RESUMO
A number of studies were published with contradictory results comparing tacrolimus (Tac) and cyclosporine A (CsA) for graft-versus-host disease (GVHD) prophylaxis, but there are only few that accounted for pharmacokinetic (PK) parameters. In this study, we created a model based on median concentrations, variability of concentrations, and failures to maintain target levels that distinguished patients with low, intermediate, and high risks of acute GVHD (hazard ratios (HR) 1.77, 95%CI 1.36-2.32, p < 0.0001). This model was used to compare 95 patients with CsA and 239 with Tac GVHD prophylaxis. In the multivariate analysis, incorporating PK risk, no differences were observed for grade II-IV acute GVHD (HR 0.73, 95%CI 0.48-1.10, p = 0.13), but grade III-IV acute GVHD was lower in the Tac group (HR 0.47, 95%CI 0.28-0.78, p = 0.004). The observed difference was due to patients with high PK risk (HR 0.377, 95%CI 0.19-0.75, p = 0.005), but not with low and intermediate PK risk (p > 0.05). Patients in the Tac group had better GVHD relapse-free survival (HR = 0.659, p = 0.01) and comparable overall survival (p > 0.05). In conclusion, PK risk should be accounted for in comparisons of GVHD prophylaxis regimens with calcineurin inhibitors, and Tac was superior to CsA in patients with high, but not intermediate and low PK risk.