Abstract
Despite improving immunosuppressive protocols in renal transplantation, chronic allograft nephropathy (CAN) remains a major impediment to long-term graft survival. The optimal immunosuppressive regimen for a patient with CAN is unknown. The aim of this study is to evaluate the various immunosuppressive management strategies of biopsy-proven CAN and of chronic allograft dysfunction (CAD) (no biopsy). A systematic review of randomized trials (n = 12 trials with 635 patients) was conducted. Studies included patients who were >6 mo post-transplant. All patients were on a calcineurin inhibitor (CNI), most often cyclosporine, and were randomized to convert to mycophenolate mofetil (MMF), tacrolimus, or sirolimus (Rapa) or to add azathioprine, MMF or Rapa to their current regimen. Follow-up time was 6 to 36 mo. The outcome measures evaluated were renal function in 11 of 12 studies and repeat renal biopsy results in one study. The methodological quality scores of the trials were generally low, using the Jadad scale (median value 2/5). Results varied between studies but suggested that CNI withdrawal is safe and that conversion to MMF or Rapa may be beneficial. The incidence of adverse effects ranged from 0% to 68% between the studies, and medication withdrawal occurred in 0% to 24% of patients. The review did not result in a consensus regarding the management of CAN and CAD. Further studies are required to determine the best therapeutic option for patients with CAD and CAN.
Despite improving immunosuppressive protocols in renal transplantation, chronic allograft nephropathy (CAN) remains a major impediment to long-term graft survival. CAN is a poorly understood condition characterized clinically by progressive renal dysfunction, associated with hypertension and proteinuria (1–3). Renal biopsies show characteristic but nonspecific histopathological changes in the vascular, glomerular and tubulointerstitial compartments of the kidney (4). Although the term CAN has been recently replaced by “interstitial fibrosis and tubular atrophy without evidence of any specific etiology” (4), we continue to use CAN in the remainder of this article, as this is the terminology used in the majority of the literature available to date. In patients receiving cyclosporine (CsA)-based immunosuppression, up to two thirds of allografts demonstrate features of CAN 5-yr after transplantation, and up to 60% of renal allografts have histopathology compatible with severe CAN 10 yr after transplantation (5). Protocol kidney biopsies in patients receiving tacrolimus (Tac)-based immunosuppression have shown an incidence ranging between 20.5% and 34.8% as early as 6 mo after renal transplantation (6). The etiology of CAN is multifactorial and involves both immune-dependent and immune-independent factors. Immune-dependent factors include cell-mediated immune responses and antibody-mediated responses, acute or chronic, to donor antigens. Important factors inciting immune responses are recipient-donor histoincompatibility and sensitization (pretransplant or post-transplant). Immune-independent factors include donor characteristics; cold ischemia time; cytomegalovirus infection; hypertension; dyslipidemia; and the use of calcineurin inhibitors (CNI), notably CsA (1–3,7,8), and more recently Tac (6). The interplay of these factors leads to cumulative and incremental damage that ultimately results in CAN (5). Chronic allograft dysfunction (CAD) is the term used when CAN is presumed to be the most likely cause of progressive renal graft dysfunction in the absence of renal graft biopsy. Several authors have examined strategies to prevent CAN and CAD, and others have evaluated protocols for delaying and/or reversing the process. Following techniques suggested by Moher et al. (9), we conducted a systematic review to summarize and evaluate the evidence on pharmacologic management of CAN and CAD.
Materials and Methods
Research Questions
The primary questions of this review were (1) What is the evidence for various pharmacologic management strategies of CAN and CAD? (2) What is the methodological quality of trials on this topic? and (3) What is the reported magnitude of benefit and risk associated with each treatment option?
Included Studies
Published full-text, English language, randomized controlled trials of management of CAN or CAD were included when there were 20 or more adults (≥18-yr of age) randomized at ≥ 6 mo post-transplant with follow-up time ≥ 6 mo. Pilot trials and abstracts from national and international nephrology conferences were not reviewed.
Search Strategies
An independent review of citations from MEDLINE (OVID 1996 to May 31, 2008) and EMBASE (OVID 1996 to May 31, 2008) was conducted by two investigators who independently reviewed citations and agreed that each article met inclusion criteria. The search terms included chronic allograft nephropathy, chronic allograft dysfunction, renal dysfunction, renal failure, interstitial fibrosis and tubular atrophy, renal transplant, randomized controlled trial, azathioprine, cyclosporine, tacrolimus, calcineurin inhibitor withdrawal, rapamycin and mycophenolate mofetil. Full text articles were retrieved for consideration of inclusion. Supplemental searches using the names of all authors in the included articles were performed.
Analysis
Randomized studies of CAN and of CAD were evaluated together. Methodological quality of randomized controlled trials was assessed using the Jadad scale, which measures blinding, randomization, withdrawals and dropouts (10). A maximum score of 5 represents the highest quality trial. Characteristics of included studies were extracted and compiled in tabular form. Outcomes were not pooled using meta-analysis because of heterogeneity among the included studies.
Results
More than 600 English language citations were screened from all sources. Studies were excluded when they did not meet any one of the inclusion criteria. Twelve randomized controlled trials that met the inclusion criteria were found (Table 1).
Management strategies of chronic allograft nephropathy or dysfunction: Randomized controlled trials
Methodologic Assessment
Methodologic quality scores of the studies were generally low, with Jadad scores ranging from 1 to 3 (Table 1). Both the average and median Jadad scores were 2/5. No study met all methodologic criteria, as none of the studies were blinded; the maximum score that can be allotted to a randomized controlled trial (RCT) that is not blinded is 3/5.
Patient Characteristics
Nine studies included patients with deteriorating renal function and biopsy-proven CAN, and three studies included patients with CAD (no biopsy) (11–13). Studies enrolled patients > 6 mo post-transplant. Sample size in randomized trials ranged from 21 to 122, with a total of 635 patients in 12 trials. Follow-up time ranged from 6 to 36 mo. Follow-up biopsies were done in two studies (14,15). There was considerable heterogeneity in baseline renal function within and among studies; for example, one study included patients with baseline creatinine clearance ranging from 18.7 to 88.2 ml/min (16). Exclusion criteria varied between studies. Two earlier studies did not mention exclusion criteria (17,18). Most studies excluded patients with an alternate cause of deteriorating renal function and pregnant patients. Several studies excluded patients with active infection or malignancy (11,12,16,19,20). Some studies also excluded patients who were diabetic (19), who began angiotensin converting enzyme inhibitor (ACEI) or angiotensin receptor blocker (ARB) within 3 mo before recruitment (16), or who had known medication allergies (11,19).
Before initiating the study, all patients were on a CNI-based immunosuppressive regimen, most often CsA. Of the 432 patients taking CsA, three studies had triple therapy with steroids and azathioprine (Aza) (18,19,21) and five studies had “CsA-based immunosuppressive therapy,” not otherwise detailed (12,13,16,17,20). The remaining 184 patients were taking a CNI, either CsA or Tac, as dual therapy with steroids (22), or as triple therapy with steroids and mycophenolate mofetil (MMF) (15) or steroids and either Aza or MMF (11).
Management Strategies
Study designs varied in terms of management strategies. Multiple interventions were evaluated in six of the included studies (14,15,17,18,20,22), and half of these studies compared a conversion to an addition strategy (15,20,22).
Conversion Studies.
Eight studies evaluated the conversion from a CNI to another immunosuppressive regimen. Patients were randomized to convert from CNI to MMF in four studies (12,18,20,22), from CsA to Tac in four studies (16,18 to 20) and from a CNI to Sirolimus (Rapa) in two studies (11,15).
Addition Studies.
Seven studies evaluated adding an immunosuppressive medication, such as Aza (14,17), MMF (13–15,17,20,22), or Rapa (21), to a CNI-based regimen. CNI doses were reduced in both treatment and control groups in three studies (14,17,21), in only the control group in three studies (15,16,20), and in the treatment group in one study (13). When CNI dose was reduced, it was usually by 40%, but it was halved in one study (13) and decreased by 24% in another study (20). In addition, when new immunosuppressive medications were added, doses were not consistent from one study to the next. For example, in studies where MMF was added, doses ranged from 1 g per day (15) to 2 g per day (17). When Rapa was added, weight-based loading (0.1 mg/kg) and daily doses (0.04 to 0.06 mg/kg per d) aiming for a trough level of 6 to 10 ng/ml were chosen in one study (15), whereas other studies introduced an 8-mg loading dose and 4 mg/d thereafter (11), or a 6-mg loading dose followed by 2 mg/d maintenance dose, aiming for a trough level of 5 to 15 ng/ml (15,21).
Outcomes
Outcome measures were heterogeneous. All studies used renal function as the primary outcome except for one study which used graft survival and pathologic evaluation of repeat graft biopsy at 6 mo (15). Measures of renal function were radio-isotope studies in five studies (11,14,18,20,21) or absolute serum creatinine or related value, such as the slope of reciprocal of the serum creatinine versus time, in the remaining seven trials.
Outcomes by Management Strategy
Findings varied, depending particularly on the management strategy used, but also varied between studies using a similar therapy.
Conversion Studies
CNI to MMF.
In the four studies where CNI was converted to MMF (12,18,20,22), renal function improved in the MMF group compared with the control group. One study was stopped early due to ethical considerations (22).
CsA to Tac.
In the four studies where CsA was converted to Tac (16,18–20), there was no significant improvement in renal function except in one study (19), where 46 patients at a single center on CsA, Aza, and steroids were randomized to continue the same regimen or were converted to Tac. Of the 24 patients who converted to Tac, compared with controls, serum creatinine improved significantly at 12 and 36 mo (254 ± 55 versus 333 ± 98 μmol/L at 12 mo [P = 0.011] and 255 ± 78 versus 317 ± 89 μmol/L at 36 mo [P = 0.048]).
CNI to Rapa.
In two studies where CNI was converted to Rapa, results were positive. Renal function improved in the Rapa group by 12.9 ml/min (95% CI, 6.1 to 19.7, P < 0.001), in one study of 40 patients (11). In a study of 84 patients, comparing conversion from CNI to Rapa versus a 40% decrease in CNI dose in combination with MMF, graft survival was better in the Rapa group (1/33 versus 8/42 graft losses, P = 0.0376), and CAN grading (histologic lesions of interstitial fibrosis and tubular atrophy) on repeat 6-mo biopsy was stable in the Rapa group but worsened in the CNI and MMF group (15).
Addition Studies
Addition of Aza Versus Addition of MMF.
Two studies published in 2001 evaluated adding Aza (14,17) versus adding MMF to a CsA-based regimen. Glomerular filtration rates (GFRs) improved in both the Aza and MMF groups.
Addition of MMF.
In a study of 106 patients with CAD, renal function improved when MMF was added to a CNI-based regimen compared with CNI continuation (13). When conversion to MMF was compared with adding MMF to a CNI-based regimen, renal function improved significantly after CNI discontinuation (P = 0.002), and the study was stopped early (22). When addition of MMF with CNI dose reduction was compared with conversion to Rapa, graft survival was better in the Rapa group (15), as mentioned above.
Addition of Rapa.
When Rapa was added to CsA-based regimens with 40% CsA dose reduction compared with 40% CsA dose reduction alone, isotope GFR decreased in the Rapa and CsA group (from 32 ± 9 to 27 ± 9 ml/min, P < 0.01) compared with controls (from 29 ± 10 to 27 ± 8 ml/min, P = NS) (21).
Acute Rejection
Acute rejection causing graft loss occurred in only two patients, in a study randomizing 34 patients 12 mo post-transplant to convert to Tac versus remaining on CsA (16). Rates of graft loss varied from zero in most studies to eight events per 42 patients (19%) in one study in which patients were converted from CNI to Rapa, compared with one event per 33 patients (3%) in the control group (15).
Adverse Effects
All studies reported adverse effects, present in 1% to 25% in cohorts that converted CsA to Tac (16,20), in up to 50% of patients that took MMF (11,13,20), and in up to 68% of patients on Rapa (11). Two studies reported only symptoms (14,17), whereas other studies also reported hematologic and metabolic changes (11–13,15,16,18–22). Side effects were occasionally reversible after dose reduction. MMF was stopped in up to 24% of patients because of persistent gastrointestinal symptoms (13). Other frequent adverse reactions attributed to MMF were anemia and/or leucopenia, insomnia, and infections (14,15,17,18,20,22). Rapa was stopped in up to 10.5% (2/19) of patients because of adverse dermatologic reactions (11). Rapa was also associated with gum hypertrophy, headache, worsened dyslipidemia, bone marrow suppression, gastrointestinal disturbances, and infection (11,15,21).
Discussion
The most important finding of this systematic review is that there is no consensus regarding the management of CAN and CAD. Twelve RCTs were reviewed but could not be easily compared because of variations in time post-transplant, study design, follow-up time, and outcome measures (Table 1). CNI withdrawal after 6 mo seems to be safe, and conversion to MMF or to Rapa appear to be beneficial for some patients. Both medications are associated with adverse effects that caused medication discontinuation in up to 24% of patients taking MMF and in up to 10.5% of those starting Rapa (11,13). Thus, although MMF and Rapa may be reasonable alternatives to CNI, their adverse effect profiles may limit their use.
Therapeutic drug monitoring (TDM) has been suggested as a means to decrease adverse reactions while maintaining efficacy. TDM of MMF with measurement of mycophenolic acid trough levels has not been shown to predict clinical events, such as acute rejection or side effects (23), but TDM of Rapa using trough levels was predictive of the occurrence of acute rejection (<5 ng/ml) and of adverse reactions (>15 ng/ml) in a study of 150 renal transplant recipients (24). Adverse reactions occurred in all studies in this review, in spite of TDM. Dose reduction is another option to decrease adverse reactions. MMF dose reduction, however, was associated with increased acute rejection rates in a retrospective cohort study of 213 renal transplant recipient (25). Combination therapy using Rapa and a CNI has been attempted; however, registry data reported that this combination is associated with decreased renal transplant graft survival in de novo renal transplant patients (26,27). A systematic review of six RCTs evaluating early CNI withdrawal from a Rapa-based regimen found a higher creatinine clearance at 1 yr but an increased risk of acute rejections (28). A systematic review of 30 trials (five RCTs and 25 nonrandomized trials) evaluating early CNI conversion to Rapa found improved short-term creatinine clearance compared with controls in five RCTs, and the creatinine clearance improved or stabilized in 66% of the patients in non-RCTs. Rapa was discontinued by 28% of patients in the randomized trials and 17% in the nonrandomized trials, largely as a result of adverse effects (29). Clearly, the optimal immunosuppressive regimen for a patient with established CAN or CAD remains enigmatic, and parallel efforts should be directed to develop regimens that prevent CAN/CAD. Potential preventative strategies include CNI avoidance with either daclizumab (an anti-CD25 antibody; 30,31), belatacept (a selective blocker of the costimulatory pathway; 32), or Rapa (33,34), the use of low-dose CNI in de novo transplant recipients (35,36), and early CNI discontinuation (37–39). Long-term follow-up with protocol biopsies are needed to confirm the clinical benefits of CNI avoidance, low-dose CNI, and early conversion strategies.
This is the first systematic review to summarize and evaluate the evidence with respect to immunosuppressive management of CAN and CAD. There are several limitations of our review. Although we included a reasonable number of studies, the methodological quality of included studies is generally low, limiting the internal validity of the results. Sample sizes are small, limiting the generalizability of the results. We were not able to pool the data, given its substantial heterogeneity. We reviewed RCTs and did not include pilot studies or abstracts because we did not feel that they contribute substantially to evaluating the evidence.
In conclusion, CAN remains a major impediment to long-term graft survival. On the basis of this review, it appears that CNI withdrawal is safe and that MMF and Rapa may be useful alternatives to CNI-based therapy. However, both medications are associated with important adverse effects, and there is no consensus regarding the management of CAN/CAD. Future directions should include the study of newer agents and combinations of immunosuppressive medications. Larger studies with rigorous design, longer follow-up, and consistent outcome measures are needed to determine an evidence-based approach for the prevention and treatment of CAN/CAD.
Disclosures
None.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
- Copyright © 2009 by the American Society of Nephrology
References
In this issue
