Risk factors associated with efficacy outcomes in kidney transplantation: analysis of a contemporary cohort of patients from the A2309 trial.

Abstract

A multivariate analysis of risk factors for a composite endpoint of treated biopsy proven acute rejection (BPAR), graft loss, death, or loss to follow-up was undertaken in a cohort of 833 de novo kidney transplants from an international trial (A2309). Patients were randomized to everolimus (trough concentration 3-8 ng/mL or 6-12 ng/mL) with reduced cyclosporine or to mycophenolic acid (MPA) with standard cyclosporine. Cox proportional hazard modeling, incorporating a range of recipient, donor, and transplant variables, showed that treatment group (i.e., randomization to either everolimus 3-8 ng/mL or 6-12 ng/mL vs. MPA) showed no significant association with risk of the composite efficacy endpoint at either month 12 or month 24 (significance level 0.05). At month 12, Cox proportional hazard modeling showed that black race (hazard ratio (HR) 1.68; 95% confidence interval (CI) 1.08, 2.60; p=0.021), increasing donor age in years (HR 1.01; 95% CI 1.00, 1.03; p=0.022), and delayed graft function (DGF; yes vs. no, HR 2.75; 95% CI 1.82, 4.16; p< 0.001) predicted higher risk of the composite endpoint; female gender (female vs. male HR 0.67; 95% CI 0.48, 0.93; p=0.017), and < 3 HLA mismatches (HR 0.70; 95% CI 0.50, 0.99; p=0.049) were associated with reduced risk. At month 24, increasing recipient age in years (HR 0.99; 95% CI 0.98, 0.99; p=0.028), black recipient race (HR 1.62; 95% CI 1.09, 2.42; p=0.018), increasing donor age in years (HR 1.01; 95% CI 1.00, 1.02; p=0.008) and delayed graft function (DGF) (HR 2.60; 95% CI 1.78, 3.82; p<0.001) were predictive of risk. These findings show that, independently from type of immunosuppression, organ quality (expressed by DGF), donor age and recipient age, race and gender appear to be the main determinants of efficacy within 2 years after kidney transplantation.