Telephone: +34981951377; FAX: +34981951378
Improvement of renal function after the switch from a calcineurin inhibitor to everolimus in liver transplant recipients with chronic renal dysfunction
Article first published online: 24 NOV 2009
Copyright © 2009 American Association for the Study of Liver Diseases
Volume 15, Issue 12, pages 1792–1797, December 2009
How to Cite
Castroagudín, J. F., Molina, E., Romero, R., Otero, E., Tomé, S. and Varo, E. (2009), Improvement of renal function after the switch from a calcineurin inhibitor to everolimus in liver transplant recipients with chronic renal dysfunction. Liver Transpl, 15: 1792–1797. doi: 10.1002/lt.21920
- Issue published online: 24 NOV 2009
- Article first published online: 24 NOV 2009
- Manuscript Accepted: 27 JUL 2009
- Manuscript Received: 29 APR 2009
Chronic renal dysfunction is a frequent and severe complication in solid-organ transplant recipients. Calcineurin inhibitors (CNIs) are the main pathogenic factors of renal dysfunction. Switching from CNIs to nonnephrotoxic drugs, such as mammalian target of rapamycin inhibitors (everolimus and sirolimus), can improve renal function in these patients, but available data about the efficacy and safety of everolimus in liver transplant recipients are scarce. Twenty-one liver transplant recipients (19 males, mean age = 60.6 ± 7.8 years) with chronic renal dysfunction (creatinine ≥ 1.5 mg/dL) were prospectively included. The basal creatinine values were 1.79 ± 0.39 mg/dL (range = 1.50–2.90 mg/dL). The basal creatinine clearance, evaluated with the Cockroft-Gault formula, was 54.64 ± 12.47 mL/minute. Everolimus was initiated at a dosage of 0.75 mg twice daily, with target levels of 3 to 8 ng/mL. The withdrawal of CNIs was initiated after the target levels of everolimus were reached. Periodic controls of the weight, arterial pressure, liver function tests, serum creatinine, everolimus levels, proteinuria, creatinine clearance, and glomerular filtration rate at days 30, 90, 180, and 360 were made. After a median follow-up of 19.8 months, the respective creatinine values at 30, 90, 180, and 360 days were 1.68 ± 0.40 (P = 0.012 with respect to basal values), 1.67 ± 0.34 (P = 0.107), 1.70 ± 0.41 (P = 0.521), and 1.57 ± 0.30 mg/dL (P = 0.047). The respective creatinine clearance values at 30, 90, 180, and 360 days were 58.64 ± 16.50 (P = 0.013 with respect to basal values), 59.49 ± 13.27 (P = 0.028), 59.82 ± 16.83 (P = 0.124), and 64.46 ± 16.79 mL/minute (P = 0.025). CNIs were withdrawn in 20 recipients (95.2%). Rejection was not detected in any case. In conclusion, the application in liver transplant recipients with chronic renal dysfunction of an immunosuppressive protocol with everolimus and the withdrawal of CNIs was associated with an initial improvement of renal function tests without an increase in the risk of rejection. Liver Transpl 15:1792–1797, 2009. © 2009 AASLD.
Chronic renal dysfunction is a frequent and severe complication in solid-organ transplant recipients. Half of liver transplant patients present a variable grade of renal dysfunction, and in 8% to 28%, advanced renal failure is established. The prevalence of renal dysfunction in this population increases as the posttransplantation follow-up period lengthens.1–5 A large study evaluating the data for 36,849 liver transplant recipients showed a 5-year accumulated incidence of chronic renal dysfunction [defined as a glomerular filtration rate (GFR) less than 29 mL/minute/m2] of 18%.6 The development of posttransplantation renal failure has a significant negative impact on the survival of liver transplant recipients.1, 4, 6, 7
Calcineurin inhibitor (CNI) immunosuppressive agents (cyclosporine and tacrolimus) are the main pathogenic factors of renal dysfunction. Thus, switching from CNIs to nonnephrotoxic drugs such as mammalian target of rapamycin (mTOR) inhibitors (everolimus and sirolimus) may improve renal function in these patients, but the efficacy and safety of everolimus in liver transplant recipients with chronic renal dysfunction have not been evaluated. Therefore, the aim of this study was to assess the feasibility, efficacy, and safety of conversion from CNIs to everolimus as a maintenance immunosuppressive drug in order to improve renal function in these patients.
PATIENTS AND METHODS
During a 24-month period, 21 liver transplant recipients (19 males, mean age = 60.4 ± 8 years, range = 42–72 years) with chronic renal dysfunction were prospectively included. Renal impairment was defined if basal creatinine values were greater than 1.49 mg/dL. The indications for transplantation were alcoholic cirrhosis in 14 patients (66.7%), cirrhosis secondary to hepatitis C virus (HCV) infection in 3 (14.3%), hemochromatosis in 2 (9.5%), autoimmune cirrhosis in 1, and fulminant liver failure in 1. Hepatocellular carcinoma was diagnosed in 3 patients with alcoholic cirrhosis, in 2 with HCV cirrhosis, and in 1 with hemochromatosis. Diabetes mellitus was present in 15 patients (71.4%), and in 12 (57.1%), arterial hypertension was detected.
The basal creatinine values were 1.79 ± 0.39 mg/dL (range = 1.50–2.90 mg/dL). The basal creatinine clearance, evaluated with the Cockroft-Gault formula,8 was 54.64 ± 12.47 mL/minute. The basal GFR, evaluated with the 4-variable Modification of Diet in Renal Disease (MDRD-4) formula,9 was 42.14 ± 8.71 mL/minute/m2. According to the GFR values, chronic kidney disease in this population was classified as stage 3 in 18 recipients (85.7%) and as stage 4 in the remaining 3 patients.10 Proteinuria greater than 1 g/L was detected in 4 recipients. The mean time between liver transplantation and inclusion in the study protocol was 62.4 ± 36.6 months. The mean evolution time of posttransplantation chronic renal dysfunction was 43.9 ± 37 months.
In 18 patients, a previous specific immunosuppressive protocol with CNIs at a low dosage plus mycophenolate mofetil (MMF) or mycophenolic acid (MPA) was applied, but improvement of renal function was not achieved; progressive deterioration was even observed in some patients. In the remaining 3 patients, MMF was not tolerated or contraindicated. No other immunosuppressive drugs such as corticosteroids or azathioprine were administered. The basal features of the study population are summarized in Table 1.
|Variable||n (%)/Mean ± SD|
|Gender (male/female)||19 (92.8%)/2|
|Age at conversion (years)||60.4 ± 8 (42–72)|
|Indication for transplantation|
|Pretransplantation diabetes mellitus||8 (38.1%)|
|Pretransplantation arterial hypertension||1 (4.8%)|
|Pretransplantation serum creatinine (mg/dL)||1.36 ± 0.82|
|Basal immunosuppressive protocol|
|Posttransplantation diabetes mellitus||15 (71.4%)|
|Posttransplantation arterial hypertension||12 (57.1%)|
|Time of renal dysfunction (months)||43.9 ± 37|
|Basal serum creatinine (mg/dL)||1.79 ± 0.39 (1.50–2.90)|
|Creatinine clearance (mL/minute)*||54.64 ± 12.47|
|Glomerular filtration rate (mL/minute/m2)†||42.14 ± 8.71|
|Stage 3||18 (85.7%)|
|Stage 4||3 (14.3%)|
Immunosuppressive Conversion Protocol
The study protocol was presented and approved by local ethics and transplant committees. Written informed consent was obtained from all patients before their inclusion in the study. Everolimus (Certican, Novartis Pharma AG, Basel, Switzerland) was initiated at a dosage of 0.75 mg twice daily, with target levels of 3 to 8 ng/mL. Blood levels of everolimus were measured with a microparticle enzyme immunoassay (Abbott IMx MEIA, Abbott Laboratories, Abbott Park, IL).11 The CNI was tacrolimus in 17 cases (80.9%), and cyclosporine was used in the remaining 4 patients. The CNI was withdrawn after the target levels of everolimus were reached. Concurrent immunosuppression with MMF or MPA at the dosages of 500 mg twice daily and 360 mg twice daily, respectively, was maintained in the aforementioned 18 patients.
Periodic controls of the weight, arterial pressure, blood cell count, liver function tests (serum bilirubin, albumin, and prothrombin time), liver biochemistry (serum aspartate aminotransferase, alanine aminotransferase, gamma glutamyl transferase, and alkaline phosphatases), serum lipids (cholesterol, triglycerides), serum creatinine, blood levels of everolimus, proteinuria, creatinine clearance (Cockroft-Gault), and GFR (MDRD-4) at 30, 90, 180, and 360 days were made. Improvement of renal function was considered if creatinine values less than 1.41 mg/dL were reached during the follow-up.
Basal and evolutive serum creatinine, clearance creatinine, and GFR values were compared by a paired t test. To determine the influence of epidemiological, clinical, and analytical variables on the improvement of renal function, Fisher's test was applied. A 2-tailed P value < 0.05 was considered significant. Statistical analysis was performed with the statistical package SPSS version 15 (SPSS, Inc., San Diego, CA). This study was not supported by an external institution or agency.
After a median follow-up of 19.8 months, the respective serum creatinine values at 30, 90, 180, and 360 days were 1.68 ± 0.40 (P = 0.012 with respect to the basal values), 1.67 ± 0.34 (P = 0.107), 1.70 ± 0.41 (P = 0.521), and 1.57 ± 0.30 mg/dL (P = 0.046; Fig. 1). The respective creatinine clearance values at 30, 90, 180, and 360 days were 58.64 ± 16.50 (P = 0.013 with respect to the basal values), 59.49 ± 13.27 (P = 0.028), 59.82 ± 16.83 (P = 0.124), and 64.46 ± 17.22 mL/minute (P = 0.025; Fig. 2). The respective GFR values, evaluated with the MDRD-4 formula, at 30, 90, 180, and 360 days were 45.81 ± 11.29 (P = 0.005 with respect to the basal values), 46.11 ± 10.12 (P = 0.024), 46.13 ± 11.75 (P = 0.148), and 49.79 ± 10.33 mL/minute/m2 (P = 0.016), respectively (Fig. 3). At 360 days, the GFR values increased more than 5 mL/minute/m2 in 42.8% of the patients, whereas a reduction of GFR values greater than 5 mL/minute/m2 was observed in 19% of the patients. The mean GFR change was 3.53 ± 8.99 mL/minute/m2 at the end of follow-up. Final chronic kidney disease staging was stage 2 in 4 recipients (9.5%), stage 3 in 14 recipients (66.7%), and stage 4 in the remaining 3 patients (13.3%).
In 20 recipients (95.2%), the CNI was completely withdrawn. The respective mean blood levels of tacrolimus at the moment of beginning of mTOR and the withdrawal of the CNI were 2.7 ± 1.4 and 2.2 ± 1.7 ng/mL (P = 0.078).
Creatinine values equal to or less than 1.40 mg/dL were reached in 38.1%, 33.3%, 25%, and 50% of evaluable patients at 30, 90, 180, and 360 days of follow-up, respectively. Basal creatinine < 1.7 mg/dL, pretreatment peak levels of creatinine < 2 mg/dL, basal creatinine clearance levels > 60 mL/minute, and basal GFR > 45 mL/minute/m2 were the only variables associated with early normalization of renal function (P < 0.05). No relationship between the normalization of serum creatinine and age, gender, time from transplantation, time of renal dysfunction, HCV infection, and presence of diabetes, arterial hypertension, or proteinuria was observed (Table 2).
|Variable||+30 Days||+90 Days||+180 Days||+360 Days|
|Age at conversion (years): ≤60 versus >60||0.203||0.642||0.347||0.614|
|Gender: male versus female||0.504||>0.999||>0.999||—|
|HCV infection: no versus yes||0.257||>0.999||>0.999||>0.999|
|Year of transplantation: before January 1, 2001 versus after January 1, 2001||0.631||0.613||0.613||>0.999|
|Time of renal dysfunction: <360 versus ≥360 days||0.618||>0.999||>0.999||0.550|
|Diabetes mellitus: no versus yes||0.631||0.613||0.613||0.302|
|Arterial hypertension: no versus yes||0.367||>0.999||0.603||>0.999|
|Proteinuria > 1 g/L: no versus yes||0.131||0.521||0.539||>0.999|
|Peak serum creatinine value (mg/dL): <2 versus ≥2||0.024||0.070||0.038||0.041|
|Basal serum creatinine (mg/dL): <1.7 versus ≥1.7||0.007||0.003||0.303||0.041|
|Basal creatinine clearance (mL/minute): >60 versus ≤60||0.018||0.356||0.347||0.041|
|Glomerular filtration rate (mL/minute/m2): >45 versus ≤45||0.007||0.003||0.303||0.003|
Everolimus was well tolerated, and early withdrawal of mTOR was needed in only 1 case because of leukopenia. This patient died during the follow-up because of bacteremia by Escherichia coli complicated by septic shock and cardiac failure 3 months after inclusion in the study protocol. No relationship of this adverse event with the study drug was considered. One additional recipient suffered graft dysfunction due to secondary biliary cirrhosis resulting from a preexisting noncorrectible chronic ischemia of the donor's biliary tree. Biochemistry liver tests and graft function were stable in the remaining 19 patients. Rejection was not detected in any case. A insignificant decrease in the white blood cell count was observed (6246.2 ± 2603.3 versus 5423.3 ± 1919.3 cells/mm3, P = 0.098). No differences between triglyceride values before and after conversion to everolimus were observed (144.6 ± 88.9 versus 188.8 ± 121.3 mg/dL, P = 0.121). Cholesterol values were significantly higher after conversion with respect to basal ones (192.8 ± 34.8 versus 241.8 ± 95.5 mg/dL, P = 0.024). Nevertheless, in only 6 patients (28.6%) was lipid-lowering therapy needed. De novo proteinuria was detected in 8 patients (38.1%). In 6, proteinuria peak values were inferior to 0.5 g/L (mean = 0.14 ± 0.13 g/L), with albuminuria peak levels of 57 ± 111 mg/L. In the remaining 2 patients, proteinuria > 3 g/L was observed during the follow-up, with respective peak values of albuminuria of 2550 and 2970 mg/L. With respect to the 4 recipients with preexisting proteinuria, proteinuria was reversed in 1, and it remained stable in 2 (the respective peak values of proteinuria were 1.43 and 1.58 g/L, and those of albuminuria were 455 and 988 mg/L). In only 1 patient did proteinuria progress to a proteinuria peak value of 6.42 g/L and an albuminuria value of 4170 mg/L at the end of follow-up.
Everolimus is a recently developed macrolide immunosuppressive agent that inhibits the proliferation and clonal expansion of interleukin-activated T lymphocytes by means of a blockade of mTOR. Its synergic immunosuppressive action when it is administered with cyclosporine and the absence of nephrotoxicity are the main features of everolimus.12
The efficacy and safety of everolimus have been reported for cardiac and renal transplant recipients,12–14 but clinical data for liver transplantation are scarce. In a recently published study,15 treatment with everolimus in combination with cyclosporine in de novo liver transplant recipients was associated with an insignificantly lower rejection rate in comparison with cyclosporine combined with placebo. Adverse events were more numerous in everolimus-treated patients, but the incidence of myelotoxicity and renal dysfunction was similar in all groups. This study suggests that everolimus is safe as an immunosuppressive agent in liver transplantation.
The efficacy of the mTOR agent sirolimus as a substitute for CNIs in liver transplant recipients with renal dysfunction has been evaluated in some studies.16–20 The withdrawal of CNIs was followed by an improvement in renal function between the first and third months of treatment, but this beneficial effect was not maintained in long-term follow-up. The absence of a long-term benefit, the increased risk of rejection when sirolimus was administered as monotherapy, and the high rate of adverse events could be the reasons that sirolimus-based immunosuppressive protocols have not been generally accepted. To date, little information about the role of everolimus as a rescue treatment in liver transplant recipients with chronic renal dysfunction is available. De Simone et al.21 reported the results of a trial with conversion from CNIs to everolimus in 40 liver transplant recipients. In 36 of them, chronic renal dysfunction was the indication for conversion. Clearance creatinine values increased 4.03 ± 12.6 mL/minute at 12 months. Nevertheless, improvement of renal function was not a primary endpoint of the study, and failure of conversion occurred in 25% of cases because of rejection, hypertransaminasemia, cholangitis, or adverse events.
In this study, the application in liver transplant recipients with long-evolution chronic renal dysfunction of an immunosuppressive protocol with everolimus (associated with MMF or MPA or not) and the later withdrawal of CNIs was associated with improvement of renal function tests. Normalization of creatinine, defined as values equal to or less than 1.40 mg/dL, was achieved in 30% to 50% of patients during the follow-up. Nevertheless, the improvement of renal function was not constant, and impairment after initial improvement was seen in some patients. Initial improvement of renal function tests could be explained by a reduction of the dosage and late withdrawal of CNIs, although differences between the CNI levels before conversion and at the moment of withdrawal were not significant. Finally, the absence of improvement or even late deterioration of renal function could be explained by concurrent etiological factors of renal failure, such as diabetic or hypertensive nephropathies.22 Longer follow-up periods will be necessary in order to assess the evolution of renal function parameters after the first year of everolimus introduction.
Basal renal function (defined as basal and peak values of creatinine), basal creatinine clearance, and GFR were the only pretreatment factors associated with the failure of renal function to improve. Likewise, baseline creatinine clearance was the only variable correlated with the probability of improvement of renal function in the study by De Simone et al.21 This fact may indicate a higher grade of reversibility in patients with less severe renal dysfunction and emphasizes the need for early treatment with everolimus in order to achieve the best results. In this series, the patients presented long-evolution renal failure with a mean evolution time greater than 40 months. In most of the recipients, previous alternative immunosuppressive protocols with a reduction of CNI levels were applied after the development of renal failure,23 without the achievement of improvement of renal function or even progressive impairment. Nevertheless, the parameters of renal function were relatively conserved. This fact may indicate the existence of a renal functional reserve that permits the improvement of renal parameters, and it emphasizes the urgency of active follow-up of posttransplantation renal function and early and vigorous action to modify the immunosuppressive protocols when renal dysfunction is detected. Undoubtedly, switching from CNIs to mTOR agents during posttransplantation terminal renal failure will not achieve any benefit with respect to the renal function or prognosis of these patients.
Treatment with everolimus was well tolerated, and in only 1 case was everolimus withdrawn because of myelotoxicity. Liver biochemistry tests remained stable during the follow-up, and no episodes of rejection were detected, even in patients with everolimus as the only immunosuppressive agent. No immunological causes of graft lost were seen. Thus, the rate of success after conversion from CNIs to everolimus in this series was greater than 95%, which is greater than the rate reported by De Simone et al.21 No sudden withdrawal of CNIs and concurrent immunosuppressant therapy with MMF or MPA in some patients of our study may have played a role in preventing the development of rejection in this population.
In conclusion, everolimus is safe and useful as a rescue immunosuppressive agent in chronic renal dysfunction after liver transplantation. Early improvements of renal function may be achieved without an increase in the risk of rejection. In contrast with previously reported results,21 in our study, improvement of renal function was the primary endpoint, the rate of successful conversion was very high, and the prevalence of adverse events, including immunological events, was low. More studies are needed in order to assess the long-term effects of everolimus-based regimens on renal function in these recipients.
- 10National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification and stratification. Am J Kidney Dis 2002; 39(suppl 1): S1–S266.