Hindawi Publishing Corporation
Journal of Transplantation
Volume 2011, Article ID 913094, 7 pages
A Decade ofExperience Using mTorInhibitorsin
1Division of Transplant Surgery, Department of Surgery, School of Medicine, University of Colorado Denver, Aurora, CO 80045, USA
2Baylor University Medical Center, Transplant, 4 Roberts, 3500 Gaston Avenue, Dallas, TX 75246, USA
Correspondence should be addressed to Jeffrey Campsen, firstname.lastname@example.org
Received 12 October 2010; Revised 20 December 2010; Accepted 4 January 2011
Academic Editor: Gian Luigi Adani
Copyright © 2011 Jeffrey Campsen et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
Some studies suggest that Sirolimus (SRL) is associated with an increased risk of death in liver transplant recipients compared
to treatment with calcineurin inhibitors (CNIs). We compared patients who received SRL or CNI in the first year after liver
transplant. Our database included 688 patients who received a liver transplant. The patients were divided into groups. (1) CNI
+ MPS (mycophenolate sodium) at time of discharge. (2) CNI + MPS at time of discharge; SRL was added within the first 6
months and continued through the first year. (3) CNI + MPS at time of discharge; SRL was added within the first 6 months
and discontinued before the first year. (4) SRL as primary immunosuppression. (5) SRL as primary immunosuppression and
discontinued before the first year. We used mortality and graft loss as the primary measures of outcome. We also quantified renal
function using the change in glomerular filtration rate (GFR), the presence of biopsy proven acute cellular reject (ACR), and
steroid-resistantrejection (SRR). There were no significantdifferences in mortalityorgraftloss.There was no difference in patient
or graft survival. Patients that received SRL as primary immunosuppression had 50% less rejection compared to controls.
Sirolimus (rapamycin) is a macrolide lactone that was
approved for use as an immunosuppressant in 1999 , but
not for use in liver transplantation. It suppresses the T-cell
response to interleukin-2 by binding to and inhibiting the
mammalian target of rapamycin (m-TOR) . There are
reports of benefits and risks for the use of Sirolimus (SRL),
in liver transplantation [2–17]. Investigators have reported
an increase in renal failure, hepatic artery thrombosis, and
overall post-transplant mortality compared to the use of
calcineurin inhibitors (CNIs) [6, 11, 18]. In contrast, other
studies report good outcomes and that SRL has a renal-
sparing effect .
Our center has routinely used SRL for immunosuppres-
sion following livertransplantation [1, 18–26]. We have used
SRL as primary therapy along with a CNI in the early post-
operative period starting in January 2000 until it received
a “black box” warning on the label for increased risk of
hepatic artery thrombosis. After that, we converted patients
hospital. Internal review of our database showed no increase
in morbidity and/or mortality in our SRL patients compared
sodium (MPS). Our preliminary data suggested that SRL
can reduce donor graft rejection and could ameliorate renal
injury secondary to increased use of CNI.
This study seeks to determine if there is an increase risk
of complications associated with the use of SRL in the first
year after liver transplantation. We collected data from all
patients who had received SRL in the first year after liver
transplant. Because of previously reported markers of poor
outcome associated with the use of SRL, this study included
Further, because our previous data suggested that SRL may
decrease the incidence of rejection, we measured the rates of
acute cellular rejection (ACR) and steroid resistant rejection
(SRR) in our patient population .
2 Journal of Transplantation
2.1. Study Design. This study was approvedby the University
of Colorado Internal Review Board. We retrospectively
reviewed the University of Colorado Denver transplant
database and collected data on all patients who received a
liver transplant between January 2000 and November 2009.
This included 688 patients. An independent investigator
extracted data from both electronic and paperfiles. Less than
3% of data was missing.
2.2. Immunosuppression Protocols. The pattern of immuno-
suppressant use allowed us to construct five study cat-
egories from the 688 patients in the database. Patients
were assigned to one of the five categories according the
immunosuppressive therapy given during the first year
The five treatment groups were.
(1) Patients received a CNI + MPS at time of discharge
(primary therapy) and through the first year of
therapy. They never received a single dose of SRL.
(2) Patients received a CNI + MPS at time of discharge;
SRL was added within the first 6 months and
continued through the first year.
(3) Patients received a CNI + MPS at time of discharge;
SRL was added within the first 6 months and
discontinued before the first year.
(4) SRL was started as primary immunosuppression
before discharge from hospital after transplantation
combined with other varying therapies including a
CNI and continued for the first year.
(5) SRL was started as primary immunosuppression
before discharge from hospital after transplantation
combined with other varying therapies including a
CNI and discontinued before the first year.
The patients in group 1 were used as Controls for
the study. Groups 2 and 3 patients represented Conversion
Groups. The Primary Treatment Group is comprised of
patients in Groups 4 and 5. Each patient was assigned to
one of the five groupings. Outcomes were then compared
between the five groups.
2.3. Primary and Secondary Outcomes of This Study. In this
study, five endpoints were compared.
(1) Graft failure rate (time to graft failure/death).
(2) Mortality rate (survival time).
(1) Acute cellular rejection rate (proven by biopsy or
(2) Steroid resistant rejection (biopsy proven and treated
with thymoglobulin or OKT3).
Graft rejection was diagnosed by liver biopsy or by elevation
of liver function test in the absence of other causes of graft
dysfunction. If the liver function tests were still elevated
after treatment with pulse dose steroids and the patient
had not received a biopsy, one was performed. ACR that
was not responsive to pulse dose steroids and had biopsy
proven rejection was then deemed SRR and treated with
thymoglobulin or OKT3.
(3) Renal function as defined by GFR was recorded at
discharge, 1 month, 6 months, and 1 year.
2.4. Statistical Analysis. Univariate: we computed the P
values for graft failure (failure-free survival) and mortality
(patient survival) rate comparisons using the log rank test.
The event-free (“survival”) curves were computed using
the Kaplan-Meier method. For graft failure, mortality, and
rejection events, the event rate in any group was computed
total person − months followup
Event rate = 1000 ∗
total number of events
This rate is in units of events per 1000 person-months
(p-mos) of followup and is a summary statistic for the
corresponding time to event curve. Rate ratios are also
reported where the denominator is the rate in the reference
control group and the numerator is the rate in the group
being compared to the control group.
The P values for comparing failure rates and mortality
rates are computed using the log rank test. For rejection
rate comparisons, P values were computed assuming an
underlying Poisson process. This assumption was made (in
part) since the time to each acute rejection episode was
not available. Analysis of deviance results showed that the
Poisson model is a good fit to the data.
A one-way ANOVA was used to compare means for
continuous variables at baseline including GFR across the
five groups after confirming that a parametric model was
appropriate for each on the original scale. The chi-square
test/Fisher test was used to compute P values for comparing
binary variables such as gender, hypertension, diabetes, and
any particular diagnosis such as hepatocellular carcinoma
(HCC) or Hepatitis C virus (HCV). The P value for
comparing race/ethnicity was also computed using the chi-
3.1. Patient Groups and Demographics. The five groups of
patients are listed with the patient demographics listed in
Table 1. The demographic features of patients in groups 2
to 5 were comparable to the Control Group 1. However,
patients differed in their MELD score and warm ischemic
time (WIT). Patients in groups 4 and 5 (Primary Treatment
Group) had lower MELD score (18 versus 22). WIT was
longer in group 4 patients compared to controls; however,
this was not statistically significant. There were also fewer
Journal of Transplantation3
Table 1: Recipient demographics.
Age (median years)
Gender male (%)
Donor age (median years)
CIT (median minutes)
WIT (median minutes)
Live donor (%)
Primary transplant indication (%)
Statisticallysignificant values are bolded:∗P < .01,∗∗P < .001.
live donor liver transplants among group 2 patients. The
Conversion Groups (2 and 3) had a lower pre-transplant
GFR compared to the Control Group 1. There were fewer
transplants done for HCV in Group 4 and less for ETOH in
Group 3 patients.
3.2. Primary Endpoints
Graft Survival. Figure 1 depicts plots and tables of graft
survival across the five groups represented by failure rates
per 1000 person-months followup. There was no statistical
differencein graft survival among the five groups ofpatients.
Patient Survival. Figure 2 depicts plots and tables of patient
survival across the five patient groups represented by failure
rates per 1000 person-months follow up. There was no
statistical difference in patient survival.
3.3. Secondary Endpoints
Acute Cellular Rejection. Table 2 depicts ACR rates per 1000
person-months of followup. ACR rates in the Groups 4 and
5 (Primary Treatment Group) where SRL was started before
Table 2: Acute cellular rejection rates per 1000 person-months.
Person mos f/u
Rate per 1000 p-mon
discharge (P value ≤ .0001 and .0007) are about 50% less
than the Control Group 1 and patients in the Conversion
Steroid Resistant Rejection. Table 3 depicts SSR rates per
1000 person-months of followup and reflects the results
from ACR. The incidence of ACR in the Primary Treatment
Groups (4 and 5) (P value = .0004 and .038)was about50%
less than the Control Group 1.
Since ACR and SRR rates show clinically and statistically
significant differences, the adjusted rates were also computed
using Poisson regression as above with all 20 covariates
included in the model. The 20 covariates were: recipient
4 Journal of Transplantation
Graft survival by group
Crude graft failure rates per 1000 person-months followup
4 12671 23
Failed rate per
Figure1: Graftsurvival by group. Itdepicts plots andtablesof graft
survival across the five groups represented by failure rates per 1000
Table 3: Steroid-resistant rejection rates per 1000 person-months.
Person mos f/u
Rate per 1000 p-mon
age, MELD score, pre-transplant GFR, BMI, donor age, CIT,
WIT, male gender, ethnicity, cadaveric donor, pre-transplant
diabetes, pre-transplant hypertension, HCC, HCV, PSC,
Laennec’s cirrhosis, HBV, AIH, NASH, and PBC. The
adjusted rate ratios are similar to the unadjusted rate ratios
showing that the Primary Treatment Group had about half
the rejection rate compared to the Control Group.
Glomerular Filtration Rate. Figure 3 depicts the glomerular
filtration rate (GFR)(mL/min) forthe five groupsofpatients
at the times of hospital discharge, 6 months, and 1 year.
As seen in Table 2, the median pre-transplant GFR of the
Conversion Groups (2 and 3) (P value = .0136 and .074)
was significantly less compared to the Control Group 1 at
time of transplantation. In Figure 3, the percent change in
GFR was calculated from pre-transplant value to the value
for GFR that was obtained at 1 year after transplantation. At
6 months, patients in categories 2 and 4 had a drop in mean
GFR (P value
= .0793 and .0465) that was significantly
more than the corresponding drop in controls. At one year,
only group 4 had a significantly worse GRF than the Control
Group with a change in GRF of −17.3% (P value = .0320).
Patient survival by group
Crude mortality rate per 1000 person-months
Group Mos F/U
Dead Mort rate per
pct alive (%)
12 2436 4860728496
Figure 2: Patient survival by group. It depicts plots and tables of
patient survival across the five patient categories represented by
failure rates per 1000 person-months followup.
While the behavior in group 5 is similar to group 4, the
sample size is lower, making the P value larger.
At one year, 46 of the original 328 persons in the Control
Group were missing, including 17 who died or were re-
transplanted. We assumed that the missing patients had
outcomes that were worse than those who remained in the
databank. Thus, the apparent mean of 6.54mL/min decrease
in GFR in the control group could be an underestimate.
The decrease in GFR is likely more than the observed
6.54mL/min if we included those patients lost to followup.
So, the one year mean GFR difference between the Control
Group and the 14mL/min drop in category 4 patients (SRL
kept) is likely smaller than the results shown.
3.4. Tacrolimus Levels. Figure 4 depicts mean Tacrolimus
levels across all groups at 1 year. Only Group 2 patients were
significantlylessthantheControlGroup1(P value = .0011).
All groups at 1 year had a mean level less than 6.0 (ng/mL).
Our data shows that there is no increase in patient death or
donor graft loss when SRL is used as primary therapy or
when patients are converted to SRL following therapy with
other immunosuppressants. Further, we have no evidence
that suggestsSRLhasdeleteriouseffectsupon renal function.
In contrast, ourdatasuggeststhat primary therapywith SRL,
but not conversion to SRL, reduced the incidence of acute
Journal of Transplantation5
Pre-TX Dischargemo1 mo6 year1
Group 1 Group 2Group 3Group 4Group 5
Figure 3: GFR (mL/min)outcome (mean). It depicts the glomeru-
lar filtration rate (GFR) (mL/min)for the five categories of patients
at the times of hospital discharge, 6 months, and 1 year.
2468 10 12
Mean tacrolimus (ng/mL) by group
Mean level (ng/mL)
Group 1 control
Group 2 SA added
Group 3 SA added/discontinued
Group 4 SA primary
Group 5 SA primary/discontinued
Figure 4: Mean Tacrolimus levels at 1 year across all five groups. It
depicts mean Tacrolimus levels across all groups at 1 year.
and steroid resistant rejection in transplant recipients. This
may be an effect of increased immunosuppression.
Sirolimus has been widely used in renal transplantation
butis not currently approved for use in liver transplantation.
There are reports that SRL has negative effects on outcome
including HAT, delayed wound healing, and increased mor-
tality compared to standard immunotherapy [14, 15]. Our
center has used SRL therapy after liver transplantation at a
number of different time points. Few other centers have this
large experience using SRL in liver transplantation coupled
with a nearly complete data set. Data from our previous
publications did not support claims that the use of SRL is
associated with negative outcomes[1, 18, 19, 22–26]. Rather,
our previous data suggested that mTor inhibitors are safe
when used as single therapy or in addition to CNI after liver
transplant. Weobservedthiswhen SRLwasusedasaprimary
immunotherapy or if SRL was added later.
We administered SRL to patients in two regiments:
we gave SRL immediately after liver transplant or added/
substituted SRL within the first year following transplanta-
tion. There was a range ofreasons for different combinations
of therapies over the first year mostly from evolving proto-
cols. The Conversion and Primary Treatment Groups repre-
sents a more heterogeneous collection of immunotherapies
and must be interpreted with caution. However, comparison
between the categories of patients within both groups did
not show significant differences. We did not find differences
between patient and/or donor graft survival between our
patients who received SRL compared to those who received
CNI + MPS. Similarly, there were no differences in patient
and graft outcome that were related to whether SRL was
primary or conversion therapy.
Oursecondary measures ofoutcomewere graftrejection.
Patients that received SRL as primary immunosuppression,
defined as SRL administered before discharge from the
hospital after liver transplant, had less ACR and SRR
rejection. This was true for patients that stayed on SRL for
one year and for patients that stopped using SRL before one
year. This finding was statistically significant, and the rates of
rejection were 50% less than the control of CNI + MPS.
Finally, we reviewed our data to see if there is a difference
in renal function when SRL is used. It is well understood
that CNI can cause progressive kidney injury. SRL has
been used to reduce the level of CNI, thus, protecting the
kidney . Our data did not support this theory. All five
groups had a reduction in GFR after transplant. We believe
there are two reasons why our study does not support
transplanted before 6 months or 1 year, their corresponding
missing GFR values create a survival biases. Thus, the
estimates of mean GFR at one year in particular may be
biased because sicker patients that have a high mortality
and also a greater reduction in GFR. Second, as seen in
Figure 3, our center keeps the Tacrolimus levels in all groups
at one year less than six. Thus, the lower levels of CNI
We suggest that these data be interpreted with caution.
First, it is a retrospective study and the Conversion and
Primary Treatment Groups contain a heterogeneous group
ofpractices. However, this data isalmost 100%completeand
accurately describes our practice patterns and subsequent
the first year. Secondly, because it is retrospective and we
only reviewed our practice patterns for the first year, our
groups are not set up to adequately assess long-term renal
6 Journal of Transplantation
outcome.Rather, we can only predict kidney function within
the first year of transplantation. Finally, we do not routinely
perform protocol biopsies to diagnose ACR, so we could
have overestimated the actual incidence of ACR in our
patient population. However, the same diagnostic criteria
were applied to all categories of patients that we studied.
Therefore, it is unlikely that our diagnostic approach favored
the diagnosis of ACR in one category of patient compared to
In conclusion, our goals for this study were to determine
if our use of SRL during the first year after liver transplant
increased mortality or morbidity. It did not. Surprisingly,
SRL use as a primary therapy decreased our rates of both
ACR and SRR. Based on this review, we will continue to
use SRL in liver transplant recipients. With the introduction
of new mTor inhibitors, such as everolimus, improved
immunosuppression combinations may be developed based
on our successful use of SRL.
Acute cellular reject
Glomerular filtration rate
Jeffrey Campsen designed and performed research, collected
and analyzed data, and wrote the paper. Michael Zimmer-
man designed research, analyzed data, and wrote the paper.
Susan Mandell analyzed data. Maria Kaplan performed
research. Igal Kam designed research, analyzed data, and
wrote the paper.
This work was supported in part by a research grant from
 J. Campsen, M. A. Zimmerman, J. F. Trotter et al., “Sirolimus
and liver transplantation: clinical implications for hepatocel-
lular carcinoma,” Expert Opinion on Pharmacotherapy, vol. 8,
no. 9, pp. 1275–1282, 2007.
 R. J. Firpi, T. T. Tran, P. Flores et al., “Sirolimus-induced
hyperlipidaemia in liver transplant recipients is not dose-
dependent,” Alimentary Pharmacology and Therapeutics, vol.
19, no. 9, pp. 1033–1039, 2004.
 A. Fisher, J. M. Seguel, A. N. de la Torre et al., “Effect of
Liver Transplantation, vol. 10, no. 2, pp. 193–198, 2004.
 D. Kniepeiss, F. Iberer, B. Grasser, S. Schaffellner, and K. H.
Tscheliessnigg, “Sirolimus and mycophenolate mofetil after
liver transplantation,” Transplant International, vol. 16, no. 7,
pp. 504–509, 2003.
 V. C. McAlister, K. M. Peltekian, D. A. Malatjalian et al.,
“Orthotopic liver transplantation using low-dose tacrolimus
and sirolimus,” Liver Transplantation, vol. 7, no. 8, pp. 701–
 M. Montalbano, G. W. Neff, N. Yamashiki et al., “A ret-
rospective review of liver transplant patients treated with
sirolimus from a single center: an analysisof sirolimus-related
complications,” Transplantation, vol. 78, no. 2, pp. 264–268,
 S. Nair, J. Eason, and G. Loss, “Sirolimus monotherapy in
nephrotoxicity due to calcineurin inhibitorsin liver transplant
recipients,” Liver Transplantation, vol. 9, no. 2, pp. 126–129,
 G. W. Neff, M. Montalbano, G. Slapak-Green et al., “A
retrospective review of sirolimus (repamune) therapy in
orthotopic liver transplant recipients diagnosed with chronic
rejection,” Liver Transplantation, vol. 9, no. 5, pp. 477–483,
 R. Perdigoto, C. Sequeira, L. Tom´ e et al., “Sirolimus: efficacy
in liver transplant patients with nephrotoxicity and renal
insufficiency,” Transplantation Proceedings, vol. 39, no. 8, pp.
 R. J. Roberts, A. C. Wells, E. Unitt et al., “Sirolimus-
induced pneumonitis following liver transplantation,” Liver
Transplantation, vol. 13, no. 6, pp. 853–856, 2007.
 C. Toso,G. A.Meeberg, D. L.Bigametal.,“Denovosirolimus-
based immunosuppression after liver transplantation for hep-
atocellular carcinoma: long-term outcomes and side effects,”
Transplantation, vol. 83, no. 9, pp. 1162–1168, 2007.
 C. Toso, S. Merani, D. L. Bigam, A. M. J. Shapiro, and
N. M. Kneteman, “Sirolimus-based immunosuppression is
associated with increased survival after liver transplantation
for hepatocellular carcinoma,” Hepatology, vol. 51, no. 4, pp.
 C. J. E. Watson, A. E. S. Gimson, G. J. Alexander et al.,
“A randomized controlled trial of late conversion from cal-
cineurin inhibitor (CNI)-based to sirolimus-based immuno-
suppression in liver transplant recipients with impaired renal
function,” Liver Transplantation, vol. 13, no. 12, pp. 1694–
 R. K. G. Weisner, S. McDiarmid, and Rapamune Liver
Transplant Study Group, “Sirolimus immunotherapy results
in reduced rates of acute rejection in de novo orthotopic liver
transplant recipients,” American Journal of Transplantation,
vol. 2, p. 464, 2002.
 R. Wiesner, “for the RapamuneLiver TransplantStudy Group.
The safety and efficacy of sirolimus and low-dose tacrolimus
vs. tacrolimus in de novo orthotopic liver transplant recipi-
ents: results from a pilot study,” Hepatology, vol. 36, p. 208A,
 Y. J. Yang, LI. X. Li, Q. He et al., “Sirolimus as primary
immunosuppressant for calcineurin inhibitor-related renal
insufficiency after liver transplantation,” Hepatobiliary and
Pancreatic Diseases International, vol. 6, no. 4, pp. 376–378,
 H. Zaghla, R. R. Selby, L. S. Chan et al., “A comparison of
sirolimus vs. calcineurin inhibitor-based immunosuppressive
therapies in liver transplantation,” Alimentary Pharmacology
and Therapeutics, vol. 23, no. 4, pp. 513–520, 2006.
 J. F. Trotter, “Sirolimus in liver transplantation,” Transplanta-
tion Proceedings, vol. 35, no. 3, supplement, pp. 193S–200S,
 J. C. Dunkelberg, J. F. Trotter, M. Wachs et al., “Sirolimus as
primary immunosuppression in liver transplantation is not
Journal of Transplantation7
associated with hepatic artery or wound complications,” Liver
Transplantation, vol. 9, no. 5, pp. 463–468, 2003.
 G. S. Jensen, A. Wiseman, and J. F. Trotter, “Sirolimus
conversion for renal preservation in liver transplantation: not
so fast,” Liver Transplantation, vol. 14, no. 5, pp. 601–603,
 M. Molinari, K. Berman, G. Meeberg et al., “Multicentric
outcome analysis of sirolimus-based immunosuppression in
252 liver transplant recipients,” Transplant International, vol.
23, no. 2, pp. 155–168, 2010.
 J. F. Trotter, M. Wachs, T. Bak et al., “Liver transplantation
using sirolimus and minimal corticosteroids (3-day taper),”
Liver Transplantation, vol. 7, no. 4, pp. 343–351, 2001.
 J. F. Trotter, M. E. Wachs, T. E. Trouillot et al., “Dyslipidemia
during sirolimus therapy in liver transplant recipients occurs
with concomitant cyclosporine but not tacrolimus,” Liver
Transplantation, vol. 7, no. 5, pp. 401–408, 2001.
 J. F. Trotter, A. Wallack, and T. Steinberg, “Low incidence
of cytomegalovirus disease in liver transplant recipients
receiving sirolimus primary immunosuppression with 3-day
corticosteroid taper,” Transplant Infectious Disease, vol. 5, no.
4, pp. 174–180, 2003.
 M. A. Zimmerman, J. F. Trotter, M. Wachs et al., “Sirolimus-
based immunosuppression following liver transplantation for
hepatocellular carcinoma,” Liver Transplantation, vol. 14, no.
5, pp. 633–638, 2008.
 M. A. Zimmerman, J. F. Trotter, M. Wachs et al., “Predictors
of long-term outcome followingliver transplantation for hep-
atocellular carcinoma: a single-center experience,” Transplant
International, vol. 20, no. 9, pp. 747–753, 2007.