Effect of long-term immunosuppression in kidney-graft recipients on cancer incidence: randomised comparison of two cyclosporin regimens

Background Long-term administration of cyclosporin carries a risk of renal toxicity, and immunosuppressants are associated with an increased rate of malignant disorders. We undertook an open randomised study of the risks and benefits of two long-term maintenance regimens of cyclosporin in kidney-allograft recipients. The primary endpoint was graft function; secondary endpoints were survival and occurrence of cancer and rejection. Methods 231 recipients of a first allograft with at most one previous rejection episode were randomised 1 year after transplantation. Most were receiving cyclosporin and azathioprine. One group received cyclosporin doses adjusted to yield trough blood concentrations of 75–125 ng/mL (low-dose group); the second received doses that yielded trough concentrations of 150–250 ng/mL (normal-dose group). Analysis was by intention to treat. Findings At 66 monthsʹ follow-up, the low-dose and normaldose groups were similar in mean serum creatinine (182 [SD 160] vs 184 [157] μmol/L; p=0•9) and mean creatinine clearance (47•5 [25•1] vs 45•3 (22•5] mL/min; p=0•6). Nine of 116 patients in the low-dose group and one of 115 in the normal-dose group had symptoms of rejection (p<0•02). There was no difference between the low-dose and normal-dose groups in survival (95 vs 92%; p=0•7) or graft survival (89 vs 82%; p=0•17) at 6 years. 60 patients developed cancers, 37 in the normal-dose group and 23 in the low-dose group (p<0•034); 66% were skin cancers (26 vs 17; p<0•05). Interpretation We found no evidence that halving of trough blood cyclosporin concentrations significantly changes graft function or graft survival. The low-dose regimen was associated with fewer malignant disorders but more frequent rejection. The design of long-term maintenance protocols for transplant recipients based on powerful immunosuppressant combinations should take these potential risks into account.