Long-term cancer risk of immunosuppressive regimens after kidney transplantation

Abstract

Cancer is a widely recognized complication of transplantation, and the effects of various immunosuppressive drugs on cancer risk remains controversial. This randomized trial allocated 489 recipients of first cadaveric renal transplants to one of three groups: Azathioprine and prednisolone, cyclosporine monotherapy, or cyclosporine monotherapy followed by a switch to azathioprine and prednisolone after 3 months. Here, we report cancer outcomes by non-skin cancer (including melanoma) and skin cancer (excluding melanoma) for 481 patients during a median follow-up of 20.6 years. A total of 226 patients developed at least one cancer: 95 with non-skin cancer and 171 with skin cancer. In the intention-to-treat analysis, mean times to first non-skin cancer (16.0, 15.3, and 15.7 years for groups 1 through 3, respectively) and first skin cancer (13.6, 14.3, and 15.2 years, respectively) were not different among the three groups or between any subgroup. In multivariate analyses, non-skin cancer associated with increasing age and previous smoking history, whereas skin cancer associated with increasing age, nonbrown eye color, fairer skin, and a functioning transplant. Treatment allocation did not associate with development of either form of cancer in multivariate analyses. In conclusion, these immunosuppressive regimens, widely used in recent decades, carry similar risks for carcinogenicity after kidney transplantation.

Figure 1.

Cumulative incidence of cancer out to 20 years posttransplant. (A) Cumulative incidence of NSC or melanoma, by treatment allocation, over time since randomization. (B) Cumulative incidence of SCEM, by treatment allocation, over time since randomization.

Figure 2.

Treatment effects upon cancer risk by potential effect modifiers. (A) Effects of randomized study of immunosuppressive treatments on NSC, stratified by potential effect modifiers. *Mismatch is the number of HLA mismatches between donor and recipient (grouped as shown); PRA is the panel-reactive antibody values for recipients (grouped as shown). (B) Effects of randomized study of immunosuppressive treatments on SCEM, stratified by potential effect modifiers. *Mismatch is the number of HLA mismatches between donor and recipient (grouped as shown); PRA is the panel-reactive antibody values for recipients (grouped as shown).

Figure 3.

Risk factors for NSA and SCEM by multivariate modeling.