- Email: [email protected]
Dyslipidemia during sirolimus therapy in liver transplant recipients occurs with concomitant cyclosporine but not tacrolimus
RAPID COMMUNICATION
Dyslipidemia During Sirolimus Therapy in Liver Transplant Recipients Occurs With Concomitant Cyclosporine But Not Tacrolimus James F. Trotter,* Michael E. Wachs,† Thomas E. Trouillot,* Thomas Bak,† Marcelo Kugelmas,* Igal Kam,† and Gregory Everson* Since its approval as an immunosuppressive agent in renal transplantation, sirolimus (RAPA) recently has been used in the primary immunosuppression regimen at several liver transplant centers. One of the major side effects of RAPA is hypercholesterolemia, which is reported in up to 44% of patients. We describe our experience in 57 primary liver transplant recipients treated with RAPA and either cyclosporine A (CSA) or tacrolimus (TAC). We report the incidence and severity of hypercholesterolemia using a prednisone-free immunosuppressive regimen. Between January 2000 and September 2000, a total of 57 patients underwent transplantation at the University of Colorado Health Sciences Center (Denver, CO) with RAPA and either CSA or TAC. The initial 10 patients who underwent transplantation under this protocol were not administered corticosteroids, and the subsequent 47 patients were administered only 3 doses of methylprednisolone days 0, 1, and 2 postoperatively (1, 0.5, and 0.5 g, respectively). Total fasting cholesterol, high-density cholesterol, low-density cholesterol, and triglyceride levels were measured at monthly intervals. Mean serum cholesterol level was significantly greater in CSA patients (200 mg/dL) compared with TAC patients (158 mg/dL; P ⴝ .0003). Serum triglyceride levels were more than 2-fold greater with CSA (292 mg/dL) compared with TAC (134 mg/dL; P ⴝ .002). Hypercholesterolemia (cholesterol > 240 mg/dL) was present in 10 of 57 patients (18%) and was significantly more common in CSA-treated patients (8 of 27 patients; 30%) compared with TAC-treated patients (2 of 30 patients; 6%; P < .05). Hypertriglyceridemia (serum triglyceride > 300 mg/dL) was present in 10 of 57 patients (18%) and was significantly more common in CSA-treated patients (9 of 27 patients; 33%) compared with TAC-treated patients (1 of 30 patients; 3%; P < .05). We conclude that (1) concomitant use of TAC with RAPA reduces the prevalence and severity of posttransplantation dyslipidemia, and (2) these findings have important implications in the prevention of complications of hypercholesterolemia in liver transplant recipients. (Liver Transpl 2001;7:401-408.)
S
irolimus (RAPA) is a macrocyclic lactone produced by Streptomyces hygroscopicus that has potent antifungal effects as well as antiproliferative and immunosuppressive effects.1-3 The mechanism of RAPA’s immunosuppressive effect is mediated through a unique intracellular protein, mTOR, or target of rapamycin (sirolimus). The RAPA-mTOR complex blocks signal
transduction in T lymphocytes and inhibits cell cycle progression from G1 to S phase. Three recent studies, largely of renal transplantation, have been published regarding the efficacy of RAPA as an immunosuppressant in transplant recipients.4-6 These studies showed that compared with conventional immunosuppressive regimens, RAPA is an effective immunosuppressive agent and reduces the occurrence and severity of acute rejection. The most common side effects noted were leukopenia, anemia, thrombocytopenia, hyperlipidemia, and mouth ulcers. Based on the positive reports of RAPA in renal transplantation, we began to use RAPA in our immunosuppressive regimen in all our primary liver transplant recipients from January 2000 to the present. Before this change in our immunosuppressive protocol, all primary liver transplant recipients at our center were administered corticosteroids for only the first 14 days after transplantation.7 The goal of adding RAPA to our regimen was to determine whether we could successfully perform transplantation on patients without corticosteroid use. Toward that end, we initially placed our first 10 transplant recipients on RAPA and either cyclosporine A (CSA) or tacrolimus (TAC) therapy without corticosteroids. We noted a high incidence of postoperative fevers, malaise, and lethargy in these patients. Consequently, we added 3 doses of methylprednisolone (1 g, 500 mg, and 500 mg) postoperative days 0, 1, and 2, respectively. In this report, we describe the lipid profiles of patients who underwent transplantation under this imFrom the Divisions of *Gastroenterology/Hepatology and †Transplant Surgery, University of Colorado Health Sciences Center, Denver, CO. Address reprint requests to James F. Trotter, MD, Division of Gastroenterology/Hepatology, 4200 E 9th Ave, B-154, Denver, CO 80262. Telephone: 303-372-8866; FAX: 303-372-8868; E-mail: James. [email protected] Copyright © 2001 by the American Association for the Study of Liver Diseases 1527-6465/01/0705-0013$35.00/0 doi:10.1053/jlts.2001.23916
Liver Transplantation, Vol 7, No 5 (May), 2001: pp 401-408
401
402
Trotter et al
munosuppressive regimen. The clinical outcomes of these patients have been reported elsewhere.8 The patient survival rate was 92%, and graft survival rate was 89%, similar to historical controls. However, the incidence of rejection was significantly less in the RAPAtreated patients (30%) compared with historical controls (70%; P ⬍ .05). In addition, the incidence of steroid-resistant rejection was also significantly less in the RAPA group (3%) compared with 37% in controls (P ⬍ .05).
Methods This is a retrospective analysis of 57 liver transplant recipients administered RAPA between January 2000 and September 2000 at the University of Colorado Health Sciences Center (Denver, CO). The disposition of all 57 patients is shown in Figure 1. Forty-eight patients received a cadaveric graft and 9 patients received a right hepatic lobe graft from a living donor. As the primary immunosuppressive regimen, patients were administered RAPA and either CSA or TAC. (The assignment to TAC or CSA therapy was performed on an alternating basis, i.e., odd-numbered transplant recipients were administered TAC and even-numbered transplant recipients were administered CSA.) The initial 10 patients who underwent transplantation under this protocol were not administered corticosteroids. However, because of a high incidence of immediate postoperative fevers, malaise, and lethargy, we added 3 doses of methylprednisolone (day 0, 1 g; days 1 and 2, 0.5 g) to alleviate these symptoms. The subsequent 47 patients were administered this initial 3-day course of methylprednisolone.
The following criteria were exclusions from analysis: retransplantation, OKT3 induction, known intolerance to RAPA, and discontinuation of RAPA within 2 weeks of transplantation. RAPA was administered orally or through a nasogastric tube at a dose of 6 mg postoperative day 0, then 2 mg/d orally thereafter. RAPA was administered at noon, whereas CSA and TAC were administered at 8:00 AM and 8:00 PM.9 CSA was administered orally at an initial dose of 5 mg/kg/d in divided doses, and TAC was administered orally or through a nasogastric tube at 0.1 mg/kg/d in divided doses. Target levels for CSA were 200 to 250 ng/mL for month 1, 175 to 200 ng/mL for month 2, and 150 ng/mL thereafter. Target levels for TAC were 8 to 10 ng/mL for month 1, 6 to 8 ng/mL month 2, and 5 to 7 ng/mL thereafter. Blood levels of RAPA were obtained sporadically, but no target level was prescribed. Fasting serum cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), and triglyceride levels were measured monthly for up to 6 months after transplantation. All available laboratory results were included in the study analysis, but not all patients had levels drawn at each interval. Hypercholesterolemia was defined as serum cholesterol level greater than 240 mg/dL, and hypertriglyceridemia was defined as serum triglyceride level greater than 300 mg/dL. Hypertension was defined as a diastolic blood pressure greater than 90 mm Hg on 3 separate occasions. Diabetes was defined as a blood glucose level greater than 200 mg/dL or the need for hypoglycemic agent or insulin. Change in weight was measured from the patient’s first outpatient visit after discharge from the hospital immediately after transplantation until the last available follow-up visit. Measures of statistical significance between immunosuppressive agents and time intervals were performed using Student’s t-test with the Microsoft Excel Spreadsheet (Microsoft Corp, Redmond, WA). Measures of statistical significance of the occurrence of hypercholesterolemia and hypertriglyceridemia comparing TAC versus CSA patients were performed using Chi-squared analysis.
Results
Figure 1. Patient disposition showing the outcome of each patient.
Demographics of the 57 patients are listed in Table 1. Mean and median follow-up after transplantation were 131 and 143 days, respectively. The disposition of all 57 patients is shown in Figure 1. Of the original 57 patients, 6 patients were excluded from analysis because they were administered RAPA for less than 2 weeks for the following reasons: respiratory failure with fever (4 patients), wound dehiscence (1 patient), and mouth ulcers (1 patient). Three of these 6 patients died less than 1 month after transplantation. RAPA therapy was discontinued in an additional 8 patients after a mean time of 54 ⫾ 26 (SD) days for the following reasons: mouth ulcers (6 patients), wound dehiscence (1 patient), and leukopenia (1 patient). Patients were censored from analysis after discontinuation of RAPA.
403
Dyslipidemia During Sirolimus Therapy
Table 1. Patient Demographics
Total no. of patients Sex distribution (M/F) Mean age (yr) Cause of liver disease Hepatitis C Hepatitic C/alcohol Sclerosing cholangitis Alcohol Other Racial distribution White Hispanic Asian Black
TAC
CSA
P
30 22/8 44.9 ⫾ 11.0
27 18/9 47.3 ⫾ 9.5
NS NS
9 (30) 10 (33) 7 (23) 2 (7) 2 (7)
10 (37) 2 (7) 5 (19) 4 (15) 6 (22)
NS .05 NS NS NS
27 (90) 2 (7) 1 (3) 0 (0)
20 (74) 4 (15) 2 (7) 1 (4)
NS NS NS NS
NOTE. Values expressed as mean ⫾ SD or number (percent).
Nine patients (19%) had to be administered intermittent or chronic prednisone after transplantation for the following indications: autoimmune hepatitis, 4 patients; autoimmune hemolytic anemia, 1 patient; gout, 1 patient; ulcerative colitis, 1 patient; and other, 2 patients. Patient and graft survival were 54 of 57 patients (95%) and 53 of 57 grafts (93%). One patient was started on CSA therapy and then developed new-onset
seizures with a speech disorder. As a result, he was converted to TAC therapy within 4 weeks of transplantation and maintained on TAC. This patient was analyzed in the TAC group. Interestingly, he was the only patient administered TAC to develop hypertriglyceridemia and was 1 of 2 patients who had hypercholesterolemia. Hyperlipidemia was present over 5 months after conversion to TAC therapy. As listed in Table 2, the mean serum cholesterol level for TAC patients was 158 mg/dL compared with 200 mg/dL for CSA patients (P ⫽ .0003). There was a trend toward a greater HDL cholesterol level in TAC-treated patients compared with CSA-treated patients, which did not reach statistical significance. Mean LDL cholesterol level was 18% greater in CSA patients (113 mg/dL) compared with TAC patients (93 mg/dL; P ⫽ .05). Mean serum triglyceride level was 292 mg/dL for CSA patients compared with 134 mg/dL in TAC patients (P ⫽ .002). The ratio of LDL-HDL was 12% greater in CSA patients (P ⫽ not significant [NS]). Hypercholesterolemia (total serum cholesterol ⬎ 240 mg/dL) was present in 10 of 57 patients (18%). The occurrence of hypercholesterolemia in patients on TAC therapy was significantly less (2 of 30 patients; 6%) than in the CSA group (8 of 27 patients; 30%; P ⬍ .05). Hypertriglyceridemia (total serum triglyceride ⬎ 300 mg/dL) occurred in 10 of 57 patients (18%), and the occurrence of hypertriglyceridemia was significantly greater in CSA-
Table 2. Occurrence of Hyperlipidemia and Risk Factors
Cholesterol (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Triglycerides (mg/dL) LDL/HDL Hypercholesterolemia Hypertriglyceridemia Creatinine (mg/dL) RAPA level (ng/mL) RAPA dose (mg/d) Hypertension Diabetes mellitus Average weight gain (kg)
TAC
CSA
Difference (%)
P
158 43 93 134 2.50 2/30 (6%) 1/30 (3%) 1.10 5.13 1.98 7 (23%) 5 (17%) 0.9
200 39 113 292 2.84 8/27 (30%) 9/27 (33%) 1.21 6.07 1.94 9 (33%) 3 (10%) 1.5
21 9 18 54 12 80 91 10 18 2 30 41 40
.0003 NS .05 .002 NS ⬍.05 ⬍.05 .195 .09 NS NS NS NS
NOTE. Lipid levels (cholesterol, HDL, LDL, and triglycerides) represent the mean of all measured levels between months 1 and 6 for patients administered either TAC or CSA. Hypercholesterolemia is defined as serum cholesterol level greater than 240 mg/dL. Hypertriglyceridemia is defined as serum triglyceride level greater than 300 mg/dL. LDL/HDL represents the mean value for all LDL/HDL values measured between months 1 and 6 for all patients administered either TAC or CSA. Serum creatinine, RAPA blood level, and daily RAPA dose represent the mean values for all available data points between months 1 and 6. The occurrence of hypertension, diabetes, and weight gain are shown. Measures of significance for cholesterol, HDL, LDL, triglyceride, LDL/HDL, serum creatinine, RAPA level, RAPA dose, and weight gain were performed using Student’s t-test. Measures of significance for the occurrence of hypercholesterolemia, hypertriglyceridemia, diabetes mellitus, and hypertension were performed using Chi-squared test.
404
Trotter et al
treated patients (9 of 27 patients; 33%) compared with TAC-treated patients (1 of 30 patients; 3%; P ⬍ .05). (Mean serum cholesterol level in the 6 CSA-treated patients administered prednisone was 219 mg/dL. Mean serum cholesterol level in the 3 TAC-treated patients administered prednisone was 177 mg/dL. These values were slightly greater, but not significantly different from those in patients not administered prednisone. Measures of statistical significance were unchanged when these patients were removed from overall analysis. Data not shown.) The mean dose of RAPA administered to patients on CSA therapy (1.98 mg/d) was the same as in TAC patients (1.94 mg/d). However, the mean RAPA blood level in CSA patients (6.07 ng/mL) was greater than in TAC patients (5.13 ng/dL; P ⫽ .08). Mean serum creatinine values were also slightly greater in CSA patients (1.21 mg/dL) than TAC patients (1.10 mg/dL; P ⫽ NS). Hypertension was present in 16 of 57 patients (28%) and was slightly more common in CSA patients (9 of 27 patients; 33%) than TAC patients (7 of 30 patients; 23%; P ⫽ NS). Diabetes mellitus was diagnosed in 8 of 57 transplant recipients (14%); 5 of 30 patients (17 %) were administered TAC and 3 of 27
patients (11%) were administered CSA. Excessive weight gain was uncommon in our patients. The average change in weight after transplantation was 1.5 kg with CSA and 0.9 kg with TAC. Only 2 patients gained more than 10 kg. Comparisons of lipid values were made between months 1 and 6, listed in Table 3 and shown in Figures 2 through 5. In CSA-treated patients, serum cholesterol level increased from 182 to 213 mg/dL at month 6 (P ⫽ .15; Fig. 2). The greatest increase in lipid fraction in CSA-treated patients was triglycerides, which increased from 226 mg/dL at month 1 to 351 mg/dL at month 6 (P ⫽ NS; Fig. 3). As shown in Figure 4, HDL cholesterol level increased from 33 mg/dL at month 1 to 42 mg/dL at month 6 (P ⫽ NS) for CSAtreated patients. LDL cholesterol level increased from 99 mg/dL at month 1 to 122 mg/dL at month 6 (P ⫽ NS; Fig. 5). In TAC patients, total cholesterol level changed negligibly in patients from month 1 (162 mg/dL) to month 6 (161 mg/dL), shown in Figure 2. However, at each monthly interval, serum cholesterol level was lower in TAC patients compared with CSA patients. Statistical significance was reached at months 2 and 3. Triglycer-
Table 3. Serum Lipid and Immunosuppressive Levels in CSA and TAC Patients Month
CSA patients Cholesterol (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Triglycerides (mg/dL) CSA (ng/mL) RAPA (ng/mL) N (cholesterol) N (lipid profile) TAC patients Cholesterol (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Triglycerides (mg/dL) TAC (ng/mL) RAPA (ng/mL) N (cholesterol) N (lipid profile)
1
2
3
4
5
6
P
182 33 99 226 254 5.9 24 7
207 39 113 244 214 5.5 24 9
212 40 96 308 200 5.8 16 11
195 38 103 237 154 8.9 14 7
207 37 125 272 173 5.8 10 5
213 42 122 351 182* 6.8 9 6
NS NS NS NS .001 NS —
162 35 85 123 7.8 4.2 28 11
154 38 90 104 6.9 5.9 26 13
163 44 93 106 6.0 5.2 17 8
167 48 111 143 8.2 6.3 17 8
159 52 87 151 5.5 4.3 9 3
161 47† 85 168 7.3 4.7 11 6
NS .01 NS NS NS NS — —
NOTE. Serum lipid and immunosuppressant levels represent the mean values measured at each time interval. N (cholesterol) represents the number of patients who had a serum cholesterol level measured at the given time interval. N (lipid profile) represents the number of patients who had HDL, LDL, and triglyceride levels measured at the given time interval. Levels of significance were measured between months 1 and 6 using Student’s t-test. *Significant difference in CSA between months 1 and 6. †Significant difference in HDL cholesterol between months 1 and 6.
Dyslipidemia During Sirolimus Therapy
405
Figure 2. Serum cholesterol levels in CSA (white bars) and TAC (black bars) patients. Each data point represents the mean serum cholesterol level for patients (CSA or TAC) at each time interval. For CSA and TAC patients, there was no significant difference in cholesterol level at any monthly interval. For the comparison of TAC versus CSA at each interval, P ⴝ .065 at month 1, P ⴝ .0001 at month 2, P ⴝ .003 at month 3, P ⴝ .11 at month 4, P ⴝ .09 at month 5, and P ⴝ .08 at month 6. *P < .05 for comparisons of TAC versus CSA.
Figure 4. Serum HDL cholesterol levels for CSA (white bars) and TAC (black bars) patients. Each data point represents the mean HDL cholesterol level for patients (CSA or TAC) at each time interval. For CSA patients, there was no significant difference in HDL level at any monthly interval. For TAC patients, there was a significant (*P < .05) difference between HDL levels at months 1 and 6. For comparisons between CSA and TAC patients, the differences were significant (#P < .05) at months 4 and 5.
ide levels, shown in Figure 3, increased only slightly from 123 mg/dL at month 1 to 166 mg/dL at month 6 (P ⫽ NS). Serum triglyceride levels were lower at each monthly interval in TAC compared with CSA patients and reached statistical significance at months 1, 2, and 3. HDL cholesterol level increased from 35 mg/dL at month 1 to 47 mg/dL at month 6 (P ⫽ .05; Fig. 4). At 5 of the 6 monthly intervals (all but month 2), HDL level was greater in TAC compared with CSA patients, reaching statistical significance at months 4 and 5. In TAC patients, LDL cholesterol level was unchanged at 85 mg/dL at months 1 and 6 (Fig. 5).
Discussion The effect of RAPA on lipid levels has been evaluated in 2 clinical trials in renal transplant recipients. In the report by Groth et al,4 renal transplant recipients were randomized to RAPA or CSA therapy. All patients were administered corticosteroids and azathioprine (AZA). RAPA was administered at relatively high doses. Initially, RAPA was administered at a loading dose of 16 to 24 mg/m2/d (9.2 to 13.9 mg/d in an average adult) until days 7 through 10, at which time the dose was reduced to 8 to 12 mg/m2/d (4.6 to 6.9 mg/d) to achieve a blood level of 30 ng/mL for 2 months, then
Figure 3. Serum triglyceride levels for CSA (white bars) and TAC (black bars) patients. Each data point represents the mean serum triglyceride levels for patients (CSA or TAC) at each time interval. For CSA and TAC patients, there was no significant difference in triglyceride levels at any monthly interval. For the comparison of TAC versus CSA at each interval, P < .05 at months 1, 2, and 3. *P < .05 for comparisons of TAC versus CSA.
Figure 5. Serum LDL cholesterol levels for CSA (white bars) and TAC (black bars) patients. Each data point represents the mean LDL cholesterol level for patients (CSA or TAC) at each time interval. For CSA and TAC patients, there was no significant difference in LDL at any monthly intervals. The comparison of CSA versus TAC shows mean LDL level was greater at months 1, 2, 3, 5, and 6 for CSA, although the differences were not significant.
406
Trotter et al
dose-reduced to 15 ng/mL. Corticosteroid therapy (prednisone or methylprednisolone) was weaned to 10 mg by month 6. There was no difference in graft survival, patient survival, incidence and severity of rejection, or OKT3 use in either group. However, hypertriglyceridemia was significantly more common in the RAPA group (51%) versus the CSA group (12%; P ⬍ .01). Hypercholesterolemia occurred in 44% of RAPA patients compared with only 14% of CSA patients (P ⬍ .01). Mean serum cholesterol and triglyceride levels were greatest at 2 months posttransplantation, when the differences between the CSA- and RAPAtreated patients were the greatest. At this follow-up point, the mean triglyceride level was 5.3 versus 2.1 mmol/L (P ⬍ .001), and total serum cholesterol level was 9.2 versus 6.4 mmol/L (P ⬍ .001). In another study, 719 primary renal transplant recipients were randomly assigned to administration of RAPA, 2 and 5 mg, or AZA, 2 to 3 mg/kg, in addition to a CSA-prednisone regimen.5 Initially, transplant recipients were administered 500 mg of methylprednisolone, followed by a tapering dose of prednisone to 10 mg/d by month 6 and 5 to 10 mg/d thereafter. The incidence of acute rejection during the first year was significantly less in both RAPA groups: 22% for 2 mg and 15% for 5 mg versus 31% for AZA. In addition, there were significantly fewer episodes of moderate and severe rejection in both RAPA groups compared with AZA patients. Graft and patient survival rates at 12 months were identical in each of the groups. At 6 months, serum cholesterol levels were significantly greater by 15% with 2 mg of RAPA (6.66 mmol/L) and 24% with 5 mg of RAPA (7.20 mmol/L) compared with AZA (5.80 mmol/L). However, at 12 months, there was no significant difference in cholesterol levels between the 2-mg RAPA group (6.39 mmol/L), 5-mg RAPA group (6.53 mmol/L), and AZA group (5.86 mmol/L). The greater cholesterol and triglyceride levels may be partially explained by the greater doses of corticosteroids administered during the first 6 months. Triglyceride levels were significantly greater in the 2-mg and 5-mg RAPA groups at 6 months compared with the AZA group. However, there was no difference in triglyceride levels in any of the groups at 12 months. The prevalence of hypercholesterolemia at 6 months was 30% and 35% in patients administered 2 and 5 mg of RAPA, respectively. At 12 months, the prevalence of hypercholesterolemia was 33% and 37%, respectively. There are important issues raised by our findings. Our results suggest that RAPA may be administered to liver transplant recipients with a relatively low occur-
rence of hypercholesterolemia. The prevalence of hypercholesterolemia in previously reported studies with RAPA ranged from 30% to 44%. However, in our patients, hypercholesterolemia was uncommon. There are several possible explanations for the decreased incidence of hypercholesterolemia. First, the exclusion of prednisone, which causes hyperlipidemia, from our immunosuppression protocol may have had an important role. Despite the absence of prednisone from our regimen, the occurrence and severity of acute cellular rejection were less than in historical controls.8 Data from the renal transplant studies described previously suggest that corticosteroids potentiate hypercholesterolemia in patients administered RAPA.4,5 In both studies reported previously, corticosteroids were administered at a tapering dose down to 5 to 10 mg at 6 months. Kahan5 showed that cholesterol levels were the lowest at 12 months when corticosteroid doses were tapered to the lowest level. Groth et al4 reported significantly greater cholesterol levels in RAPA-treated patients at month 2, when corticosteroid doses were greater, compared with month 6, when corticosteroid doses were lower. Second, the low incidence of hypercholesterolemia in our patients may be caused by the relatively low doses of RAPA (2 mg/d). As noted in the studies cited previously, the incidence of hypercholesterolemia is associated with increased doses and blood levels of RAPA. Finally, the choice of calcineurin inhibitor administered with RAPA may influence the incidence of hypercholesterolemia. Although the studies by Groth et al4 and Kahan5 both used CSA, our data suggest that the use of TAC was associated with a significantly lower total serum cholesterol level compared with CSA. The occurrence of hypercholesterolemia was reduced by 80% in our patients administered TAC (6%) compared with CSA (30%). Furthermore, only 1 of our patients administered TAC had hypertriglyceridemia compared with 1 of 3 patients administered CSA. The findings of this report could have important implications for the long-term postoperative immunosuppression management of liver transplant recipients. With the technical refinements in liver transplantation over the past 20 years, current survival rates exceed 80% for most recipients. Consequently, the number of liver transplant recipients with measurable long-term survival increases each year. In addition, older patients are undergoing transplantation with greater frequency. As the cohort of older liver transplant recipients increases, greater attention is focused on the prevention of longterm postoperative complications. One of the most important of these is cardiovascular
Dyslipidemia During Sirolimus Therapy
disease. Liver transplant recipients often have concomitant risk factors for atherosclerosis, including hypertension, diabetes, and obesity. A report from our center showed that 31% of long-term liver transplant recipients have hypercholesterolemia and 40% are obese.10 Sheiner et al11 reported that compared with the general population, long-term (survival ⬎ 5 years) liver transplant recipients have a significantly greater risk for hypertension (3.07 times the general population) and diabetes (5.99 times the general population). Therefore, immunosuppressive regimens such as the one presented in this report could potentially reduce the incidence of hypercholesterolemia. This could in turn prevent the development of atherosclerosis in long-term survivors of liver transplantation. Our practice has been to follow the American Heart Association guidelines for the treatment of hyperlipidemia.12 However, we typically wait at least 6 months after transplantation before initiating cholesterolreducing medication. The follow-up period in our patients was relatively short. Consequently, we treated only one patient for hyperlipidemia; gemfibrozil was successfully used to treat one patient with a serum triglyceride level greater than 800 mg/dL. Previous studies have documented more favorable lipid profiles associated with TAC compared with CSA. Guckelberger et al13 reported lipid profiles in 302 patients with a median follow-up of 18 months. Hypercholesterolemia was significantly less common in TAC (53%) than CSA patients (76%; P ⬍ .05). Hypertriglyceridemia was present in 40% of patients administered CSA and 33% administered TAC (P ⫽ NS). Canzanello et al14 published similar results in 63 patients. These investigators showed that hypercholesterolemia was present in 33% of patients administered CSA at 12 months after liver transplantation compared with 0% of patients administered TAC. Serum cholesterol levels were significantly greater with CSA (225 mg/dL) than with TAC (159 mg/dL; P ⬍ .05). Corticosteroids were administered in both these studies and may have contributed to the development of hyperlipidemia. There is evidence that RAPA may interact with CSA to increase patient exposure to CSA. This may increase the toxic effects of CSA when both drugs are administered concomitantly. Kahan5 reported significantly greater serum creatinine levels in renal transplant patients administered RAPA compared with AZA. He further noted that lower oral doses of CSA were required to achieve the same blood CSA levels and area under the curve values in RAPA-treated patients compared with AZA-treated patients. These effects oc-
407
curred despite the delayed administration of RAPA 4 hours after CSA. The exact mechanism of the interaction of CSA and RAPA has not been fully delineated. However, possible mechanisms include competitive metabolism of both drugs through the cytochrome P-450 system or altered distribution or clearance of each drug.5 The additive renal toxic effect of CSA seen with concomitant administration of RAPA may be caused in part by increased distribution of CSA to renal tissue. Napoli et al15 reported increased concentrations of CSA in renal tissue in rats administered RAPA with CSA. We also noted greater serum creatinine levels in our patients coadministered RAPA and CSA compared with TAC. In addition, CSA patients had RAPA blood levels 18% greater than TAC patients. This suggests the possibility that RAPA exposure may be greater with the concomitant administration of CSA. This observation awaits confirmation with further studies. Our report has several flaws. Data collection was retrospective. As such, we had difficulty obtaining LDL, HDL, and triglyceride levels at each follow-up interval. This was especially true in patients who had laboratory tests drawn at sites far from our center. Consequently, not all patients had these values recorded at each follow-up interval. The number of patients with serum cholesterol and lipid profiles drawn at each follow-up interval is listed in Table 3. Finally, although our primary immunosuppressive protocol was steroid free, some patients were administered prednisone for clinical indications. These patients were included in the analysis and as noted, their removal did not change our findings. In summary, we report the lipid profiles in primary liver transplant recipients administered RAPA and either TAC or CSA with only 3 days of corticosteroid therapy. The prevalence and severity of hyperlipidemia was extremely low in patients treated concomitantly with RAPA and TAC. We speculate that prevention of hyperlipidemia could prevent atherosclerosis and its complications in long-term survivors of liver transplantation. Currently, we are formally evaluating the efficacy and side-effect profile of RAPA in liver transplant recipients in a trial underway at our institution.
References 1. Watson CJ, Friend PJ, Jamieson NV, Frick TW, Alexander G, Gimson AE, Calne R. Sirolimus: A potent new immunosuppressant for liver transplantation. Transplantation 1999;67:505-509. 2. Abraham RT, Wiederrecht GJ. Immunopharmacology of rapamycin. Annu Rev Immunol 1996;14:483-510. 3. Sehgal SN. Rapamune®: Mechanism of action of immunosup-
408
4.
5.
6.
7.
8.
9. 10.
Trotter et al
pressive effect results from blockade of signal transduction and inhibition of cell cycle progression. Clin Biochem 1998;31:335340. Groth CG, Backman L, Morales JM, Calne R, Kreis H, Lang P, et al. Sirolimus (rapamycin)-based therapy in human renal transplantation. Transplantation 1999;67:1036-1042. Kahan BD for the Rapamune® US Study Group. Efficacy of sirolimus compared with azathioprine for reduction of acute renal allograft rejection. Lancet 2000;356:194-202. McAlister VC, Gao Z, Peltekian K, Domingues J, Mahalati K, MacDonald AS. Sirolimus-tacrolimus combination immunosuppression. Lancet 2000;355:376-377. Stegall MD, Wachs ME, Everson G, Steinberg T, Bilir B, Shrestha R, et al. Prednisone withdrawal 14 days after liver transplantation with mycophenolate: A prospective trial of cyclosporine and tacrolimus. Transplantation 1997;64:1755-1760. Trotter JF, Wachs M, Bak T, Trouillot T, Stolpman N, Everson GT, Kam I. Liver transplantation using sirolimus and minimal corticosteroids (3-day taper). Liver Transpl 2001;7:343-351. Wyeth-Ayerst Laboratories. Production information, Rapamune®. Philadelphia, PA: Wyeth-Ayerst Laboratories, 1999. Stegall MD, Everson G, Schroter G, Bilir B, Karrer F, Kam I.
11.
12. 13.
14.
15.
Metabolic complications after liver transplantation. Diabetes, hypercholesterolemia, hypertension, and obesity. Transplantation 1995;60:1057-1060. Sheiner PA, Magliocca JF, Bodian CA, Kim-Schluger L, Altaca G, Guarrera JV, et al. Long-term medical complications in patients surviving ⬎ or ⫽ 5 years after liver transplant. Transplantation 2000;69:781-789. Consensus Panel Statement. Guide to primary prevention of cardiovascular diseases: Summary. Circulation 1995;92:2-4. Guckelberger O, Bechstein WO, Neuhaus R, Luesebrink R, Lemmens HP, Kratschmer B, et al. Cardiovascular risk factors in long-term follow-up after orthotopic liver transplantation. Clin Transplant 1997;11:60-65. Canzanello VJ, Schwartz L, Taler SJ, Textor SC, Wiesner RH, Porayko MK, Krom RAF. Evolution of cardiovascular risk after liver transplantation: A comparison of cyclosporine A and tacrolimus. Liver Transpl Surg 1997;3:1-9. Napoli KL, Wang ME, Stepkowski SM, Kahan BD. Relative tissue distributions of cyclosporine and sirolimus after concomitant peroral administration to the rat: Evidence for pharmacokinetic interactions. Ther Drug Monit 1998;20:123-133.
Dyslipidemia During Sirolimus Therapy in Liver Transplant Recipients Occurs With Concomitant Cyclosporine But Not Tacrolimus James F. Trotter,* Michael E. Wachs,† Thomas E. Trouillot,* Thomas Bak,† Marcelo Kugelmas,* Igal Kam,† and Gregory Everson* Since its approval as an immunosuppressive agent in renal transplantation, sirolimus (RAPA) recently has been used in the primary immunosuppression regimen at several liver transplant centers. One of the major side effects of RAPA is hypercholesterolemia, which is reported in up to 44% of patients. We describe our experience in 57 primary liver transplant recipients treated with RAPA and either cyclosporine A (CSA) or tacrolimus (TAC). We report the incidence and severity of hypercholesterolemia using a prednisone-free immunosuppressive regimen. Between January 2000 and September 2000, a total of 57 patients underwent transplantation at the University of Colorado Health Sciences Center (Denver, CO) with RAPA and either CSA or TAC. The initial 10 patients who underwent transplantation under this protocol were not administered corticosteroids, and the subsequent 47 patients were administered only 3 doses of methylprednisolone days 0, 1, and 2 postoperatively (1, 0.5, and 0.5 g, respectively). Total fasting cholesterol, high-density cholesterol, low-density cholesterol, and triglyceride levels were measured at monthly intervals. Mean serum cholesterol level was significantly greater in CSA patients (200 mg/dL) compared with TAC patients (158 mg/dL; P ⴝ .0003). Serum triglyceride levels were more than 2-fold greater with CSA (292 mg/dL) compared with TAC (134 mg/dL; P ⴝ .002). Hypercholesterolemia (cholesterol > 240 mg/dL) was present in 10 of 57 patients (18%) and was significantly more common in CSA-treated patients (8 of 27 patients; 30%) compared with TAC-treated patients (2 of 30 patients; 6%; P < .05). Hypertriglyceridemia (serum triglyceride > 300 mg/dL) was present in 10 of 57 patients (18%) and was significantly more common in CSA-treated patients (9 of 27 patients; 33%) compared with TAC-treated patients (1 of 30 patients; 3%; P < .05). We conclude that (1) concomitant use of TAC with RAPA reduces the prevalence and severity of posttransplantation dyslipidemia, and (2) these findings have important implications in the prevention of complications of hypercholesterolemia in liver transplant recipients. (Liver Transpl 2001;7:401-408.)
S
irolimus (RAPA) is a macrocyclic lactone produced by Streptomyces hygroscopicus that has potent antifungal effects as well as antiproliferative and immunosuppressive effects.1-3 The mechanism of RAPA’s immunosuppressive effect is mediated through a unique intracellular protein, mTOR, or target of rapamycin (sirolimus). The RAPA-mTOR complex blocks signal
transduction in T lymphocytes and inhibits cell cycle progression from G1 to S phase. Three recent studies, largely of renal transplantation, have been published regarding the efficacy of RAPA as an immunosuppressant in transplant recipients.4-6 These studies showed that compared with conventional immunosuppressive regimens, RAPA is an effective immunosuppressive agent and reduces the occurrence and severity of acute rejection. The most common side effects noted were leukopenia, anemia, thrombocytopenia, hyperlipidemia, and mouth ulcers. Based on the positive reports of RAPA in renal transplantation, we began to use RAPA in our immunosuppressive regimen in all our primary liver transplant recipients from January 2000 to the present. Before this change in our immunosuppressive protocol, all primary liver transplant recipients at our center were administered corticosteroids for only the first 14 days after transplantation.7 The goal of adding RAPA to our regimen was to determine whether we could successfully perform transplantation on patients without corticosteroid use. Toward that end, we initially placed our first 10 transplant recipients on RAPA and either cyclosporine A (CSA) or tacrolimus (TAC) therapy without corticosteroids. We noted a high incidence of postoperative fevers, malaise, and lethargy in these patients. Consequently, we added 3 doses of methylprednisolone (1 g, 500 mg, and 500 mg) postoperative days 0, 1, and 2, respectively. In this report, we describe the lipid profiles of patients who underwent transplantation under this imFrom the Divisions of *Gastroenterology/Hepatology and †Transplant Surgery, University of Colorado Health Sciences Center, Denver, CO. Address reprint requests to James F. Trotter, MD, Division of Gastroenterology/Hepatology, 4200 E 9th Ave, B-154, Denver, CO 80262. Telephone: 303-372-8866; FAX: 303-372-8868; E-mail: James. [email protected] Copyright © 2001 by the American Association for the Study of Liver Diseases 1527-6465/01/0705-0013$35.00/0 doi:10.1053/jlts.2001.23916
Liver Transplantation, Vol 7, No 5 (May), 2001: pp 401-408
401
402
Trotter et al
munosuppressive regimen. The clinical outcomes of these patients have been reported elsewhere.8 The patient survival rate was 92%, and graft survival rate was 89%, similar to historical controls. However, the incidence of rejection was significantly less in the RAPAtreated patients (30%) compared with historical controls (70%; P ⬍ .05). In addition, the incidence of steroid-resistant rejection was also significantly less in the RAPA group (3%) compared with 37% in controls (P ⬍ .05).
Methods This is a retrospective analysis of 57 liver transplant recipients administered RAPA between January 2000 and September 2000 at the University of Colorado Health Sciences Center (Denver, CO). The disposition of all 57 patients is shown in Figure 1. Forty-eight patients received a cadaveric graft and 9 patients received a right hepatic lobe graft from a living donor. As the primary immunosuppressive regimen, patients were administered RAPA and either CSA or TAC. (The assignment to TAC or CSA therapy was performed on an alternating basis, i.e., odd-numbered transplant recipients were administered TAC and even-numbered transplant recipients were administered CSA.) The initial 10 patients who underwent transplantation under this protocol were not administered corticosteroids. However, because of a high incidence of immediate postoperative fevers, malaise, and lethargy, we added 3 doses of methylprednisolone (day 0, 1 g; days 1 and 2, 0.5 g) to alleviate these symptoms. The subsequent 47 patients were administered this initial 3-day course of methylprednisolone.
The following criteria were exclusions from analysis: retransplantation, OKT3 induction, known intolerance to RAPA, and discontinuation of RAPA within 2 weeks of transplantation. RAPA was administered orally or through a nasogastric tube at a dose of 6 mg postoperative day 0, then 2 mg/d orally thereafter. RAPA was administered at noon, whereas CSA and TAC were administered at 8:00 AM and 8:00 PM.9 CSA was administered orally at an initial dose of 5 mg/kg/d in divided doses, and TAC was administered orally or through a nasogastric tube at 0.1 mg/kg/d in divided doses. Target levels for CSA were 200 to 250 ng/mL for month 1, 175 to 200 ng/mL for month 2, and 150 ng/mL thereafter. Target levels for TAC were 8 to 10 ng/mL for month 1, 6 to 8 ng/mL month 2, and 5 to 7 ng/mL thereafter. Blood levels of RAPA were obtained sporadically, but no target level was prescribed. Fasting serum cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), and triglyceride levels were measured monthly for up to 6 months after transplantation. All available laboratory results were included in the study analysis, but not all patients had levels drawn at each interval. Hypercholesterolemia was defined as serum cholesterol level greater than 240 mg/dL, and hypertriglyceridemia was defined as serum triglyceride level greater than 300 mg/dL. Hypertension was defined as a diastolic blood pressure greater than 90 mm Hg on 3 separate occasions. Diabetes was defined as a blood glucose level greater than 200 mg/dL or the need for hypoglycemic agent or insulin. Change in weight was measured from the patient’s first outpatient visit after discharge from the hospital immediately after transplantation until the last available follow-up visit. Measures of statistical significance between immunosuppressive agents and time intervals were performed using Student’s t-test with the Microsoft Excel Spreadsheet (Microsoft Corp, Redmond, WA). Measures of statistical significance of the occurrence of hypercholesterolemia and hypertriglyceridemia comparing TAC versus CSA patients were performed using Chi-squared analysis.
Results
Figure 1. Patient disposition showing the outcome of each patient.
Demographics of the 57 patients are listed in Table 1. Mean and median follow-up after transplantation were 131 and 143 days, respectively. The disposition of all 57 patients is shown in Figure 1. Of the original 57 patients, 6 patients were excluded from analysis because they were administered RAPA for less than 2 weeks for the following reasons: respiratory failure with fever (4 patients), wound dehiscence (1 patient), and mouth ulcers (1 patient). Three of these 6 patients died less than 1 month after transplantation. RAPA therapy was discontinued in an additional 8 patients after a mean time of 54 ⫾ 26 (SD) days for the following reasons: mouth ulcers (6 patients), wound dehiscence (1 patient), and leukopenia (1 patient). Patients were censored from analysis after discontinuation of RAPA.
403
Dyslipidemia During Sirolimus Therapy
Table 1. Patient Demographics
Total no. of patients Sex distribution (M/F) Mean age (yr) Cause of liver disease Hepatitis C Hepatitic C/alcohol Sclerosing cholangitis Alcohol Other Racial distribution White Hispanic Asian Black
TAC
CSA
P
30 22/8 44.9 ⫾ 11.0
27 18/9 47.3 ⫾ 9.5
NS NS
9 (30) 10 (33) 7 (23) 2 (7) 2 (7)
10 (37) 2 (7) 5 (19) 4 (15) 6 (22)
NS .05 NS NS NS
27 (90) 2 (7) 1 (3) 0 (0)
20 (74) 4 (15) 2 (7) 1 (4)
NS NS NS NS
NOTE. Values expressed as mean ⫾ SD or number (percent).
Nine patients (19%) had to be administered intermittent or chronic prednisone after transplantation for the following indications: autoimmune hepatitis, 4 patients; autoimmune hemolytic anemia, 1 patient; gout, 1 patient; ulcerative colitis, 1 patient; and other, 2 patients. Patient and graft survival were 54 of 57 patients (95%) and 53 of 57 grafts (93%). One patient was started on CSA therapy and then developed new-onset
seizures with a speech disorder. As a result, he was converted to TAC therapy within 4 weeks of transplantation and maintained on TAC. This patient was analyzed in the TAC group. Interestingly, he was the only patient administered TAC to develop hypertriglyceridemia and was 1 of 2 patients who had hypercholesterolemia. Hyperlipidemia was present over 5 months after conversion to TAC therapy. As listed in Table 2, the mean serum cholesterol level for TAC patients was 158 mg/dL compared with 200 mg/dL for CSA patients (P ⫽ .0003). There was a trend toward a greater HDL cholesterol level in TAC-treated patients compared with CSA-treated patients, which did not reach statistical significance. Mean LDL cholesterol level was 18% greater in CSA patients (113 mg/dL) compared with TAC patients (93 mg/dL; P ⫽ .05). Mean serum triglyceride level was 292 mg/dL for CSA patients compared with 134 mg/dL in TAC patients (P ⫽ .002). The ratio of LDL-HDL was 12% greater in CSA patients (P ⫽ not significant [NS]). Hypercholesterolemia (total serum cholesterol ⬎ 240 mg/dL) was present in 10 of 57 patients (18%). The occurrence of hypercholesterolemia in patients on TAC therapy was significantly less (2 of 30 patients; 6%) than in the CSA group (8 of 27 patients; 30%; P ⬍ .05). Hypertriglyceridemia (total serum triglyceride ⬎ 300 mg/dL) occurred in 10 of 57 patients (18%), and the occurrence of hypertriglyceridemia was significantly greater in CSA-
Table 2. Occurrence of Hyperlipidemia and Risk Factors
Cholesterol (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Triglycerides (mg/dL) LDL/HDL Hypercholesterolemia Hypertriglyceridemia Creatinine (mg/dL) RAPA level (ng/mL) RAPA dose (mg/d) Hypertension Diabetes mellitus Average weight gain (kg)
TAC
CSA
Difference (%)
P
158 43 93 134 2.50 2/30 (6%) 1/30 (3%) 1.10 5.13 1.98 7 (23%) 5 (17%) 0.9
200 39 113 292 2.84 8/27 (30%) 9/27 (33%) 1.21 6.07 1.94 9 (33%) 3 (10%) 1.5
21 9 18 54 12 80 91 10 18 2 30 41 40
.0003 NS .05 .002 NS ⬍.05 ⬍.05 .195 .09 NS NS NS NS
NOTE. Lipid levels (cholesterol, HDL, LDL, and triglycerides) represent the mean of all measured levels between months 1 and 6 for patients administered either TAC or CSA. Hypercholesterolemia is defined as serum cholesterol level greater than 240 mg/dL. Hypertriglyceridemia is defined as serum triglyceride level greater than 300 mg/dL. LDL/HDL represents the mean value for all LDL/HDL values measured between months 1 and 6 for all patients administered either TAC or CSA. Serum creatinine, RAPA blood level, and daily RAPA dose represent the mean values for all available data points between months 1 and 6. The occurrence of hypertension, diabetes, and weight gain are shown. Measures of significance for cholesterol, HDL, LDL, triglyceride, LDL/HDL, serum creatinine, RAPA level, RAPA dose, and weight gain were performed using Student’s t-test. Measures of significance for the occurrence of hypercholesterolemia, hypertriglyceridemia, diabetes mellitus, and hypertension were performed using Chi-squared test.
404
Trotter et al
treated patients (9 of 27 patients; 33%) compared with TAC-treated patients (1 of 30 patients; 3%; P ⬍ .05). (Mean serum cholesterol level in the 6 CSA-treated patients administered prednisone was 219 mg/dL. Mean serum cholesterol level in the 3 TAC-treated patients administered prednisone was 177 mg/dL. These values were slightly greater, but not significantly different from those in patients not administered prednisone. Measures of statistical significance were unchanged when these patients were removed from overall analysis. Data not shown.) The mean dose of RAPA administered to patients on CSA therapy (1.98 mg/d) was the same as in TAC patients (1.94 mg/d). However, the mean RAPA blood level in CSA patients (6.07 ng/mL) was greater than in TAC patients (5.13 ng/dL; P ⫽ .08). Mean serum creatinine values were also slightly greater in CSA patients (1.21 mg/dL) than TAC patients (1.10 mg/dL; P ⫽ NS). Hypertension was present in 16 of 57 patients (28%) and was slightly more common in CSA patients (9 of 27 patients; 33%) than TAC patients (7 of 30 patients; 23%; P ⫽ NS). Diabetes mellitus was diagnosed in 8 of 57 transplant recipients (14%); 5 of 30 patients (17 %) were administered TAC and 3 of 27
patients (11%) were administered CSA. Excessive weight gain was uncommon in our patients. The average change in weight after transplantation was 1.5 kg with CSA and 0.9 kg with TAC. Only 2 patients gained more than 10 kg. Comparisons of lipid values were made between months 1 and 6, listed in Table 3 and shown in Figures 2 through 5. In CSA-treated patients, serum cholesterol level increased from 182 to 213 mg/dL at month 6 (P ⫽ .15; Fig. 2). The greatest increase in lipid fraction in CSA-treated patients was triglycerides, which increased from 226 mg/dL at month 1 to 351 mg/dL at month 6 (P ⫽ NS; Fig. 3). As shown in Figure 4, HDL cholesterol level increased from 33 mg/dL at month 1 to 42 mg/dL at month 6 (P ⫽ NS) for CSAtreated patients. LDL cholesterol level increased from 99 mg/dL at month 1 to 122 mg/dL at month 6 (P ⫽ NS; Fig. 5). In TAC patients, total cholesterol level changed negligibly in patients from month 1 (162 mg/dL) to month 6 (161 mg/dL), shown in Figure 2. However, at each monthly interval, serum cholesterol level was lower in TAC patients compared with CSA patients. Statistical significance was reached at months 2 and 3. Triglycer-
Table 3. Serum Lipid and Immunosuppressive Levels in CSA and TAC Patients Month
CSA patients Cholesterol (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Triglycerides (mg/dL) CSA (ng/mL) RAPA (ng/mL) N (cholesterol) N (lipid profile) TAC patients Cholesterol (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Triglycerides (mg/dL) TAC (ng/mL) RAPA (ng/mL) N (cholesterol) N (lipid profile)
1
2
3
4
5
6
P
182 33 99 226 254 5.9 24 7
207 39 113 244 214 5.5 24 9
212 40 96 308 200 5.8 16 11
195 38 103 237 154 8.9 14 7
207 37 125 272 173 5.8 10 5
213 42 122 351 182* 6.8 9 6
NS NS NS NS .001 NS —
162 35 85 123 7.8 4.2 28 11
154 38 90 104 6.9 5.9 26 13
163 44 93 106 6.0 5.2 17 8
167 48 111 143 8.2 6.3 17 8
159 52 87 151 5.5 4.3 9 3
161 47† 85 168 7.3 4.7 11 6
NS .01 NS NS NS NS — —
NOTE. Serum lipid and immunosuppressant levels represent the mean values measured at each time interval. N (cholesterol) represents the number of patients who had a serum cholesterol level measured at the given time interval. N (lipid profile) represents the number of patients who had HDL, LDL, and triglyceride levels measured at the given time interval. Levels of significance were measured between months 1 and 6 using Student’s t-test. *Significant difference in CSA between months 1 and 6. †Significant difference in HDL cholesterol between months 1 and 6.
Dyslipidemia During Sirolimus Therapy
405
Figure 2. Serum cholesterol levels in CSA (white bars) and TAC (black bars) patients. Each data point represents the mean serum cholesterol level for patients (CSA or TAC) at each time interval. For CSA and TAC patients, there was no significant difference in cholesterol level at any monthly interval. For the comparison of TAC versus CSA at each interval, P ⴝ .065 at month 1, P ⴝ .0001 at month 2, P ⴝ .003 at month 3, P ⴝ .11 at month 4, P ⴝ .09 at month 5, and P ⴝ .08 at month 6. *P < .05 for comparisons of TAC versus CSA.
Figure 4. Serum HDL cholesterol levels for CSA (white bars) and TAC (black bars) patients. Each data point represents the mean HDL cholesterol level for patients (CSA or TAC) at each time interval. For CSA patients, there was no significant difference in HDL level at any monthly interval. For TAC patients, there was a significant (*P < .05) difference between HDL levels at months 1 and 6. For comparisons between CSA and TAC patients, the differences were significant (#P < .05) at months 4 and 5.
ide levels, shown in Figure 3, increased only slightly from 123 mg/dL at month 1 to 166 mg/dL at month 6 (P ⫽ NS). Serum triglyceride levels were lower at each monthly interval in TAC compared with CSA patients and reached statistical significance at months 1, 2, and 3. HDL cholesterol level increased from 35 mg/dL at month 1 to 47 mg/dL at month 6 (P ⫽ .05; Fig. 4). At 5 of the 6 monthly intervals (all but month 2), HDL level was greater in TAC compared with CSA patients, reaching statistical significance at months 4 and 5. In TAC patients, LDL cholesterol level was unchanged at 85 mg/dL at months 1 and 6 (Fig. 5).
Discussion The effect of RAPA on lipid levels has been evaluated in 2 clinical trials in renal transplant recipients. In the report by Groth et al,4 renal transplant recipients were randomized to RAPA or CSA therapy. All patients were administered corticosteroids and azathioprine (AZA). RAPA was administered at relatively high doses. Initially, RAPA was administered at a loading dose of 16 to 24 mg/m2/d (9.2 to 13.9 mg/d in an average adult) until days 7 through 10, at which time the dose was reduced to 8 to 12 mg/m2/d (4.6 to 6.9 mg/d) to achieve a blood level of 30 ng/mL for 2 months, then
Figure 3. Serum triglyceride levels for CSA (white bars) and TAC (black bars) patients. Each data point represents the mean serum triglyceride levels for patients (CSA or TAC) at each time interval. For CSA and TAC patients, there was no significant difference in triglyceride levels at any monthly interval. For the comparison of TAC versus CSA at each interval, P < .05 at months 1, 2, and 3. *P < .05 for comparisons of TAC versus CSA.
Figure 5. Serum LDL cholesterol levels for CSA (white bars) and TAC (black bars) patients. Each data point represents the mean LDL cholesterol level for patients (CSA or TAC) at each time interval. For CSA and TAC patients, there was no significant difference in LDL at any monthly intervals. The comparison of CSA versus TAC shows mean LDL level was greater at months 1, 2, 3, 5, and 6 for CSA, although the differences were not significant.
406
Trotter et al
dose-reduced to 15 ng/mL. Corticosteroid therapy (prednisone or methylprednisolone) was weaned to 10 mg by month 6. There was no difference in graft survival, patient survival, incidence and severity of rejection, or OKT3 use in either group. However, hypertriglyceridemia was significantly more common in the RAPA group (51%) versus the CSA group (12%; P ⬍ .01). Hypercholesterolemia occurred in 44% of RAPA patients compared with only 14% of CSA patients (P ⬍ .01). Mean serum cholesterol and triglyceride levels were greatest at 2 months posttransplantation, when the differences between the CSA- and RAPAtreated patients were the greatest. At this follow-up point, the mean triglyceride level was 5.3 versus 2.1 mmol/L (P ⬍ .001), and total serum cholesterol level was 9.2 versus 6.4 mmol/L (P ⬍ .001). In another study, 719 primary renal transplant recipients were randomly assigned to administration of RAPA, 2 and 5 mg, or AZA, 2 to 3 mg/kg, in addition to a CSA-prednisone regimen.5 Initially, transplant recipients were administered 500 mg of methylprednisolone, followed by a tapering dose of prednisone to 10 mg/d by month 6 and 5 to 10 mg/d thereafter. The incidence of acute rejection during the first year was significantly less in both RAPA groups: 22% for 2 mg and 15% for 5 mg versus 31% for AZA. In addition, there were significantly fewer episodes of moderate and severe rejection in both RAPA groups compared with AZA patients. Graft and patient survival rates at 12 months were identical in each of the groups. At 6 months, serum cholesterol levels were significantly greater by 15% with 2 mg of RAPA (6.66 mmol/L) and 24% with 5 mg of RAPA (7.20 mmol/L) compared with AZA (5.80 mmol/L). However, at 12 months, there was no significant difference in cholesterol levels between the 2-mg RAPA group (6.39 mmol/L), 5-mg RAPA group (6.53 mmol/L), and AZA group (5.86 mmol/L). The greater cholesterol and triglyceride levels may be partially explained by the greater doses of corticosteroids administered during the first 6 months. Triglyceride levels were significantly greater in the 2-mg and 5-mg RAPA groups at 6 months compared with the AZA group. However, there was no difference in triglyceride levels in any of the groups at 12 months. The prevalence of hypercholesterolemia at 6 months was 30% and 35% in patients administered 2 and 5 mg of RAPA, respectively. At 12 months, the prevalence of hypercholesterolemia was 33% and 37%, respectively. There are important issues raised by our findings. Our results suggest that RAPA may be administered to liver transplant recipients with a relatively low occur-
rence of hypercholesterolemia. The prevalence of hypercholesterolemia in previously reported studies with RAPA ranged from 30% to 44%. However, in our patients, hypercholesterolemia was uncommon. There are several possible explanations for the decreased incidence of hypercholesterolemia. First, the exclusion of prednisone, which causes hyperlipidemia, from our immunosuppression protocol may have had an important role. Despite the absence of prednisone from our regimen, the occurrence and severity of acute cellular rejection were less than in historical controls.8 Data from the renal transplant studies described previously suggest that corticosteroids potentiate hypercholesterolemia in patients administered RAPA.4,5 In both studies reported previously, corticosteroids were administered at a tapering dose down to 5 to 10 mg at 6 months. Kahan5 showed that cholesterol levels were the lowest at 12 months when corticosteroid doses were tapered to the lowest level. Groth et al4 reported significantly greater cholesterol levels in RAPA-treated patients at month 2, when corticosteroid doses were greater, compared with month 6, when corticosteroid doses were lower. Second, the low incidence of hypercholesterolemia in our patients may be caused by the relatively low doses of RAPA (2 mg/d). As noted in the studies cited previously, the incidence of hypercholesterolemia is associated with increased doses and blood levels of RAPA. Finally, the choice of calcineurin inhibitor administered with RAPA may influence the incidence of hypercholesterolemia. Although the studies by Groth et al4 and Kahan5 both used CSA, our data suggest that the use of TAC was associated with a significantly lower total serum cholesterol level compared with CSA. The occurrence of hypercholesterolemia was reduced by 80% in our patients administered TAC (6%) compared with CSA (30%). Furthermore, only 1 of our patients administered TAC had hypertriglyceridemia compared with 1 of 3 patients administered CSA. The findings of this report could have important implications for the long-term postoperative immunosuppression management of liver transplant recipients. With the technical refinements in liver transplantation over the past 20 years, current survival rates exceed 80% for most recipients. Consequently, the number of liver transplant recipients with measurable long-term survival increases each year. In addition, older patients are undergoing transplantation with greater frequency. As the cohort of older liver transplant recipients increases, greater attention is focused on the prevention of longterm postoperative complications. One of the most important of these is cardiovascular
Dyslipidemia During Sirolimus Therapy
disease. Liver transplant recipients often have concomitant risk factors for atherosclerosis, including hypertension, diabetes, and obesity. A report from our center showed that 31% of long-term liver transplant recipients have hypercholesterolemia and 40% are obese.10 Sheiner et al11 reported that compared with the general population, long-term (survival ⬎ 5 years) liver transplant recipients have a significantly greater risk for hypertension (3.07 times the general population) and diabetes (5.99 times the general population). Therefore, immunosuppressive regimens such as the one presented in this report could potentially reduce the incidence of hypercholesterolemia. This could in turn prevent the development of atherosclerosis in long-term survivors of liver transplantation. Our practice has been to follow the American Heart Association guidelines for the treatment of hyperlipidemia.12 However, we typically wait at least 6 months after transplantation before initiating cholesterolreducing medication. The follow-up period in our patients was relatively short. Consequently, we treated only one patient for hyperlipidemia; gemfibrozil was successfully used to treat one patient with a serum triglyceride level greater than 800 mg/dL. Previous studies have documented more favorable lipid profiles associated with TAC compared with CSA. Guckelberger et al13 reported lipid profiles in 302 patients with a median follow-up of 18 months. Hypercholesterolemia was significantly less common in TAC (53%) than CSA patients (76%; P ⬍ .05). Hypertriglyceridemia was present in 40% of patients administered CSA and 33% administered TAC (P ⫽ NS). Canzanello et al14 published similar results in 63 patients. These investigators showed that hypercholesterolemia was present in 33% of patients administered CSA at 12 months after liver transplantation compared with 0% of patients administered TAC. Serum cholesterol levels were significantly greater with CSA (225 mg/dL) than with TAC (159 mg/dL; P ⬍ .05). Corticosteroids were administered in both these studies and may have contributed to the development of hyperlipidemia. There is evidence that RAPA may interact with CSA to increase patient exposure to CSA. This may increase the toxic effects of CSA when both drugs are administered concomitantly. Kahan5 reported significantly greater serum creatinine levels in renal transplant patients administered RAPA compared with AZA. He further noted that lower oral doses of CSA were required to achieve the same blood CSA levels and area under the curve values in RAPA-treated patients compared with AZA-treated patients. These effects oc-
407
curred despite the delayed administration of RAPA 4 hours after CSA. The exact mechanism of the interaction of CSA and RAPA has not been fully delineated. However, possible mechanisms include competitive metabolism of both drugs through the cytochrome P-450 system or altered distribution or clearance of each drug.5 The additive renal toxic effect of CSA seen with concomitant administration of RAPA may be caused in part by increased distribution of CSA to renal tissue. Napoli et al15 reported increased concentrations of CSA in renal tissue in rats administered RAPA with CSA. We also noted greater serum creatinine levels in our patients coadministered RAPA and CSA compared with TAC. In addition, CSA patients had RAPA blood levels 18% greater than TAC patients. This suggests the possibility that RAPA exposure may be greater with the concomitant administration of CSA. This observation awaits confirmation with further studies. Our report has several flaws. Data collection was retrospective. As such, we had difficulty obtaining LDL, HDL, and triglyceride levels at each follow-up interval. This was especially true in patients who had laboratory tests drawn at sites far from our center. Consequently, not all patients had these values recorded at each follow-up interval. The number of patients with serum cholesterol and lipid profiles drawn at each follow-up interval is listed in Table 3. Finally, although our primary immunosuppressive protocol was steroid free, some patients were administered prednisone for clinical indications. These patients were included in the analysis and as noted, their removal did not change our findings. In summary, we report the lipid profiles in primary liver transplant recipients administered RAPA and either TAC or CSA with only 3 days of corticosteroid therapy. The prevalence and severity of hyperlipidemia was extremely low in patients treated concomitantly with RAPA and TAC. We speculate that prevention of hyperlipidemia could prevent atherosclerosis and its complications in long-term survivors of liver transplantation. Currently, we are formally evaluating the efficacy and side-effect profile of RAPA in liver transplant recipients in a trial underway at our institution.
References 1. Watson CJ, Friend PJ, Jamieson NV, Frick TW, Alexander G, Gimson AE, Calne R. Sirolimus: A potent new immunosuppressant for liver transplantation. Transplantation 1999;67:505-509. 2. Abraham RT, Wiederrecht GJ. Immunopharmacology of rapamycin. Annu Rev Immunol 1996;14:483-510. 3. Sehgal SN. Rapamune®: Mechanism of action of immunosup-
408
4.
5.
6.
7.
8.
9. 10.
Trotter et al
pressive effect results from blockade of signal transduction and inhibition of cell cycle progression. Clin Biochem 1998;31:335340. Groth CG, Backman L, Morales JM, Calne R, Kreis H, Lang P, et al. Sirolimus (rapamycin)-based therapy in human renal transplantation. Transplantation 1999;67:1036-1042. Kahan BD for the Rapamune® US Study Group. Efficacy of sirolimus compared with azathioprine for reduction of acute renal allograft rejection. Lancet 2000;356:194-202. McAlister VC, Gao Z, Peltekian K, Domingues J, Mahalati K, MacDonald AS. Sirolimus-tacrolimus combination immunosuppression. Lancet 2000;355:376-377. Stegall MD, Wachs ME, Everson G, Steinberg T, Bilir B, Shrestha R, et al. Prednisone withdrawal 14 days after liver transplantation with mycophenolate: A prospective trial of cyclosporine and tacrolimus. Transplantation 1997;64:1755-1760. Trotter JF, Wachs M, Bak T, Trouillot T, Stolpman N, Everson GT, Kam I. Liver transplantation using sirolimus and minimal corticosteroids (3-day taper). Liver Transpl 2001;7:343-351. Wyeth-Ayerst Laboratories. Production information, Rapamune®. Philadelphia, PA: Wyeth-Ayerst Laboratories, 1999. Stegall MD, Everson G, Schroter G, Bilir B, Karrer F, Kam I.
11.
12. 13.
14.
15.
Metabolic complications after liver transplantation. Diabetes, hypercholesterolemia, hypertension, and obesity. Transplantation 1995;60:1057-1060. Sheiner PA, Magliocca JF, Bodian CA, Kim-Schluger L, Altaca G, Guarrera JV, et al. Long-term medical complications in patients surviving ⬎ or ⫽ 5 years after liver transplant. Transplantation 2000;69:781-789. Consensus Panel Statement. Guide to primary prevention of cardiovascular diseases: Summary. Circulation 1995;92:2-4. Guckelberger O, Bechstein WO, Neuhaus R, Luesebrink R, Lemmens HP, Kratschmer B, et al. Cardiovascular risk factors in long-term follow-up after orthotopic liver transplantation. Clin Transplant 1997;11:60-65. Canzanello VJ, Schwartz L, Taler SJ, Textor SC, Wiesner RH, Porayko MK, Krom RAF. Evolution of cardiovascular risk after liver transplantation: A comparison of cyclosporine A and tacrolimus. Liver Transpl Surg 1997;3:1-9. Napoli KL, Wang ME, Stepkowski SM, Kahan BD. Relative tissue distributions of cyclosporine and sirolimus after concomitant peroral administration to the rat: Evidence for pharmacokinetic interactions. Ther Drug Monit 1998;20:123-133.