Protocol renal allograft biopsies and the design of clinical trials aimed to prevent or treat chronic allograft nephropathy.

Abstract

The minimum sample size to perform a clinical trial aimed to modify the natural history of chronic allograft nephropathy (CAN) is very large. Since the presence of chronic tubulointerstitial damage in renal protocol biopsy specimens is an independent predictor of late outcome, we evaluated whether protocol biopsies could facilitate the design of trials aimed to prevent or treat CAN. Two hundred eighty-two protocol biopsy specimens were obtained 3 months after transplantation in 280 patients with serum creatinine levels <300 micromol/L, proteinuria <1000 mg/day, and stable function. The specimens were evaluated according to the Banff criteria. Graft survival depended on the presence of CAN and renal transplant vasculopathy (RTV). Thus, biopsy specimens were classified as: (a) no CAN (n=174); (b) CAN without RTV (n=87); and (c) CAN with RTV (n=21). Graft survival at 10 years was 95%, 82%, and 41%, respectively (P=0.001). Total serum cholesterol before transplantation was 4.5+/-1.1, 4.6+/-1.1, and 5.3+/-1.6 mmol/L, respectively (P=0.009) and it was the only predictor of RTV. Power analysis (beta=20%, alpha=5%) was done to evaluate whether protocol biopsies can facilitate the design of clinical trials aimed either to prevent or treat CAN. We showed that the most feasible approach would be to use the presence of CAN as the primary efficacy end point in a prevention trial. To demonstrate a 50% reduction in the incidence of CAN at 3 months, 570 patients would be required. Protocol biopsies may allow a reduction of sample size and especially the time of follow-up in a trial aimed to prevent CAN.