Effect of Corticosteroid Withdrawal on Tacrolimus and Mycophenolate Mofetil Exposure in a Randomized Multicenter Study


Corresponding author: Fuad S. Shihab, fuad.shihab@hsc.utah.edu


As corticosteroid-sparing protocols are increasingly utilized in kidney transplant recipients, it is crucial to understand potential drug interactions between tacrolimus (TAC) and the effect of corticosteroid withdrawal as well as to characterize dose adjustments of mycophenolate mofetil (MMF) in this setting. This prospective, multicenter, randomized, double-blind study included 397 patients who were randomized on posttransplant day 8 to receive either placebo (CSWD) or corticosteroid continuance (CCS). TAC trough levels at week two posttransplant were significantly greater in the CSWD group whereas TAC doses were comparable to the CCS group. This interaction was not observed in the African American subgroup. Higher serum creatinine and potassium levels were also observed in the CSWD group. MMF dose was significantly reduced in the CSWD group by the investigators because of decreased WBC counts, mostly outside of study protocol criteria, despite similar incidence of neutropenia and reported cytomegalovirus infection. Understanding TAC and MMF exposure in the context of corticosteroid-sparing protocols should allow for improved dosing of immunosuppressants and better management of posttransplant patients.


absolute neutrophil count


chronic allograft nephropathy


calcineurin inhibitor


cytochrome P450 3A4


interleukin-2 receptor


mycophenolate mofetil




posttransplant day


serum creatinine


serum potassium




white blood cell.


Corticosteroids have been a mainstay of immunosuppressive therapy for kidney transplant recipients for over 50 years. Although corticosteroids are effective at preventing rejection in transplant recipients, the consequence of the adverse events associated with its long-term use has prompted interest in corticosteroid-sparing immunosuppressive regimens. Literature in kidney transplant recipients has demonstrated that corticosteroid withdrawal immunosuppressive regimens provide similar long-term patient and graft survival and improved cardiovascular profile compared to chronic corticosteroid use [1-4]. As a result, there has been a decrease in the percentage of kidney transplant recipients who receive corticosteroids at time of transplant from 95.1% in 1998 to 65.7% in 2009 [5].

The move away from the use of corticosteroids in kidney transplant recipients may have clinical consequences as an interaction between corticosteroids and tacrolimus (TAC) has been hypothesized based on pharmacokinetic and pharmacodynamic characteristics of the two drugs where TAC is a known cytochrome P450 3A4 (CYP3A4) and P-glycoprotein (PGP) substrate and corticosteroid is a known CYP3A4 and PGP inducer [6, 7]. This interaction has been clinically observed in small single-center studies where TAC trough levels increase in conjunction with corticosteroid taper [8-12]. TAC is known to have a narrow therapeutic index with considerable intra- and interpatient variation and therefore, it is crucial to better understand this interaction and its clinical impact as the use of corticosteroid withdrawal protocol in kidney transplant recipients becomes increasingly prevalent.

Furthermore, the effect of corticosteroid withdrawal on mycophenolate mofetil (MMF) dosing has not been fully investigated. Because of the leukocytosis-inducing effects of corticosteroids, it is anticipated that corticosteroid withdrawal would exacerbate the myelosuppressive effects of MMF. This is concerning because MMF dose reduction secondary to side effects has been associated with increased risk for acute rejection and decreased graft function in kidney transplant recipients [13-15].

The Astellas Corticosteroid Withdrawal Trial Study Group recently reported the 5-year results of early corticosteroid withdrawal in combination with TAC and MMF compared to continued corticosteroid use in kidney transplant recipients [3]. The study did not demonstrate a difference between patients maintained on corticosteroids and those who underwent corticosteroid withdrawal with respect to the composite of death, graft loss or moderate/severe acute rejection at 5 years posttransplant. However, the corticosteroid withdrawal group experienced greater rates of biopsy proven acute rejection at 5 years posttransplant but the difference was not statistically significant. In addition, post hoc analysis demonstrated a relatively higher incidence of chronic allograft nephropathy (CAN) in the corticosteroid withdrawal group. Herein, we sought to confirm the presence and evaluate the clinical significance of the pharmacokinetic drug–drug interaction between corticosteroids and TAC. Furthermore, we analyzed and described dose adjustment of MMF in the setting of corticosteroid withdrawal.

Materials and Methods

This clinical trial is registered at www.clinicaltrials.gov (NCT00650468) and meets all CONSORT criteria for publication of randomized trials. The study protocol was institutional review board approved at each study site and conducted according to IRB standards at each institution. Informed consent was obtained before enrollment.

The overall study design, methods and outcomes of the Astellas Corticosteroid Withdrawal Trial have been previously published; this study is a subanalysis of the aforementioned trial and no additional inclusion/exclusion criteria were imposed [3]. In brief, this was a 5-year, prospective, multicenter, randomized, double-blind study of early corticosteroid withdrawal in kidney transplant recipients. All patients received the following unblinded corticosteroid regimen through posttransplant day (PTD) 7: methylprednisolone perioperative 10 mg/kg; PTD 1, 5 mg/kg; PTD 2, 3 mg/kg; prednisone PTD 3, 2 mg/kg; PTD 4, 1 mg/kg; PTD 5, 0.7 mg/kg; PTD 6, 0.5 mg/kg; PTD 7, 0.4 mg/kg. On PTD 8, patients received blinded study drug as capsules (CSWD group–placebo; CCS group–prednisone). Prednisone dosing in the CCS group was: PTD 8–14, 0.4 mg/kg; PTD 15 to 29, 0.3 mg/kg; PTD 30–89, 0.2 mg/kg; PTD 90–119, 0.15 mg/kg; PTD 120–180, 0.1 mg/kg; PTD > 180, 5 mg/day. Study drug dosing (steroid or placebo) was based on actual body weight and rounded to the nearest 5 mg. All patients received either antithymocyte globulin (Thymoglobulin; Genzyme, Cambridge, MA, USA; 1.5 mg/kg × 4 doses) or interleukin-2 receptor (IL-2R) antibody (basiliximab [Simulect; Novartis, East Hanover, NJ, USA] or daclizumab [Zenapax; Roche, Nutley, NJ, USA] dosed per package insert) induction therapy according to center preference. Maintenance immunosuppressive regimen consisted of TAC (Prograf; Astellas, Deerfield, IL, USA) and MMF (Cellcept®; Genentech, San Francisco, CA, USA). TAC was initiated within 72 h posttransplant at a starting oral dose of 0.15-0.2 mg/kg/day using actual body weight. Changing body weight was accounted for when calculating TAC dose (mg/kg/day) by using actual body weight at each clinic visit. TAC dose was titrated to target trough levels of 10–20 ng/mL by PTD 7 through PTD 90, then 5–15 ng/mL beyond PTD 90. TAC trough levels were measured by microparticle enzyme immunoassay in whole blood at the time points specified in the original study protocol [4]. MMF was given preoperatively and patients were maintained on 2 gm/day long-term with the following dose adjustment guideline for leukopenia: WBC > 3000 cells/μL, no change; WBC 2000–3000 cells/μL, decrease dose to 500-1000 mg/day; WBC < 2000 cells/μL with ANC > 1000 cells/μL, decrease dose by 50%; ANC < 1000 cells/uL, discontinue MMF until ANC > 1000 cells/μL.

Cytomegalovirus disease (CMV), defined as CMV tissue invasive disease as well as CMV infection/syndrome, was reported.

Randomization was 1:1 with stratification by race (African American or not) and donor type (living or deceased) to CSWD or CCS group. Data were analyzed on an intention-to-treat basis. Mean values for continuous variables were compared using a t-test. Statistical analyses were performed using SAS and two-tailed tests of significance were used in all instances. The level of significance of all statistical tests was a p value of 0.05.


A total of 397 patients were randomized with 195 CCS and 191 CSWD evaluable patients. No differences were observed with respect to patient demographics, immunologic risks, induction therapy and CMV serostatus (Table 1).

Table 1. Demographic, immunologic risk, induction therapy and stratification data*
 CCS (n = 195 patients)CSWD (n = 191 patients)
  1. *No significant difference between groups.

  2. n = number of patients at the time of randomization.

Age (mean ± SD, median)46.2 ± 12.7, 4646.6 ± 12.2, 48
Female gender (%)36.430.9
Race (% of patients)  
African American21.517.8
BMI (kg/m2) (mean ± SD, median)28.3 ± 14.3, 26.427.1 ± 5.1, 26.6
Donor source (%)  
Living related36.432.5
Living unrelated21.024.1
Deceased donor cold ischemic time (h; mean ± SD, median)17.2 ± 7.2, 17.518.4 ± 5.7, 18.0
HLA mismatches (mean, median)3.5, 3.03.5, 3.0
Current PRA (mean ± SD, median)1.8 ± 5.5, 01.6 ± 5.3, 0
CMV serology (%)  
D+/R+ or D-/R+81.680.1
Pretransplant diabetes mellitus (%)30.825.7
Induction therapy (%)  
IL-2R antagonist30.334.6

TAC dose and trough level

All patients

Despite similar mean TAC dose, significantly greater mean TAC trough level was observed in the CSWD group compared to the CCS group at week two posttransplant, which is 1 week after discontinuing corticosteroids (Figure 1A). Following discontinuation of corticosteroids, the subsequent mean TAC dose necessary to maintain similar trough levels was consistently less in the CSWD group compared to the CCS group for the remainder of the study until 5 years posttransplant; significant differences were observed from week four through week eight posttransplant (Figure 1B).

Figure 1.

Mean (±SD) tacrolimus trough level (A) and tacrolimus dose (B) in all patients randomized to receive corticosteroid (CCS) therapy or early steroid withdrawal (CSWD). *p < 0.05.

African American subgroup

When data analysis was performed separately in the African American subgroup, no significant differences in mean TAC dose and trough level were observed between the CSWD and CCS groups at any time points following corticosteroid withdrawal (Figures 2A and 2B). However, analysis of the non-African American subgroup demonstrated similar results as those observed in all patients (Figures 2C and 2D).

Figure 2.

Mean (±SD) tacrolimus trough level (A) and tacrolimus dose (B) in African American patients randomized to receive corticosteroid (CCS) therapy or early steroid withdrawal (CSWD). Mean tacrolimus trough level (C) and tacrolimus dose (D) in non-African American patients randomized to receive corticosteroid (CCS) therapy or early steroid withdrawal (CSWD). *p < 0.05.

Another important observation is that the African American subgroup consistently required significantly greater mean TAC dose compared to their non-African American counterpart to maintain similar mean trough level (Figure 2), regardless of whether they were randomized to CSWD or CCS group.

Additional subgroup analyses

Patients were also stratified based on pretransplant diabetes status, induction therapy and donor source and subgroup analyses were performed (Tables 2A and 2B). Results observed were similar to all patients where CSWD patients in each stratified group had a significantly greater mean TAC trough level at week two after transplant compared to their CCS counterparts even though the mean TAC dose was comparable between the two groups. Furthermore, the mean TAC dose remained consistently less in CSWD patients compared to CCS patients in each subgroup in most instances.

Table 2A. Mean tacrolimus trough level
Time posttransplantWeek 1Week 2Week 4Month 12Month 60
  1. *p < 0.05 (Gp. 1 vs. 2).

  2. †p < 0.05 (Gp. 3 vs. 4).

  3. €p < 0.05 (Gp. 1 vs. 3).

  4. ‡p < 0.05 (Gp. 2 vs. 4).

  5. All results mean ± SD; Values of n shown indicate the number of subjects who had a value at Week 1, subsequent time points may have fewer subjects; DM, diabetes mellitus; Thymo, Thymoglobulin; IL-2R, interleukin-2 receptor blocker; LD, living donor; DD, deceased donor.

Pretransplant diabetes
Gp. 1
DM CCS11.76 ± 4.6312.71 ± 4.98*11.32 ± 2.709.38 ± 2.246.91 ± 3.27
(n = 60)     
Gp. 2
DM CSWD12.18 ± 4.3016.29 ± 5.9411.13 ± 2.639.05 ± 2.288.3 ± 4.28
(n = 48)     
Gp. 3
Non-DM CCS10.47 ± 4.4112.76 ± 4.9611.88 ± 3.58.89 ± 2.157.5 ± 3.99
(n = 134)     
Gp. 4
Non-DM CSWD10.85 ± 4.3515.06 ± 5.6111.95 ± 3.418.97 ± 2.426.65 ± 2.24
(n = 142)     
Induction therapy
Gp. 1
Thymo CCS10.74 ± 4.6913.34 ± 5.08*11.70 ± 3.539.28 ± 2.227.79 ± 4.22
(n = 135)     
Gp. 2
Thymo CSWD11.42 ± 4.5216.46 ± 5.5712.24 ± 3.099.05 ± 2.407.38 ± 3.23
(n = 124)     
Gp. 3
IL-2R CCS11.18 ± 4.0911.40 ± 4.4111.74 ± 2.598.51 ± 2.026.31 ± 2.35
(n = 59)     
Gp. 4
IL-2R CSWD10.76 ± 4.0513.37 ± 5.4410.76 ± 3.338.88 ± 2.346.38 ± 2.01
(n = 66)     
Donor source
Gp. 1
LD CCS10.81 ± 4.3212.57 ± 5.17*11.91 ± 3.118.99 ± 2.086.64 ± 3.01
(n = 112)     
Gp. 2
LD CSWD11.59 ± 4.414.87 ± 5.6411.37 ± 3.369.16 ± 2.356.65 ± 2.75
(n = 108)     
Gp. 3
DD CCS10.96 ± 4.7812.99 ± 4.6611.45 ± 3.489.11 ± 2.318.18 ± 4.45
(n = 82)     
Gp. 4
DD CSWD10.66 ± 4.2916.06 ± 5.7612.23 ± 3.038.77 ± 2.417.48 ± 2.99
(n = 82)     
Table 2B. Mean tacrolimus dose
Time posttransplantWeek 1Week 2Week 4Month 12Month 60
  1. *p < 0.05 (Gp. 1 vs. 2).

  2. †p < 0.05 (Gp. 3 vs. 4).

  3. €p < 0.05 (Gp. 1 vs. 3).

  4. ‡p < 0.05 (Gp. 2 vs. 4).

  5. All results mean ± SD; Values of n shown indicate the number of subjects who had a value at Week 1, subsequent time points may have fewer subjects; DM, diabetes mellitus; Thymo, Thymoglobulin; IL-2R, interleukin-2 receptor blocker; LD, living donor; DD, deceased donor.

Pretransplant diabetes
Gp. 1
DM CCS0.11 ± 0.060.11 ± 0.060.11 ± 0.070.08 ± 0.050.05 ± 0.03
(n = 60)     
Gp. 2
DM CSWD0.12 ± 0.060.11 ± 0.060.10 ± 0.060.08 ± 0.050.05 ± 0.03
(n = 48)     
Gp. 3
Non-DM CCS0.14 ± 0.060.13 ± 0.070.14 ± 0.070.10 ± 0.060.06 ± 0.04
(n = 133)     
Gp. 4
Non-DM CSWD0.13 ± 0.080.12 ± 0.070.11 ± 0.070.08 ± 0.050.05 ± 0.03
(n = 142)     
Induction therapy
Gp. 1
Thymo CCS0.13 ± 0.060.14 ± 0.070.14 ± 0.08*0.09 ± 0.060.06 ± 0.04
(n = 134)     
Gp. 2
Thymo CSWD0.14 ± 0.070.13 ± 0.070.11 ± 0.070.08 ± 0.060.05 ± 0.03
(n = 124)     
Gp. 3
IL-2R CCS0.11 ± 0.060.11 ± 0.060.11 ± 0.070.08 ± 0.060.05 ± 0.04
(n = 59)     
Gp. 4
IL-2R CSWD0.11 ± 0.070.10 ± 0.070.10 ± 0.060.08 ± 0.040.05 ± 0.03
(n = 66)     
Donor source
Gp. 1
LD CCS0.13 ± 0.070.13 ± 0.07*0.12 ± 0.07*0.09 ± 0.050.05 ± 0.04
(n = 111)     
Gp. 2
LD CSWD0.12 ± 0.070.11 ± 0.070.10 ± 0.070.08 ± 0.050.05 ± 0.03
(n = 107)     
Gp. 3
DD CCS0.12 ± 0.060.13 ± 0.070.14 ± 0.080.10 ± 0.070.06 ± 0.04
(n = 82)     
Gp. 4
DD CSWD0.14 ± 0.070.13 ± 0.070.11 ± 0.060.08 ± 0.050.05 ± 0.03
(n = 83)     

A notable observation is that pre-transplant diabetics in either CSWD or CCS groups required lower mean TAC dose for the majority of time points throughout the study to maintain similar trough levels compared to patients without pretransplant diabetes. However, significant differences were observed only in the CCS group and not in the CSWD group (Table 2B).

Renal function and hyperkalemia

Mean serum creatinine (SCr; Figure 3) level was significantly greater and mean calculated creatinine clearance using the Cockcroft-Gault equation (data not shown) was significantly lower at weeks two, four and eight posttransplant in the CSWD group. In addition, mean serum potassium (SrK+) level was significantly elevated in the CSWD group at week two posttransplant and remained significantly elevated through month 48 posttransplant (Figure 4). Specifically, the incidence of SrK+ ≥ 6.0 mEq/L was significantly greater in the CSWD group during the first 3 months posttransplant (CSWD 22.5% vs. CCS 12.3%; p = 0.01).

Figure 3.

Mean (±SD) serum creatinine in all patients randomized to receive corticosteroid (CCS) therapy or early steroid withdrawal (CSWD). *p < 0.05.

Figure 4.

Mean (±SD) serum potassium in all patients randomized to receive corticosteroid (CCS) therapy or early steroid withdrawal (CSWD). *p < 0.05.

Because acute rejection may play a role in elevating SCr and SrK+ levels, we performed a subanalysis in patients who did not experience an acute rejection episode. We found that the mean SCr was consistently higher throughout the study period in the CSWD group compared to the CCS group, with values reaching statistical significance at weeks two, four, eight and months three and six. As for SrK+, a similar trend was observed with higher mean values recorded in the CSWD group compared to the CCS group throughout the study period with significance achieved at all time points from week two through month 60.

MMF dose and risk for acute rejection

Mean MMF dose was significantly lower in the CSWD group from week four until 3 years posttransplant compared to the CCS group (Figure 5). Gastrointestinal side effects was the reported cause of MMF dose reduction in a similar proportion of patients in both the CSWD and CCS groups (CSWD 29.8% vs. CCS 28.2%; p = 0.72) whereas leukopenia was the reason for MMF dose reduction in a significantly greater number of patients in the CSWD group (CSWD 52.4% vs. CCS 26.7%; p < 0.001). This is despite the fact that only 20% of the CSWD group and 17.3% of the CCS group had a WBC < 3000 cells/μL at the time of dose reduction, and therefore, only a small number of patients met criteria for MMF dose reduction per protocol. Although the overall mean WBC count in the CSWD group was significantly decreased following corticosteroid withdrawal in comparison to the CCS group (Figure 6) and the incidence of WBC < 3000 cells/μL was significantly greater in the CSWD group (CSWD 27.2% vs. CCS 14.9%; p = 0.004), no significant differences were observed in the incidence of more severe leukopenia with WBC < 2500 cells/μL or < 2000 cells/μL (CSWD 14.1% vs. CCS 9.2% for WBC < 2500 cells/μL, p = 0.154; CSWD 4.7% vs. CCS 5.1% for WBC < 2000 cells/μL, p = 1.000; Figure 7).

Figure 5.

Mean (±SD) MMF dose in all patients randomized to receive corticosteroid (CCS) therapy or early withdrawal (CSWD). *p < 0.05.

Figure 6.

Mean (±SD) WBC count in all patients randomized to receive corticosteroid (CCS) therapy or early steroid withdrawal (CSWD). *p < 0.05.

Figure 7.

Incidence of leukopenia defined as either WBC <3000 cells/μL or <2000 cells/μL, low absolute neutrophil count (ANC) defined as ANC <1000 cells/μL or <500 cells/μL and cytomegalovirus (CMV) disease in patients randomized to receive corticosteroid (CCS) therapy or early steroid withdrawal (CSWD).

With regards to neutropenia or CMV disease, which could both result in MMF dose reduction, the incidence of either was similar between both CSWD and CCS groups. The overall incidence of severe neutropenia with ANC < 500 cells/μL during the study in the CSWD and CCS groups were 5.2% and 3.6%, respectively (p = 0.459). The incidence of CMV disease in the CSWD and CCS groups were 7.3% and 10.3%, respectively (p = 0.37; Figure 7).


Although the interaction between TAC and corticosteroids has been previously reported, this is the first large prospective, multicenter, randomized, double-blind study to address this clinically significant pharmacokinetic interaction [8-12]. We demonstrated that corticosteroid withdrawal leads to a significant increase in the TAC trough level within 7 days of the discontinuation. This difference in trough level is only temporary as dose adjustments are made over time to maintain trough levels within range. However, the effect of corticosteroid withdrawal continues to be observed throughout the study as greater TAC doses were necessary in the CCS group to achieve similar trough levels. The interaction between TAC and corticosteroids is thought to be secondary to the induction of the CYP3A4 and PGP systems by corticosteroids in a dose-dependent manner. As a result, as the corticosteroid dose is tapered over time, the inducing effects that corticosteroids exert on the CYP3A4 and PGP systems diminish and lead to decreased clearance of TAC and therefore decreased TAC dose necessary to maintain similar trough level [11, 12]. Significant difference in mean TAC dose between the CSWD and CCS groups was observed until month three posttransplant, suggesting that prednisone dose 0.15 mg/kg or less does not exert a significant effect on the CYP3A4 and PGP enzyme systems.

When analyzed separately, the difference in mean TAC trough levels was not observed in African American patients nor was there any difference demonstrated with regards to the mean TAC doses between African American CSWD and CCS patients, suggesting a lack of effect of corticosteroid withdrawal in this patient population. Furthermore, it was observed that African American patients consistently required significantly greater TAC doses than non-African American patients to maintain comparable TAC trough levels, confirming findings of previously published studies [16-19]. It remains unclear as to why African American patients require greater TAC doses or do not display significant corticosteroid interactions. However, pharmacogenetic studies have demonstrated polymorphic differences in expression of cytochrome P450 3A4 and 3A5, as well as PGP bet-ween African American and non-African American kidney transplant recipients, resulting in altered TAC absorption and metabolism [20-23]. Our current findings shed light on the importance of considering differences in polymorphic expression in African American transplant recipients who are already at greater risk of rejection. Additional studies focused on the African American population are warranted to better understand this difference.

In addition, patients who were diabetics pretransplant required lower TAC doses than their nondiabetic counterparts to maintain similar trough levels. This observation is counterintuitive because the prevailing thought is that absorption is impaired in diabetic patients because of diabetic enteropathy. A possible explanation may be related to the higher incidence of nonalcoholic fatty liver disease, which may potentially affect the metabolism of TAC. Additional studies are warranted to better understand this interesting observation.

In our CSWD patients, a slower decline in SCr was observed following corticosteroid withdrawal between weeks two and eight posttransplant which coincided with the temporary increase in TAC trough levels. In comparison to CCS patients, mean SCr was significantly greater and calculated creatinine clearance was significantly less in the CSWD group for the first 8 weeks posttransplant. Moreover, mean SrK+ level was significantly higher throughout the majority of the study following discontinuation of corticosteroids. It is well known that renal dysfunction in association with hyperkalemia is suggestive of acute calcineurin inhibitor (CNI) nephrotoxic effects. This may suggest that patients who underwent corticosteroid withdrawal may have been subjected to greater renal CNI exposure despite similar trough levels achieved in both groups; however, formal pharmacokinetics are undoubtedly needed to confirm this hypothesis. Although mean SCr level was similar in both groups after eight weeks, it is known that SCr is an insensitive biomarker of kidney injury and may not reflect kidney histology in the absence of protocol biopsies. However, indication biopsies in this study did demonstrate a greater incidence of CAN changes in patients who underwent corticosteroid withdrawal [3]. Although this may be explained by a higher incidence of acute rejection in this patient population [3], chronic CNI nephrotoxicity is a known major contributor to the CAN histology [24]. Moreover, hyperkalemia was a universal finding in patients who were withdrawn from corticosteroids, which may represent a lack of mineralocorticoid activity or it can also be explained by a greater TAC exposure or effect on the distal tubules.

As expected, CSWD patients were on a lower dose of MMF because of the absence of corticosteroid-associated leukocytosis. Although the CSWD group had more patients overall with mild leukopenia, the incidence of severe leukopenia with WBC < 2500 cells/μL or < 2000 cells/μL and the incidence of clinically significant neutropenia (defined as ANC < 1000 cells/μL or < 500 cells/μL) were not different between the groups. Despite that, CSWD patients experienced more MMF dose reductions, mostly outside of protocol indications. Dose adjustments were, therefore, made prematurely in the majority of patients. Although it may be disconcerting to observe a decrease in WBC count and not react to it, providers must remember that reducing MMF dose may have clinical repercussions such as significantly increased incidence of acute rejection and decreased graft survival, especially if the dose reduction occurred early posttransplant [25-28]. Therefore, MMF dose reductions must be made only when necessary and it is crucial to routinely assess whether there is a continued need to maintain the dose reduction as an attempt should always be made to optimize the MMF dose as long as the clinical picture permits [28].

In summary, corticosteroid withdrawal protocols seem to profoundly affect TAC levels and dosing in addition to MMF dosing. As corticosteroid sparing protocols become more commonplace, proper knowledge of these interactions will allow the clinicians to better manage their patients. Future studies including formal pharmacokinetics, pharmacodynamics and pharmacogenetics are needed to better understand these numerous and complex interactions. This may allow tailoring of the immunosuppressive medication requirements for each individual or patient population to maximize benefits and minimize side effects.


This research has been supported by funding from Astellas Pharma Global Development, Inc.

Astellas Corticosteroid Withdrawal Study Group: Laurence Chan MD PhD, University of Colorado, Denver, Colorado; Mark Stegall MD, Mayo Clinic, Rochester, Minnesota; Brian Stevens MD, PhD, University of Nebraska, Omaha, Nebraska; Jonathan Bromberg MD, PhD, Mount Sinai Hospital, New York, New York; Okechukwu Ojogho MD, Loma Linda University; Loma Linda, California; Kenneth Washburn MD, University of Texas Health Science Center, San Antonio, Texas; Kristene Gugliuzza MD, U. of Texas Medical Branch, Galveston, Texas; Ravi Parasuraman MD, Henry Ford Hospital, Detroit, Michigan; Oleh Pankewycz MD, SUNY Buffalo, Buffalo, New York; Eugene Schweitzer MD, University of Maryland, Baltimore, Maryland; Willem Van der Werf MD, Latter Day Saints Hospital, Salt Lake City, Utah; Christopher Johnson MD, Medical College of Wisconsin, Milwaukee, Wisconsin; George Loss MD, PhD, Ochsner Clinic, New Orleans, Louisiana; George Francos MD, Thomas Jefferson University, Philadelphia, Pennsylvania; Paul Morrisey MD, Rhode Island Hospital, Providence, Rhode Island; Raphael Mendez MD, St. Vincent's Hospital, Los Angeles, California; David Shaffer MD, Vanderbilt University, Nashville, Tennessee; Sandip Kapur MD, Cornell University, New York, New York; Richard Thistlethwaite MD, PhD, University of Chicago, Chicago, IL; Richard Freeman MD, New England Medical Center, Boston, Massachusetts; David Laskow MD, R. W. Johnson Medical School, New Brunswick, New Jersey; Thomas Johnston MD, University of Kentucky, Lexington, Kentucky; Arthur Matas MD, University of Minnesota, Minneapolis, Minnesota; Don Hricik MD, University Hospitals of Cleveland, Cleveland, Ohio; S. Abul-Ezz MD, PhD, University of Arkansas, Little Rock, Arkansas; Rita Alloway PharmD, University of Cincinnati, Cincinnati, Ohio, Linda W. Moore RD, The Methodist Hospital Houston, TX; Nosratollah Nezakatgoo MD, University of Tennesse, Memphis; The EMMES Corporation–Paul Van Veldhuisen, PhD.


This study is a subanalysis of a multicenter study sponsored by Astellas Pharma US under mutually agreed contracts with each participating transplant center. Sponsors allowed authors complete access to all data.

The authors of this manuscript have conflicts of interest to disclose as described by the American Journal of Transplantation. The following authors are full-time employees of Astellas Pharma Global Development: M.R. First, W. Fitzsimmons, and J. Holman. R. Reisfield is a full-time employee of Astellas Scientific and Medical Affairs, Inc. A. O. Gaber received grants from Astellas Pharma US during the time in which the study was performed and the manuscript was prepared. E. S. Woodle and J.D. Pirsch received grants and honoraria from Astellas Pharma US during the time in which the study was performed and the manuscript was prepared.