved renal function at 12 months, although with higher rates of biopsy-proven acute rejection and hyperlipidemia [8] RCTs using early introduction of everolimus at 4 weeks post-transplant with either CNI withdrawal or dose minimization showed significant improvement in renal function at 2 and 3 years [9]. It should be noted, however, that in the larger Phase III study, complete elimination of CNI was associated with more allograft rejection and that arm of the study had to be stopped prematurely [8]. In this study, patients who did not develop allograft rejection and continued on long-term everolimus without CNI had the best renal function at 3 years, compared to patients either on low-dose CNI plus everolimus or standard CNI.

With respect to patients with chronic renal failure post-liver transplant, it is generally well recognized that once there is established significant renal dysfunction, the percentage of patients who improve on CNI withdrawal is reduced; however, a switch to MMF or an mTORi with minimization or withdrawal of CNI has shown improvement in renal function in some studies reviewed [10]. Monotherapy with either an mTORi or MMF has been associated with a significant risk of allograft rejection and additional low-dose CS or CNI therapy may be appropriate in such circumstances.

Exception 2: HCV infection

Recurrent HCV infection is an important cause of graft loss and reduced survival post-transplant. CS have been implicated in HCV recurrence, and there is ongoing debate about whether CS should be removed from the IS protocol. It is recognized that outcomes in HCV recurrence are heavily influenced by the pattern of early HCV recurrence including the level of HCV replication in the first 3 months post-transplant. CS have been shown to increase HCV replication, HCV entry into hepatocytes and increase cell-to-cell spread [11]. All of these effects will contribute to higher HCV levels in the first instance and potential HCV cytopathic injury. Two meta-analyses have shown less severe HCV recurrence in CS-free protocols [12,13]; however, a large RCT showed no benefit of steroid-free IS in terms of allograft rejection, HCV recurrence or patient or graft survival [14].

Another issue of debate is whether cyclosporine or tacrolimus is better for HCV patients. A recent RCT indicated less severe recurrence in cyclosporine-treated patients but no difference in survival at 2 years [15]. In contrast, a retrospective review of almost 9000 patients in the United Network for Organ Sharing/Organ Procurement and Transplantation Network database suggested better outcomes in tacrolimus-treated patients [16]. There has been much discussion on the role of MMF in HCV infection post-liver transplant. A recent RCT also showed that an azathioprine-containing regimen was associated with less fibrosis and disease progression than tacrolimus monotherapy [17]. There has also been much discussion regarding the use of mTORi to slow down fibrosis progression in HCV patients. This concept was supported in a single-center analysis [18]; however, a review of the large US Scientific Registry of Transplant Recipients database showed that sirolimus-containing IS was associated with an increased risk of death and graft loss in HCV patients [19].

Exception 3: hepatocellular carcinoma

The major consideration here is whether the use of an mTORi can prevent the recurrence of hepatocellular carcinoma. A retrospective analysis of the SRTR database in the US suggested less recurrence in patients exposed to sirolimus. Two meta-analyses came to the same conclusion [20,21]. In addition, minimization of CNI exposure also seems to reduce recurrence. An approach with an analysis of molecular and cellular phenotype of the actual hepatocellular carcinoma tissue in the explant may in future guide individual IS in these patients.

Exception 4: metabolic risk

CS increase the likelihood of hypertension, diabetes and hyperlipidemia after liver transplantation. Meta-analyses have demonstrated a reduction in hypertension and diabetes with steroid-free IS [12,13]. Furthermore, it is well recognized that tacrolimus therapy increases the risk of diabetes [1]. Thus, it can be argued that a CS-free regimen without tacrolimus may provide the lowest metabolic risk.

Exception 5: extrahepatic malignancy

It is well known that liver transplant patients have an increased risk of extrahepatic cancer, particularly those related to viral infections [22]. High CNI exposure and long-term azathioprine use have been implicated in predisposition to extrahepatic malignancy [23]. There is some evidence that mTORi therapy may decrease the incidence of non-melanotic skin cancers.

Exception 6: autoimmune diseases

Recurrence of autoimmune hepatitis may be reduced with CS as part of the long-term IS protocol [24]. There is also evidence that recurrence of primary biliary cirrhosis is reduced on long-term cyclosporine rather than tacrolimus [25]. Autoimmune diseases, in general, may be more predisposed to severe allograft rejection, so under-IS is best avoided. The activity of inflammatory bowel disease in patients transplanted for primary sclerosing cholangitis may be influenced by the choice of IS protocol [26].

Exception 7: positive antibody cross match

Patients with a high titer of donor-specific HLA antibodies have more allograft rejection and occasionally lose the allograft to fulminant antibody-mediated rejection. The incidence of severe antibody-mediated rejection seems to be reduced if antibody induction is part of the IS protocol [27].

Exception 8: tolerance

Some patients may come off IS altogether without the loss of the liver allograft. This is uncommon, but in many series about 20–30% of patients without autoimmune disease, who have not had episodes of allograft rejection and who have been weaned onto long-term minimal IS, can do this. Recently, molecular signatures in blood leucocytes and the allograft itself were able to predict successful withdrawal [28]. Although the benefits of IS withdrawal are questioned, improvements in HCV-related allograft injury and metabolic risk have been observed in some series [29].

Conclusion

In conclusion, protocol IS is generally an outdated concept. IS protocols are now ‘tailored’ to the individual situation. It should be remembered, however, that such approaches should be on the basis of evidence. Where no evidence exists or is minimal, the original standard protocol of a CNI consisting of a CS and an anti-metabolite may still form a reasonable basis for IS that we should only deviate from with good indications and evidence.

Financial & competing interests disclosure

GW McCaughan has reported associations with Novartis, Astellas/Janssen and Roche Products. S Strasser has reported associations with Janssen and Roche Products. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.