Mycophenolate mofetil monotherapy in liver transplant recipients: A single center experience
The long-term use of calcineurin inhibitors (CIs) is associated with significant morbidity in liver transplant recipients. Although mycophenolate mofetil (MMF) is well tolerated, two small studies reported an unacceptable rate of acute allograft rejection in liver transplant recipients receiving MMF monotherapy. In this study, we retrospectively investigated the safety and efficacy of MMF monotherapy in liver transplant recipients. We reviewed the medical records of all patients who underwent liver transplant at our institution. Sixteen patients were identified who received MMF either as monotherapy (n = 13) or with corticosteroids (n = 3; 2 of them for other comorbid conditions), and these patients were studied to determine the efficacy and complications. Fifteen (15/16) patients were converted from a CI to MMF because of renal insufficiency. Patients were converted to MMF monotherapy after a median of 2,056 days (range, 606-5,893) after liver transplantation. The median postconversion follow-up was 668 days (range, 60-1,509). Four patients required dialysis despite conversion; of those patients not requiring dialysis, serum creatinine stabilized and showed a trend toward improvement (2.51 ± 1.12 mg/dL to 1.85 ± .58 mg/dL, P = .1). However, there were 3 episodes (47, 107, and 1,203 days after conversion) of severe, irreversible allograft rejection after conversion resulting in death in 2 patients and necessitating retransplantation in 1 patient. There were no patient characteristics, except perhaps African-American race, that predicted the development of rejection. In conclusion, MMF monotherapy was associated with a significant risk (19%) of unpredictable, severe, and irreversible allograft rejection even among long-term transplant survivors. Caution should be exercised before converting patients to MMF monotherapy. (Liver Transpl 2004;10:1189–1194.)
Standard immunosuppression after liver transplantation typically includes a calcineurin inhibitor (CI)—that is, tacrolimus or cyclosporine—with or without corticosteroids. This regimen affords a patient and graft survival rate of approximately 85% at 1 year and 77% at 3 years.1 However, tacrolimus and cyclosporine are associated with a considerable risk of neurotoxicity, nephrotoxicity, and diabetes mellitus; this often requires dose adjustment or discontinuation.2–4
Mycophenolate mofetil (MMF) is a morpholinoethyl ester prodrug of mycophenolic acid. Mycophenolic acid inhibits the proliferation of B and T lymphocytes by reversible inhibition of inosine monophosphate dehydrogenase, an enzyme in the synthetic pathway of guanine nucleotides.5, 6 A reduced incidence of acute rejection was reported in both renal and cardiac transplant recipients when MMF was used in combination with CI.7, 8 The role of MMF in the liver transplant population has traditionally been in the immediate posttransplant period. However, recently it has been used as a CI-sparing drug in patients who developed renal insufficiency.9–12 Although MMF monotherapy is a very attractive option for liver transplant recipients, the efficacy of MMF as monotherapy has been questioned. A randomized controlled trial was prematurely discontinued when acute cellular rejection occurred in 5 of 14 patients converted from CI to MMF monotherapy, compared with no rejection episodes in the “control” patients.10 In another case series, severe acute allograft rejection was reported in 3 of 5 patients on MMF monotherapy, and 2 of these patients required retransplantation.13 In this retrospective analysis, we determined safety, effects on renal function, risk of rejection, and patient survival after conversion from a CI to MMF monotherapy.
Patients and Methods
We performed a retrospective chart review of all adult patients who had liver transplantation between April 1985 and November 2003 at our institution. Of the 562 patients, 16 (6 male, 10 female) were identified whose immunosuppressant regimen consisted of either MMF alone (n = 13) or MMF with the addition of low-dose (≤10 mg/d) prednisone (n = 2), or hydrocortisone for adrenal insufficiency (n = 1). We collected data on demographics, renal function before and after conversion to MMF, number of rejection episodes and severity of rejection, and graft and patient survival.
The study was conducted after obtaining approval from the local institutional review board.
Individual patient (5 female; 14 White American, 2 African-American) characteristics are summarized in Table 1. The median age at the time of transplant was 48 years (range, 25–65 years). All transplants received cadaver transplantation with the exception of patient 16, who received a live-donor transplant from his son. The median interval from liver transplantation to MMF monotherapy conversion was 2,056 days (range, 606–5,893 days), while the median follow-up after conversion to MMF was 668 days (range, 60–1,509 days). Fourteen of the 16 patients underwent a single liver transplant. One patient (patient 15) required a second cadaver transplantation 2 days after his first due to primary graft nonfunction. A second patient (patient 10) with recurrent hepatitis C developed acute rejection that progressed to chronic rejection following MMF conversion and received a combined liver (retransplantation) and kidney transplant 2,108 days after his first (primary) transplantation.
Table 1. Characteristics of Patients Who Were Converted to MMF
|3†||49||M/W||Alcohol||5154||645||3.4||5.4||C, M, S||M|
|7†||47||M/W||Sarcoid||2179||886||2.3||2.4||C, S||M, S|
|12||45||M/W||HCV/HBV||683||1509||4.8||1.7||S, T||M, S|
|13||61||M/W||Alcohol||1079||1026||1.4||1.4||C, S||M, S|
The majority of patients (15/16) were on tacrolimus or cyclosporine immediately prior to conversion to MMF. One patient (patient 2), who was not on a CI immediately prior to introduction of MMF, discontinued rapamycin due to infection with histoplasmosis. Of the 3 patients who remained on corticosteroids in addition to MMF at last follow-up, 2 had steroid-responsive medical conditions (sarcoidosis in patient 7 and adrenal insufficiency in patient 12).
Renal insufficiency was the indication for discontinuation of CI in 15 of 16 patients. Four patients required dialysis during or after conversion to MMF. Patient 1 had gradually deteriorating renal function, requiring dialysis in 20 months despite initiation of MMF monotherapy; renal biopsy in this patient revealed changes consistent with hypertensive nephrosclerosis. Patient 3 had received a cadaver renal transplantation 5 years prior to his liver transplant due to end-stage renal disease from hypertensive and diabetic nephropathy. His transplanted kidney failed 5 months after withdrawal of CIs due to biopsy-proven de novo focal segmental glomerulosclerosis on a background of chronic transplant glomerulopathy. Patient 14 required the initiation of dialysis at the time of conversion to MMF due to presumed CI nephrotoxicity. Patient 10 required dialysis 5 months after conversion to MMF monotherapy and received a renal transplant 4 months later. All patients except patient 10 remained on dialysis at last follow-up. Patients who required dialysis were converted from CI to MMF at a longer interval from the time of transplant (median 3,503 days) than those who did not require dialysis (median 2,056 days; P = .1). Of the patients not requiring dialysis, serum creatinine stabilized and showed a trend toward improvement (2.51 ± 1.12 mg/dL immediately prior to conversion to 1.85 ± 0.58 mg/dL at last follow-up; P = .1).
There were 3 episodes of biopsy-proven acute cellular rejection in 2 patients from the time of transplant to the time of conversion to MMF, and there were 3 episodes in 3 patients after conversion to MMF at 47, 107, and 1,203 days after conversion. None of the patients with rejection after conversion had a history of biopsy-proven rejection prior to conversion. Patient 4, who was over 13 years post-liver transplant for cryptogenic cirrhosis, developed rejection 1,203 days after a stable clinical course on MMF monotherapy. She was initially noted to have an asymptomatic elevation in aminotransferases between 600 and 700 IU/L without cholestasis. Her initial liver biopsy revealed severe hepatitis with intense lymphoplasmacytic inflammation. Oral prednisone was started, and her MMF dose was increased. Although aminotransferases improved, she developed severe cholestasis (total serum bilirubin 12–13 mg/dL). A second liver biopsy 3 months after the first revealed moderate chronic portal inflammation and moderate lobular activity with focal endothelialitis and bile duct injury, consistent with acute cellular rejection, but without evidence of chronic rejection. Oral prednisone was continued with the addition of rapamycin. Because of concurrent fever and pneumonia, high-dose steroids or antilymphocyte preparations were not used. Valganciclovir was started due to a positive cytomegalovirus blood culture, but an immunostain for cytomegalovirus in liver tissue was negative. A workup for posttransplant lymphoproliferative disease, graft-versus-host disease, and biliary obstruction was negative. Cholestasis progressed, and she died 4 months after her second liver biopsy due to overwhelming sepsis with evidence of liver failure.
Patient 10 (primary disease: hepatitis C [HCV]) was diagnosed with rejection 107 days after conversion to MMF monotherapy. He presented with jaundice and was noted to have a total bilirubin of 12 mg/dL with aminotransferases approximately 200 IU/L. Liver biopsy revealed an intense chronic portal inflammation with moderate periportal and focal lobular hepatitis with mild focal venous endothelialitis. These findings could not reliably differentiate between acute cellular rejection and recurrent hepatitis C. He was started on Neoral and maintained on MMF at a reduced dose. Despite improvement in aminotransferases, bilirubin continued to increase. Rapamycin was added as salvage, but because of progressive cholestasis, liver biopsy was repeated. This again showed portal inflammation, cholestasis, and focal venous endothelialitis, and pulse steroids were initiated. While receiving triple immunosuppression (steroids, rapamycin, and tacrolimus), the patient developed bacteremia and a bilobar pneumonia requiring intubation. He required retransplantation 5 months later. Histological evaluation of his explant revealed chronic portal inflammation with bile duct epithelial injury, and severe obstructive intimal fibrosis of intrahepatic arteries, consistent with chronic rejection.
Patient 15 (primary disease: HCV), had evidence of recurrent cirrhosis at the time of initiation of MMF monotherapy. His bilirubin was approximately 2 mg/dL with only mild disturbances (<2× upper limit of normal) in aminotransferases. He re-presented 47 days after conversion with change in mental status and jaundice. Labs were notable for aspartate aminotransferase (≈700 IU/L), alanine aminotransferase (≈350 IU/L), and bilirubin (≈27 mg/dL); liver biopsy showed cirrhosis with moderate portal chronic inflammation and mild bile duct injury, consistent with acute cellular rejection. Although the aminotransferases improved with initiation of pulse steroids, his hyperbilirubinemia and coagulopathy worsened, and he died of multisystem organ failure and massive gastrointestinal bleeding.
There were no significant differences in age at transplant, sex, or time interval (2,336 ± 1,210 vs. 2,501 ± 1,823 days) from transplantation to conversion to MMF monotherapy between patients with and without rejection after MMF conversion. Patients with rejection after conversion received lower doses of MMF than those in whom rejection did not occur (mean 1,333 ± 577 mg/d vs. 1,923 ± 706 mg/d; P = .15), although this difference did not reach statistical significance. However, it is worth noting that both African-Americans (1 with cryptogenic cirrhosis and 1 with HCV) in our cohort (patients 4 and 15, respectively) developed acute rejection after conversion and died from it.
The primary liver disease recurred in 5 patients on MMF monotherapy, all of whom underwent transplantation for cirrhosis due to hepatitis C. Of these 5 patients, 4 had moderate to severe recurrence of hepatitis C, and only 1 patient had minimal histological changes (patient 2). Patient 2 has had only mild histological activity and mild fibrosis on serial protocol liver biopsies over 5 years in the setting of normal liver function tests and has not required antiviral therapy. Patient 10 required retransplantation 9 months after conversion to MMF due chronic rejection and had evidence of recurrent hepatitis C at the time of retransplantation. Patients 14 and 15 died at 705 and 1,503 days after transplant, respectively, due to sepsis with histological evidence of cirrhosis. Patient 16 was lost to follow-up 2 years after his transplant, at which time he had evidence of recurrent hepatitis C with a cholestatic hepatitis and coagulopathy.
Laboratory data prior to conversion were compared to those at last follow-up. We excluded 3 patients who died, in an effort to avoid confounding variables, such as infection, sepsis and hypotension, that would affect the laboratory tests at last follow-up. There were no statistically significant differences between pre- and postconversion laboratory values (Table 2). Except for serious rejection episodes in 3 patients, MMF was well tolerated with minimal side effects. Only 2 patients (patients 16 and 10) developed symptoms that were attributed to immunosuppression. One patient had nausea in the setting of narcotic use for a chronic pain syndrome, and the second had mild diarrhea. At last follow-up, however, both remained on MMF.
Table 2. Laboratory Values Before and After Conversion to MMF
|TB (mg/dL)||3.14 ± 8.32||1.82 ± 3.6||.6|
|AST (IU/L)||38.9 ± 44.3||45.2 ± 50.6||.8|
|ALT (IU/L)||37.2 ± 30.5||46.5 ± 55.0||.6|
|AP (IU/L)||121 ± 57.1||190 ± 233||.3|
|Alb (g/dL)||3.89 ± 0.56||3.94 ± 0.47||.8|
MMF is an immunosuppressant with reversible antiproliferative effects on lymphocytes.14 When used in combination with CI, MMF has shown beneficial effects in the prevention of acute liver allograft rejection15–17 and in the rescue of acute and chronic rejection.9, 15, 16, 18, 19 Previous small studies have shown unacceptably high rejection rates of 36 to 60% with MMF. In our study, approximately 20% of patients developed serious rejection, resulting in 2 deaths and one retransplantation. Although rejection rate in our series was lower than 2 previous small studies (36% and 60%),10, 13 our observations reinforce the need for extreme caution when MMF monotherapy is considered even in long-term liver transplant survivors.
An important question that we could not answer in the present study was whether we could have predicted those patients who developed serious rejection with MMF monotherapy. There were many limitations to answering this question, including the retrospective nature of our study and the small sample size. These patients were selected by their treating hepatologist for MMF monotherapy at least in part because of a previously stable clinical course; moreover, these patients had no prior history of rejection. Patient characteristics, age at transplant, and the interval between liver transplant and conversion were similar in those with and without postconversion rejection. Dosages of MMF were also similar, although there was a non-statistically significant trend of higher MMF doses in patients who did not have rejection. Although a much larger sample would be required to make a firm statement, our observations may suggest that the development of rejection and the interval from conversion to rejection are both unpredictable. Therefore, patients who are converted to MMF monotherapy should be followed closely for an indefinite period. One interesting finding in our study was that both African-Americans in our cohort developed severe rejection. Although it is difficult to draw any conclusions from this small study, it is prudent to observe African-Americans more closely when switching from one class of immunosuppression to another. Another uncertainty is whether a more aggressive approach in the treatment of patients with biopsy-proven rejection could have prevented death or the need for retransplantation. In our series, the use of high-dose pulse steroids or antilymphocyte therapy was limited by concurrent infection and the presence of HCV. In 1 patient, rejection occurred more than 3 years after conversion; in this patient, it is possible that early aggressive treatment may have changed the outcome.
There is significant interindividual20, 21 and intraindividual22, 23 variability in the pharmacokinetics of mycophenolic acid (MPA), the active form of MMF. Disease states and patient characteristics can both alter drug concentrations. MPA free fraction and MPA area under the curve are both increased in renal dysfunction24, 25 and with concurrent use of tacrolimus,26 while cyclosporine has been shown to decrease MPA serum concentration.27 A major limitation of our study was that we did not measure MPA concentration in any of our patients; therefore, we do not know if patients who had acute rejection had therapeutic concentrations of MPA. Any future study should include measurement of drug concentration.
One of the potential benefits of MMF monotherapy is the lack of renal toxicity. Both cyclosporine and tacrolimus are associated with acute and chronic nephrotoxicity. These drugs cause acute toxicity because of a reversible decrease in renal blood flow due to vasoconstriction of the afferent renal artery.28 With continued exposure, structural changes that are often progressive and irreversible occur in the kidneys.29, 30 In our study, conversion of patients from CI to MMF resulted in improvement in renal function. Those patients who did not require dialysis had either stable or improved renal function, and our data are consistent with the observations made by many other investigators9–12, 31 who have shown improvement in renal function upon withdrawal of CI. Of the 4 patients who required dialysis before or after conversion to MMF, 3 were felt clinically and/or pathologically to have renal failure due to CI.
In the past, there was some interest that MMF may have had some benefit in the treatment of hepatitis C because of similarities between ribavirin and MMF.32 Similar to ribavirin, MPA, the breakdown product of MMF, also inhibits inosine monophosphate dehydrogenase. However, there is no current evidence to suggest that MMF has any direct antiviral action,33; moreover, MMF therapy was not associated with delayed posttransplant recurrence or improved response to ribavirin and interferon.34 In our study, 4 of 5 patients who had clinical recurrence developed cirrhosis resulting in death or retransplantation.
In summary, despite excellent tolerability and minimal side effects, we urge caution in the use of MMF as monotherapy in “stable” long-term survivors of liver transplantation. If monotherapy is attempted, patients should have very close follow-up on an indefinite basis. Future studies should examine the utility of drug monitoring and various drug interactions to determine whether optimal therapeutic levels could prevent acute or chronic rejection.