Sirolimus Delays Renal Allograft Recovery

Introduction

Sirolimus use in the setting of delayed graft function (DGF) is being encouraged because it is not perceived to be nephrotoxic (1–3). Recent papers by Hong and Kahan and Chang et al. (4, 5) have, in fact, suggested that sirolimus may be the drug of choice for post-transplant immunosuppression in renal allograft recipients with DGF. Prior to this recommendation, immunosuppressive management of DGF often included the administration of an antilymphocyte preparation, mycophenolate mofetil or azathioprine and steroids (6–9). More recently, IL-2 receptor blockers have been used in low-risk patients in place of antilymphocyte preparations, to limit toxic side-effects as well as the risks of over-immunosuppression (7, 10, 11). Calcineurin inhibitors have long been avoided or used in decreased doses in patients at high risk for DGF because of the increased probability of added renal injury and prolongation of renal recovery (6–9, 11).

Contrary to recent recommendations that sirolimus should be a mainstay of immunosuppressive therapy in patients with DGF, McTaggart et al., in this issue of American Journal of Transplantation, advise instead that ‘it may not be the optimal immunosuppressive choice in the DGF setting’ (12). The authors provide evidence that sirolimus prolongs the duration of DGF. These investigators retrospectively evaluated 132 patients with DGF transplanted between 1/1/97 and 6/30/01 for risk factors associated with the time to renal recovery. Immunosuppressive medications including sirolimus, donor characteristics such as age and cause of death, peri-operative factors such as cold ischemia time, and recipient characteristics such as presensitization were evaluated for an association with duration of DGF. Sirolimus exposure was found to be the most important factor associated with prolongation of DGF (hazards ratio 0.49, p = 0.0009). This finding is even more remarkable if one considers that the authors of this report come from the same center that had previously advised the use of sirolimus in the setting of DGF, and thus represents a departure from previous conclusions (5, 12).

The findings of McTaggart et al. are congruent with the findings of Smith et al., who have recently reported an increased risk of developing DGF in patients receiving sirolimus on the day of transplant compared to those not receiving the drug (p = 0.02) (13). Multivariate analysis revealed that the development of DGF was very significantly associated with an increasing dose of sirolimus (OR 1.13 per additional mg of sirolimus, p = 0.004). Calcineurin inhibitor use was not responsible for the development of DGF in this study, as tacrolimus administration was delayed until allograft function was established or, in those with DGF, after 14 days from surgery. Furthermore, although not statistically significant, there was a trend toward increased duration of DGF in patients receiving sirolimus. Most notably, an increased number of patients requiring dialysis up to and beyond 60 days from transplant was found. Additionally, the discontinuance of sirolimus and tacrolimus under the cover of Thymoglobulin, mycophenolate mofetil and prednisone resulted in the return of renal function. The DGF rate, which had increased from 8 to 25% with the peri-operative use of sirolimus, fell back to 11% after delaying the administration of sirolimus as well as the calcineurin inhibitor. Finally, Groth et al., in 1999, in the seminal European randomized trial of sirolimus use, reported that sirolimus was associated with an increased rate of DGF (17%) (although not statistically significant) compared to cyclosporine (10%)-based therapy (1).

Animal studies predicted this clinical outcome. Andoh et al. reported in a rat model of renal ischemia that sirolimus impaired recovery compared to vehicle or mycophenolate mofetil (14). Seven days after ischemic injury, sirolimus-treated animals manifested decreased glomerular filtration rate, cc/min (GFR) and renal blood flow, while showing more pronounced interstitial fibrosis than animals treated with vehicle or mycophenolate mofetil. Lieberthal et al. also reported on the effect of sirolimus on recovery from renal ischemia induced by renal artery clamping in rats (15). Animals given sirolimus had a marked prolongation of the time to recovery (by GFR and fractional excretion of sodium measurements), prolonged presence of a high proportion of necrotic tubules after injury, increased tubular apoptosis and reduced tubular epithelial cell proliferation (by proliferating cell nuclear antigen staining) compared to animals receiving vehicle. Sirolimus treatment also decreased the activity of p70S6 kinase in ischemic kidneys compared to vehicle-treated controls. Furthermore, in vitro studies of mouse proximal tubular cells by the same authors showed that growth-factor-induced proliferation was inhibited by sirolimus, as was the activation of p70S6 kinase, while apoptosis was enhanced. p70S6 kinase is an enzyme critical to cell cycle progression.

Stepping back and looking at sirolimus from a physiologic perspective could have predicted the results of McTaggart, Smith, Andoh and Lieberthal, if the machinery inhibited by sirolimus is present in the kidney. Indeed, it has been shown that the cellular pathways affected by sirolimus, denoted by FK binding protein and target of rapamycin (TOR, a kinase), are present in renal epithelium (13). and p70S6 kinase is activated in the setting of renal injury and is inhibited by sirolimus (15). Sirolimus complexes with FKBP, and this complex inhibits growth-factor-stimulated TOR activity. TOR helps control the activation of proteins that modulate translation of mRNAs encoding regulators of the cell cycle. Such regulators include binding protein for eukaryotic translation initiation factor 4F and p70S6 kinase. These factors help to facilitate the assembly of ribosomes and the translation of selected mRNAs important to initiate cellular proliferation. Signals activating these pathways stimulate continued cell survival and proliferation. Inhibition of these pathways results in apoptosis and decreased cellular proliferation. Apoptosis is not directly promoted by inhibition of this pathway, but instead occurs due to inhibition of growth factor ability to promote cell survival, thereby sending cells to the default pathway of apoptosis. In addition to influences on proliferation and apoptosis, Andoh et al. have shown direct tubular injury by sirolimus (16). These investigators administered cyclosporine, tacrolimus and sirolimus independently to animals on a low salt diet for 14 days. Although renal function and blood flow was not diminished during the study period in sirolimus-treated animals, histologic analysis showed an increase in tubular dilation, tubular atrophy, tubular cast formation and thickening of the tubular basement membrane, comparable to that seen in calcineurin-treated animals. The possible impact (or not) of direct injury on long-term renal function in human renal and extrarenal allografts remains to be determined.

Compilation of the information provided by McTaggart, Smith, Andoh and Lieberthal leads one to consider avoiding agents that promote injury or delay cellular recovery in the peri-transplant period or at least until signs of recovery are demonstrated. Suggestions for peri-operative management from an injury/repair perspective could include the following. In patients at low risk for the development of DGF, an IL2R blocker with MMF and prednisone could be considered as first-line agents with sirolimus and/or calcineurin inhibitor use initiated once renal function is apparent (11). Anti-lymphocyte therapy or IL2R blockers along with mycophenolate mofetil and steroids could be considered as initial agents in high-risk recipients with sirolimus and/or a calcineurin inhibitor started once renal recovery is clearly established (e.g. creatinine less than 3 mg/dL) (6, 9, 11, 17). Alternatively, no matter what the risk for DGF, calcineurin inhibitors or sirolimus could be initiated soon after transplantation at very low doses (11). Sirolimus and/or calcineurin inhibitors should probably not be given before transplant unless there is a low likelihood of DGF (12, 13). In patients with DGF, delayed administration or markedly reduced doses of sirolimus or calcineurin inhibitors may assist in the renal recovery process.

In summary, delayed administration of calcineurin inhibitors and sirolimus should be considered in allografts with an increased risk for DGF, preferably delayed until renal allograft function is established. New non-nephrotoxic agents that do not prevent allograft cellular proliferation are awaited for use in the setting of DGF. and investigations targeting the physiology behind why some allografts appear to be more sensitive to the antiproliferative/pro-apoptotic effects of sirolimus are needed.