Objectives: Newly developed, direct-acting antiviral therapy is effective in over 90% of cases to eradicate hepatitis C virus infection. Direct-acting antiviral therapy is also effective in liver transplant recipients with recurrent hepatitis C virus infection. However, hepatic function after sustained virologic response in transplant recipients is unknown. Here, we aimed to uncover the incidence of hepatic dysfunction in this patient group at our center.
Materials and Methods: Our study included 40 consecutive (January 2014 to February 2016) and compliant posttransplant recipients who achieved sustained viral response from direct-acting antiviral therapy. Patients were investigated for incidence and causes of hepatic dysfunction.
Results: In our patient group, 4 (10%) experienced hepatic dysfunction with stable baseline immunosuppression, with 2 having drastic increases in alanine aminotransferase at 15 and 32 weeks after direct-acting antiviral therapy. Biopsies showed hepatitis, and both patients were treated with hydrocortisone, which increased their baseline immunosuppression. The 3rd patient had an increase in bilirubin at 21 weeks posttherapy, with biopsy showing macrovascular steatosis. The 4th patient had a rapid increase in bilirubin at 7 weeks after direct-acting antiviral therapy, with biopsy showing significant duct loss.
Conclusions: During the study period, 10% of patients experienced hepatic dysfunction after sustained viral response. Presumed causative factors included partial immune reconstitution and nonalcoholic fatty liver disease.
Key words : Chronic rejection, Immune system reconstitution, Immunosuppression, Nonalcoholic fatty liver disease
Chronic hepatitis C viral (HCV) infection is the most common indication for liver transplant.1-5 After successful liver transplant, biochemical and histologic recurrence of HCV in the allograft liver is almost universal. Furthermore, the progression of HCV-related histologic changes is much faster, resulting in allograft failure and retransplant.6 With newly developed direct-acting antiviral (DAA) therapy for HCV infection, the success rates for liver transplant recipients are comparable to those for nontransplant patients.7-9 However, the incidence and causes of hepatic biochemical abnormalities in liver transplant recipients post-DAA therapy after sustained virologic response (SVR) have not been widely reported.
In this study, our aims were (1) to examine the rate of SVR in our liver transplant patient population after DAA therapy, (2) to define the rate of hepatic biochemical abnormalities after SVR, (3) to correlate these biochemical abnormalities with immunosuppression and histologic changes, (4) to identify likely causative factors, and (5) to consider management and prevention strategies.
Materials and Methods
All liver transplant recipients who received DAA therapy for recurrent HCV at our transplant center were retrospectively examined after institutional review board approval. Existing electronic data bases for clinical management of the patient and mandatory compliance requirements by the United Network for Organ Sharing were utilized. After an eligible patient population was identified, the dataset was deidentified. During our study period (January 2014 to February 2016), 44 liver transplant recipients received DAA therapy as per the American Association of Study of Liver Disease and Infectious Diseases Society of America guidelines. Included patients were (1) those who received whole deceased donor hepatic allografts and (2) those who were followed for at least 1 year after start of DAA therapy. In addition, for this study, age, sex, HCV genotype, time from liver transplant to DAA therapy, and duration of DAA therapy were recorded. All liver biopsy findings had been reviewed by experienced pathologists.
Hepatitis activity index (HAI) and fibrosis score (as defined by Ishak and associates10) and rejection activity index (RAI; as per as per Banff criteria11) were recorded in addition to any other clinically relevant findings.
Response rates to DAA therapy were categorized as nonresponders (those who did not clear HCV RNA), end of treatment responders (those who cleared HCV RNA initially but relapsed within 12 weeks of completion of therapy), and SVR (those who maintained undetectable HCV RNA for more than 12 weeks after completion of therapy). Sequential hepatic biochemical parameters and immunosuppression changes were cataloged before, during, and after completion of DAA therapy. Any other known viral, anatomic, vascular, or biliary causes were ruled out. Other clinical variables such as noncompliance and weight changes were noted.
Our study population consisted of 36 male (81.8%) and 8 female (18.2%) liver transplant recipients with mean age of 54.9 ± 6.8 years. Mean time from transplant to DAA therapy was 54.5 ± 57.4 months. Thirty-eight patients had HCV genotype 1 (86.4%), 3 had HCV genotype 2, and 3 had HCV genotype 3 (Table 1).
Of 44 patients, 41 (93.2%) achieved SVR. One patient was a nonresponder (patient 11). Two patients relapsed after end of treatment response (patient 38 and patient 42). One patient discontinued her immunosuppressive medications after SVR (patient 13). Of the remaining 40 patients, 4 patients (10%) developed hepatic dysfunction changes after achieving SVR. The clinical course, biochemical changes, immunosuppressive findings, histologic manifestations, and management for each patient are described below.
Patient 9. Patient 9 was a 44-year-old man who had received sofosbuvir and ledipasvir 36 months after liver transplant (genotype 3A) for 24 weeks. A pre-DAA therapy biopsy showed portal inflammation with HAI (score of 6/18) and mild portal fibrosis (score of 1/6) (Figure 1b). His liver function tests (LFT) before DAA therapy showed alanine aminotransferase (ALT) level of 113 U/dL, aspartate aminotransferase (AST) level of 105 U/L, bilirubin level of 17.1 µmol/L, and alkaline phosphatase (ALP) level of 380 U/L. His immunosuppression regimen included cyclosporine 75 mg twice daily with trough concentrations ranging from 62 to 70 ng/mL and mycophenolic acid (MPA) 180 mg twice daily. His immunosuppression remained stable during and after therapy. Eight weeks after completion of antiviral therapy, his ALT increased from 74 to 200 U/L (Figure 1a). He was negative for cytomegalovirus (CMV) infection. His hepatic vascular anatomy was patent with normal biliary anatomy. His cyclosporine and MPA doses were increased, and an oral prednisone taper was commenced starting at 40 mg/day, which decreased by 10 mg every 3 weeks (Figure 1a, bottom).
Fifteen weeks after DAA therapy had started, his ALT level increased to 352 U/L. A biopsy showed focal portal and lobular inflammation (HAI score of 2/18) without fibrosis (Figure 1b). His ALT level slowly improved to 106 U/L. Thirty-two weeks after start of DAA therapy, ALT increased again to 470 U/L. A repeat liver biopsy was essentially identical to the previous biopsy. A second oral prednisone taper was repeated. His ALT level improved to baseline by 37 weeks after starting DAA. At 39 weeks post-DAA therapy, ALT increased for the third time to 241 U/L. He was converted to tacrolimus, and a third course of oral prednisone taper was initiated. By 10 weeks after tacrolimus conversion, his ALT normalized and remained stable at 12 months post-DAA therapy (Figure 1a).
Patient 20. Patient 20 was a 58-year-old man who had been receiving sofosbuvir and ledipasvir for 12 weeks at 11.8 months after transplant. A pretherapy biopsy showed portal and lobular inflammation (HAI score of 6/18) without any rejection or fibrosis (Figure 2b). Before the start of DAA therapy, his LFT showed ALT level of 92 U/L, AST level of 61 U/L, bilirubin level of 11.97 µmol/L, ALP level of 55 U/L, and gamma-glutamyltranspeptidase (GGTP) of 78 U/L. He was on tacrolimus 1 mg twice daily with trough concentration of 2.9 ng/mL and MPA of 180 mg twice daily. Four days before completion of DAA therapy, his ALT was elevated at 692 U/L and further increased to 1414 U/L by 14 days post-DAA therapy (Figure 2a). A biopsy showed portal inflammation with interface activity (HAI score of 7/18) and mild portal fibrosis (score 1/6) (Figure 2b). Similar to the previous case, he was negative for CMV and had patent hepatic vascular and normal biliary anatomy. His immunosuppression was stable during DAA therapy. He received 2.5 g of intravenous methylprednisolone over 4 days. Tacrolimus dose and MPA dose were increased, and oral prednisone was started. His ALT levels initially improved but elevated again to 1012 U/L at 27 days post-DAA therapy. A repeat liver biopsy was similar to the previous biopsy. Tacrolimus level was maintained around 10 ng/mL with MPA 720 mg twice daily and prednisone 20 mg/day. His liver function slowly normalized by 126 days post-DAA. At his last follow-up 25 months post-DAA therapy, LFTs remained normal with tacrolimus level of 5.1 ng/mL and MPA 360 mg twice daily without prednisone.
Patient 37. Patient 37 was a 50-year-old man who was 7 months posttransplant and was being treated with a 24-week course of sofosbuvir and ledipasvir. His LFTs before therapy showed bilirubin level of 25.65 μmol/L, AST level of 60 U/L, ALT level of 71 U/L, GGTP level of 35 U/L, and ALP level of 75 U/L (Figure 3a). Liver biopsy showed mild portal inflammation (HAI score of 3/18) without endotheliitis and fibrosis (score of 0/6) (Figure 3b). He was on MPA 180 mg twice daily with tacrolimus level of 9.7 ng/ mL (Figure 3a). Twenty-one weeks after start of DAA, his serum bilirubin level increased to 53.01 μmol/L (AST, ALT, ALP, and GGTP were normal) (Figure 3a). His tacrolimus level was 11.6 ng/mL. He was also negative for CMV with patent hepatic vascular and normal biliary anatomy. A biopsy showed steatohepatitis with mild macrovesicular steatosis occupying 25% of biopsy and perivenular and periportal fibrosis (Figure 3b). At 17 weeks before DAA, his weight was 92.5 kg and body mass index (BMI) was 30.7 kg/m2. This gradually increased to 132 kg with BMI of 43.1 kg/m2 at 21 weeks post-DAA therapy (Figure 3a) with a total weight gain of 39.5 kg (39.5%) and a BMI increase of 40.39% in a 62-week period. During this period, his HbA1c ranged from 7.3 to 8.9. At 1 year post-DAA follow-up, his weight had decreased to 125 kg and his BMI had decreased to 41.2 kg/m2. His bilirubin level had decreased to 37.62 μmol/L.
Patient 44. Patient 44 was a 58-year-old man who was at 9 months posttransplant (HCV genotype 1) and who had commenced a 12-week course of sofosbuvir and ledipasvir. His bilirubin level was 56.43 μmol/L, and his ALT level was 282 U/L (Figure 4a). A biopsy showed portal and lobular inflammation with apoptotic hepatocytes and focal bile duct injury (RAI score of 2/9, HAI score of 5/18, and fibrosis score of 1/6; Figure 4b). A diagnosis of recurrent HCV was favored. His immunosuppressive regimen included everolimus 4 mg twice daily with levels of 7.1 ng/mL, prednisone of 5 mg/day, and cyclosporine of 100 mg twice daily without detectable levels (Figure 4a). Thirty-nine days after start of DAA therapy, his bilirubin was elevated at 58.14 µmol/L and had rapidly increased to 388.17 μmol/L by 53 days after start of DAA therapy. He was negative for CMV with normal hepatic vasculature. He had a mild bile duct anastomotic stricture, which was dilated and stented before DAA therapy. The stent was subsequently removed. His immunosuppression was stable during the DAA therapy. Biopsy showed chronic ductopenic rejection with portal inflammation, periportal fibrosis, cholestasis, and 8 of 10 portal triads without interlobular bile ducts (Figure 4b). He received methylprednisolone of 2.7 g total intravenously over 9 days (Figure 4a). At 67 days after start of DAA therapy, he was switched to tacrolimus. However, he later went into multiorgan system failure, and life support was withdrawn.
Historically, treatment of HCV is more demanding and challenging in liver transplant patients than in nontransplant patients. The first challenge is immunosuppressive management. Immunosuppression plays a dual role. It not only reduces the host immune response required to protect the hepatic allograft from rejection but also allows a permissive environment for HCV to exist within the liver transplant recipient. The second challenge is to distinguish between HCV recurrence and allograft rejection from a liver biopsy.12 This is primarily because certain criteria outlined in the Banff scoring system for RAI11 overlap with the HAI scoring system.10 In addition, certain suggestions of features regarding C4d stains and acute rejection after liver transplant to differentiate from recurrent HCV have not been utilized.13,14
The idea of an algorithmic approach to the treatment of cellular rejection or recurrent hepatitis based on RAI and HAI score on biopsy has been neither explored nor accepted.12,15 Furthermore, interferon-based therapies in liver transplant recipients have not shown uniform results in many existing studies.16-20 Utilization of ribavirin with interferon has shown a better response rate than interferon-alpha alone in nontransplant patients.21 However, in the setting of liver transplant, this regimen has posed a different problem. That is, many liver transplant recipients have renal dysfunction as a result of calcineurin inhibitors (CNI), and ribavirin dosage was not adjusted based on renal dysfunction.22 This can result in higher withdrawal rates with increasing rate of anemia from hemolysis, requiring erythropoietin factor or even blood transfusions.23,24 The introduction of a first-generation DAA therapy for HCV (NS3-4A protease inhibitor) resulted in considerable drug interactions with CNI, adding further challenges to immunosuppression management.25 With the advent of a nucleotide NS5B analog (sofosbuvir) in conjunction with the DAA NS5A inhibitor (ledipasvir) or the NS3/4A protease inhibitor (simeprevir) with or without ribavirin, interferon-free regimens have been found to be much more potent and effective for various genotypes in pretransplant patients with SVRs, with response of more than 90% in many studies.26-30
The same interferon-free regimen has been successfully utilized in liver transplant recipients. The main focus of the reports has been to address the efficacy of the drug regarding SVR.7,25,31 However, data on the incidence of deterioration in hepatic dysfunction after SVR in liver transplant recipients are lacking. More importantly, potential causes of the biochemical deterioration and histologic changes in the allograft have not been studied in detail. Saxena and associates32 reported 4 cases of rejection in 347 liver transplant recipients. The incidence appears to be much lower than in our report. In our study population, 42% recipients had cirrhosis, with 53% of patients with hepatic decompensation before DAA therapy. Furthermore, there was a wide variation in ALT levels in the patients. However, the study did not account for incidence or causes of deterioration in hepatic dysfunction besides rejection.32
In our study of 40 consecutive compliant liver transplant recipients, patients had close longitudinal follow-up by the same team of physicians. Our study was specifically designed to look at the incidence of hepatic dysfunction after SVR from DAA therapy, to identify the cause, and to determine the most appropriate treatment. We found that 10% of our recipients developed abnormal liver function after SVR. The most striking finding of our study is that the biopsies did not show acute rejection as described by RAI in Banff criteria.11 However, the 2 patients (patients 9 and 20) with this dilemma responded to corticosteroids and augmentation in baseline immunosuppression. Another patient developed chronic rejection with an undetectable concentration of CNI (patient 44). Chronic rejection has been well-established in the absence of tacrolimus.33-35
We believe that the immunologic response of the host with HCV infection is impaired by defective function of lymphocytes, particularly CD4 and CD8. With the rapid eradication of HCV infection, there could be at least partial reconstitution of CD4 and CD8 function. Furthermore, the presence of baseline immunosuppression in liver transplant recipients impairs host immune response and could also affect HCV virulence. This is potentially a double-edged sword for transplant hepatocytes and could potentially affect biochemical changes in the allograft.
Lafrado and associates reported that mitogen-induced lymphocyte proliferation is lower in viral (hepatitis B and non-A, and non-B)-infected chimpanzees than in naïve animals.36 The adaptive immune response with HCV infection in the acute and chronic phases of hepatitis C viremia has been described by Bowen and associates.37 CD8 T-cell dysfunction with HCV infection has also been reported.38,39 In one study, it was suggested that infection with hepatitis C virus manipulated the immune system by disrupting both innate and adaptive immunity.38 The investigators claimed that the net liver damage from HCV infection depends on the balance between the host’s antiviral mechanism and the virus’ ability to subvert it.38 With the rapid eradication of HCV infection, it is conceivable that there could be at least partial reconstitution of CD4 and CD8 function, which could cause hepatocyte injury.
On further detailed examination of our patients, we observed some similarities between patients 9 and 20. It appears that the baseline maintenance of immunosuppression could have been a factor.
Posttransplant management of immunosuppression has changed over the past 2 decades. Previously, management of immunosuppression was based on improving hepatic biochemical parameters and previous history of rejection in a given patient. If biochemical parameters remain normal without any history of rejection, baseline immunosuppression is then gradually reduced. Many patients with indolent HCV show abnormal hepatic function; hence, reduction of immunosuppression can be delayed.18 Samonakis and associates summarized the role of immunosuppression with SVR and recurrent HCV in liver transplant recipients in a meta-analysis study.40 Sheiner and associates showed that patients who received antibody preparation and steroids performed poorly.41 Berenguer and associates suggested that a slower reduction of steroids is better in liver transplant recipients than a faster reduction.42 Over time, it was realized that HCV recipients need lower baseline immunosuppression.43 However, the rapid elimination of HCV in interferon-free regimens appears to improve the host’s immune system.
For the third case (patient 37), de novo nonalcoholic fatty liver disease (NAFLD) was the cause of increased LFTs, since baseline immunosuppression was adequate but there was an increase of weight with suboptimal glycemic control. Seo and associates showed that de novo NAFLD posttransplant is 19,38 times more likely when there is an accompanying increase in BMI of > 10% posttransplant.44 Recently, Hejlova and associates reported the prevalence of steatosis risk factors after liver transplant.45
In the fourth case (patient 44), baseline immunosuppression before DAA therapy included everolimus, prednisolone (5 mg), and cyclosporine without detectable cyclosporine levels. After SVR, the patient experienced a rapid rise in bilirubin level, and the liver biopsy showed chronic rejection with 80% bile duct loss. Before the tacrolimus era, chronic rejection was a well-known complication after liver transplant. Fortunately, the development of tacrolimus reduced the rate of chronic rejection under cyclosporine in most cases.33,46 Subsequently, primary liver transplant with tacrolimus showed virtual freedom from chronic rejection.35,47 Hepatic fibrosis posttransplant has also been reported with everolimus and low exposure to tacrolimus.48
After rapid eradication of HCV, it is possible that partial reconstitution of the host’s immune function with improvement in hypoactive CD4 and CD8 lymphocytes could have been a factor in hepatic dysfunction, as discussed in our first 2 patients. Currently, there are 2 available biomarkers. The Cylex Immuknow assay (Cyclex Inc., Columbia, MD, USA) provides a surrogate marker by ATP generation from stimulated recipient lymphocytes with phytohemagglutinin.49 The Pleximmune blood test (Plexision, Pittsburgh, PA, USA) measures the inflammatory immune response of recipient T cells to the donor in cocultures of lymphocytes from both inflammatory response sources of donor cytotoxic T cells.50
The most surprising part of our observation was that the liver biopsies were consistent with hepatitis without detectable HCV RNA and without signs of acute cellular rejection (patients 9 and 20). Despite this, both patients responded to steroid boluses and augmentation in immunosuppression. These immune markers could be utilized in future trials to adjust immunosuppression during and after DAA therapy.
Currently used DAA therapies in liver transplant recipients are just as potent and effective as those for nontransplant patients. However, liver transplant recipients, after achieving SVR, can exhibit worsened LFTs in up to 10% of cases. Our observations showed that this phenomenon is multifactorial. It is likely that, with the rapid eradication of hepatitis C viremia, there is a partial reconstitution of the host’s immune system. Patients who are on a stable lower maintenance immunosuppression regimens may benefit by increasing the baseline immunosuppression to allow a rapid improvement in viremia. Second, even with adequate baseline immunosuppression, prevention of weight gain with better glycemic control is crucial to avoid de novo NAFLD. More frequent monitoring of LFT with DAA therapy may allow early detection of abnormalities and prompt adjustment of immunosuppression. Further prospective studies with the use of immune function assays may provide better guidelines for the adjustment of immunosuppression during and after DAA therapy in liver transplant recipients.
Volume : 18
Issue : 3
Pages : 345 - 352
DOI : 10.6002/ect.2018.0127
From the 1Department of Surgery, the 2Department of Medicine, Division of
Gastroenterology, and the 3Department of Pathology, The Pennsylvania State
University, College of Medicine, Hershey, Pennsylvania; and the 4The
Pennsylvania State University, College of Medicine, Hershey, Pennsylvania, USA
Acknowledgements: The authors have no sources of funding for this study and have no conflicts of interest to declare. Parts of the data were presented at the American Society of Transplant Surgeons Annual State of the Art Winter Symposium (Miami Beach, FL, USA, January 26-29, 2017) and the Joint International Congress of ILTS, ELITA, and LICAGE (Prague, Czech Republic, May 24-27, 2017). We thank Nancy Sabb (transplant coordinator) for providing patient updates of clinical status, Michelle Carraher (data manager and analyst quality assurance) for providing help with patient demographics, and Jenna Weller (CRNP) for assisting with typing the manuscript.
Corresponding author: Ashokkumar Jain, The Pennsylvania State University, College of Medicine, Department of Surgery, Division of Transplantation, 500 University Drive, H062, PO Box 850, Hershey, PA 17033-0850, USA
Phone: +1 717 531 5921
Figure 1. Patient 9
Figure 2. Patient 20
Figure 3. Patient 37
Figure 4. Patient 44
Table 1. Patient Demographics