Immunosuppressive Therapy in Immune-Mediated Liver Disease in the Non-Transplanted Patient

Autoimmune liver disease management goals are primarily slowing disease progression and symptomatic treatment. There are few options for curative medical management other than transplant for a spectrum of autoimmune liver disease that encompasses autoimmune hepatitis, primary biliary cirrhosis, primary sclerosing cholangitis as well as their overlap syndromes. These diseases are managed primarily with immunosuppressive therapy. Herein, we review the current literature, detailing the promise and pitfalls of the recommended immunosuppressive therapy for these challenging diseases.


Introduction
Immunosuppressant therapy is a critical component of treatment in autoimmune mediated liver disease. Owing to the complex nature of immune mediated organ injury, immunosuppressive therapy has two principal effects, each resulting in specific benefits but also harms. Immunosuppressants modulate or abolish immune processes which have the effect of dampening autoimmunity but they also suppress normal immune function. The utility and power of these medications is predicated on the OPEN ACCESS clinician's ability to balance their benefits and harms. Herein, we review the benefits as well as the risks of standard pharmacotherapeutics of the following immune mediated liver diseases: autoimmune hepatitis (AIH), primary biliary cirrhosis (PBC), primary sclerosing cholangitis (PSC) as well as their overlap syndromes. median of 67 (12-128) months. Four patients had myelosuppression (multiple cell lines afflicted), 2 of which relapsed when the azathioprine was withdrawn. Many more developed lymphopenia. There is little international consensus on two important issues regarding this drug. First, in America, the common initial dose is 50-100 mg daily with prednisone during induction and 50-200 mg per day as monotherapy during maintenance. In Europe, a weight-based approach is favored with 1-2 mg/kg body weight depending on the phase of therapy. Second, there is no guideline consensus on whether to test for thiopurine methyltransferase enzyme activity to stratify for risk of toxic side effects. Clearly, however, the dose can be adjusted to optimize risks and benefits. One can also consider using 6-mercaptopurine (azathioprine is its pro-drug) or 6-thioguanine (the active metabolite of both), though the published experience is limited [8,9]. Alternate regimens derivative from experience in other diseases are many, though supporting literature may be sparse. Cyclosporine, a calcineurin inhibitor, is a promising agent for the 10% of patients with steroid refractory disease or those who are known to be corticosteroid intolerant [10]. For patients failed by front-line corticosteroid therapy, a trial of cyclosporine at 2-5 mg/kg daily in divided doses can be considered with positive outcomes in as many as 93% of patients, albeit in case-report level data [11][12][13]. It is not clear from the literature whether a particular trough level ought to be targeted, though 100-250 ng/mL is suggested by some groups [11,14,15]. The class specific side-effects of calcineurin inhibitors are notable and include hypertension and drug-level dependent renal dysfunction and neurotoxicity. However, amongst the pediatric population, cyclosporine use in AIH has been shown to have better tolerance with milder side effects including hypertrichosis and gingival hypertrophy with similar dosing and outcomes compared to adults [16,17]. Similarly, tacrolimus, a second-generation calcineurin inhibitor, has also been used. In a case-series from van Thiel et al. [18], 21 patients followed for a year received tacrolimus at 3 mg twice daily with median trough levels of 0.5 ng/mL resulting in a 70%-80% reduction in aminotransferases.
For patients with contraindications to or intolerance of azathioprine, a different antimetabolite drug, Mycophenolate mofetil (MMF), has been investigated. The published experience with this drug is mixed and dates to 1998 and 2000 with the first reports by Schuppan and Richardson [19,20]. Inductivo-Yu [21] reported that, at typical doses (1 gram twice daily), 15 azathioprine intolerant patients, three on MMF monotherapy and 12 on prednisone and MMF, experienced favorable outcomes with reductions in biochemical as well as histological markers. The following year, Hlivko et al. [22] reported their series of 29 patients who received MMF, 12 of whom were switched to MMF after intolerance or nonresponse to prednisone-azathioprine and 17 of whom received MMF-prednisone as initial therapy. Their results present a characteristically mixed picture. Of the 29 patients who received MMF, 34% discontinued it as a result of side effects while 16 (84%) of the remaining patients on MMF achieved remission. It appears that failure to respond to azathioprine predicts failure of MMF. Of the eight patients treated by Hennes et al. [23] with MMF who had prior nonresponse to azathioprine, six (75%) did not respond to MMF. Meanwhile, of the 28 patients in their series with azathioprine intolerance, remission was reached in 12 (43%). MMF's side effects include myelosuppression, notably leukopenia and gastrointestinal distress.
Finally, it is likely that ursodeoxycholic acid (UDCA) has immunomodulating properties. However, it is not particularly effective for AIH without an overlap syndrome. It can be used in combination with prednisone plus or minus azathioprine [2]. Czaja et al. [24] enrolled 37 subjects with refractory disease and randomized them to receive either placebo or 13-15 mg/kg of UDCA daily with corticosteroids. After 6 months of therapy, patients treated with UDCA were no more likely to have a dose reduction in corticosteroids or attain biochemical or histological improvement.

Primary Biliary Cirrhosis (PBC)
PBC is characterized by chronic bile duct inflammation resulting in cholestasis and liver injury manifesting most frequently as fatigue and pruritus with elevated serum alkaline phosphatase. Up to 90% of patients with PBC are women. The diagnosis of PBC is based on the presence of cholestasis, a positive antimitochondrial antibody (AMA) which is present in as much as 95% of cases or diagnostic liver biopsies [25]. Overlap syndromes have also been described between AIH-PBC.
Frontline therapy consists of UDCA for induction and maintenance therapy at a dose of 13-15 mg/kg divided into two daily doses ( Table 2). UDCA is the only FDA approved drug for treatment of PBC. A systematic review of 16 treatment trials including 1,447 patients showed biochemical and symptomatic improvement with UDCA, but failed to show any mortality benefit [26]. In a combined analysis of three randomized, placebo controlled trials performed in France, America and Canada including 548 patients, UDCA treated was associated with a reduced likelihood of liver transplantation or death after up to 4 years of follow up [27]. UDCA has been shown to slow disease progression [27][28][29]. At ten years, treated patients have significantly better survival than non-treated patients and only slightly lower survival than age-and sex-matched controls from the general population [27]. In general, most investigators argue that early initiation and prolonged follow up gives the best results. Expected side effects of UDCA include headaches and gastrointestinal symptoms. Several groups have looked at combination therapy with UDCA. Colchicine and methotrexate are the most common co-therapies. While early results were encouraging, methotrexate, generally given at 0.25 mg/kg weekly, has had mixed results over time. Bonis et al. [30] added low dose methotrexate to the treatment regimen (UDCA and colchicine) of 10 patients who were not responding to therapy with significant biochemical improvements. Kaplan et al. [31] compared patients with PBC who received UDCA plus methotrexate versus UDCA plus colchicine. No difference in survival was found and neither colchicine nor methotrexate improved transplant free survival above that predicted by a conventional risk assessment tool-Mayo prognostic model-which incorporates age, synthetic function (bilirubin, albumin and prothrombin time) and volume status (peripheral edema and diuretic use) [32]. Combes et al. [29] performed a randomized trial with 265 randomized to UDCA plus methotrexate and 133 to UDCA and placebo. In this study, methotrexate, where added to UDCA for a median of 7.6 years, showed no significant effect on disease progression or survival. The study was stopped early for futility.
Again, early data was supportive of colchicine. Colchicine, commonly at a dose of 1.2 mg daily in divided doses, has been shown most helpful in symptom control and can show biochemical improvement [31,33]. In a small trial of 90 patients randomly assigned to UDCA (at a suboptimal dose of 500 mg daily) with or without colchicine, Almasio et al. [34] showed that the addition of colchicine to UDCA resulted in a small but significant reduction of disease progression. However, as above, in the later trials, there was no benefit [31]. A systematic review studying 631 patients showed no improvement in mortality [35]. The balance of the evidence for colchicine so far has shown no improvement in disease progression [35][36][37][38][39].
In patients with UDCA-refractory PBC, additional therapy should be considered. Kaplan et al. [40] studied methotrexate and colchicine in UDCA refractory PBC. This study investigated 91 PBC patients who failed to respond to UDCA after 1 year of treatment were started on colchicine for 6 months and if there was no response then methotrexate (0.25 mg/kg per week) was added. Their results showed a significant improvement in biochemical tests and histology in 73 of 91 patients. Drugs such as prednisone, cyclosporine, azathioprine and D-penicillamine are generally not recommended owing to their toxic side effects and limited clinical experience [41][42][43][44].

Primary Sclerosing Cholangitis (PSC)
PSC is a chronic, progressive disease of the bile ducts. The pathogenesis is thought to be autoimmune. There is also strong association with inflammatory bowel disease. The long-term sequelae of PSC include cirrhosis and a 10%-20% lifetime risk of cholangiocarcinoma.
There is no FDA approved drug for PSC. This is consistent with 2010 AASLD guidelines which do not recommend UDCA for PSC treatment because of insufficient proven benefit on mortality or liver transplantation [45]. Many studies have investigated the benefit of UDCA. Earlier studies suggested that UDCA therapy (13-15 mg/kg) could attain biochemical improvement [46] (Table 3). However, by 1997, the results of high quality trials were less favorable when the Mayo PSC group published their randomized trial concluding that UDCA failed to improve outcomes even while it improved biochemical features [47]. Two large studies followed. First, Olssen et al. [48] reported their randomized controlled trial where 219 patients were randomized to either 17-23 mg/kg UDCA or placebo for 5 years. In this trial, the combined end point of death or liver transplantation occurred in 7.2% of patients receiving UDCA versus 10.9% of patients receiving placebo (p = 0.368). Lindor et al. [49] reported a multicenter, randomized placebo controlled trial in 2009 that was terminated early for futility. One hundred and fifty patients were enrolled and randomized to UDCA doses of 28-30 mg/kg or placebo. During therapy, biochemical tests improved in the UDCA group compared to placebo but these improvements were not associated with clinical benefit. The relative risk of a reaching primary endpoint was 2.3 times greater for patients on UDCA (30 versus 19 patients) and 2.1 times greater for death, transplantation, or minimal listing criteria (22 vs. 15 patients). Notably 15 patients on UDCA also developed esophageal or gastric varices compared to five on placebo.
There is insufficient evidence on the use of steroid, calcineurin inhibitors, methotrexate, azathioprine and penicillamine therapy.

Overlap Syndromes
Autoimmune hepatitis, PBC and PSC may co-exist as a so-called overlap syndrome. Treatment is derivative of published experience with the primary diseases. Overlap syndromes are described anywhere between 5%-20% of AIH patients by clinicians [50]. Below we discuss treatment options for overlap syndromes.

AIH-PBC
The data on treatment and even diagnosis of AIH-PBC is limited. The longest standing criteria, "Paris criteria" suggest that patients with overlap have PBC defined by: (i) alkaline phosphatase levels at least twice the upper limit of normal, (ii) positive AMA, and (iii) a liver biopsy showing florid bile duct lesion and AIH defined by (i) alanine transaminase levels at least five times the upper limit of normal, (ii) Immunoglobulin G levels at least twice the upper limit of normal or a positive anti-SMA, and (iii) a suggestive liver biopsy [51]. Some reports suggest that it may be underdiagnosed or not detected when the disease transitions from AIH to overlap. Confirming earlier data, Bonder et al. [52] argued this point in a retrospective, single institution review. In this study, several patients who did not meet Paris Criteria had a complete response to prednisone, confirming a major component of AIH [52,53]. Therapy is, at this point, a hybrid of front-line therapy for both conditions. Chazouilleres et al. [54] looked at 17 AIH-PBC overlap patients. Eleven received UDCA alone and six received UDCA and corticosteroids. Twenty-seven percent in the UDCA alone group showed a biochemical and immunological response whereas 67% in the combination therapy group showed biochemical response. Combination treatment with UDCA (13-15 mg/kg) and corticosteroids (+/− azathioprine) is recommended based on non-randomized non-blinded studies [50].

AIH-PSC
A study by Floreani et al. [55] reviewed seven patients (7/41 included patients) followed for at least 2 years. Diagnosis of AIH-PSC was made by: a total AIH score of more than 15; ANA or anti SMA of at least 1:40 and based on histology. Their results showed biochemical improvement and survival benefit with combination induction therapy with UDCA 15-20 mg/kg, prednisolone 0.5 mg/kg and azathioprine 2 mg/kg, followed by maintenance daily dosing with prednisolone 10-15 mg and 50-75 mg azathioprine plus UDCA as above stated dose.

Conclusions
In this review we give an overview of the use of immunosuppressants in chronic liver disease, specifically AIH, PBC and PSC. Immunosuppressants have many different functions ranging from slowing disease progression to purely symptomatic control and it is often a balance between clinical improvement and toxic side effects. This is a dynamic field with questions yet to be answered about the exact mechanism of disease and discovery of new treatments.

Conflicts of Interest
The authors declare no conflict of interest.