Graft-versus-tumor effect according to type of graft-versus-host disease defined by National Institutes of Health consensus criteria and associated outcomes.

The impact of National Institutes of Health consensus criteria (NCC) graft-versus-host disease (GVHD) on survival has rarely been investigated in a large cohort of patients with GVHD presenting before and after day 100 posttransplantation. We retrospectively investigated 775 patients who underwent allogeneic stem cell transplantation and assessed the GVHD effects on survival by the time-dependent covariates in Cox proportional hazards regression models. Using the NCC, the patients were classified into 4 groups: (1) no GVHD (n = 251); (2) acute GVHD (aGVHD) only (n = 199), including 26 patients with late aGVHD; (3) classic chronic GVHD (cGVHD; n = 232); and (4) overlap syndrome (OS; n = 93). Multivariate analyses showed that classic cGVHD (hazard ratio [HR], 0.46; 95% confidence interval [CI], 0.27-0.77) and OS (HR, 0.52; 95% CI, 0.28-0.96) were associated with significantly decreased risk of relapse, whereas aGVHD only was not associated with relapse rate (HR, 1.11; 95% CI, 0.76-1.63). All aGVHD events, including the period of aGVHD in patients who developed cGVHD after aGVHD, also did not affect the risk of relapse (HR, 0.74; 95% CI, 0.49-1.12). All types of GVHD were significantly associated with higher nonrelapse mortality in common. Finally, patients with aGVHD only had significantly lower overall survival and disease-free survival compared with those without GVHD, in contrast to favorable survival outcomes in patients with cGVHD without previous aGVHD. This study demonstrates that NCC GVHD type is associated with different graft-versus-tumor effects. Further studies are needed to investigate risk factors, pathogenesis, and biomarkers for each type of NCC GVHD.