Skip to main content
. 2023 Mar 23;108(9):2305–2315. doi: 10.3324/haematol.2022.281876

Figure 3.

Figure 3.

Five-year predicted probabilities of graf-versus-host disease and relapse/rejection-free survival (GRFS) and causes of GRFS failure in the upfront and relapsed or refractory cohorts. Predictions are provided by the Cox model shown in Table 2 and are given for some combinations of selected covariates: (A) age and timing of allogeneic hematopoietic stem cell transplantation (allo-HSCT) for the upfront cohort. Other covariates were arbitrarily set as follows: Cytomegalovirus (CMV): “other than seronegative for both donor and recipient (D-/R-)”; graft source: “bone marrow [BM]”; in vivo T-cell depletion: “yes”. (B) Age and donor type for the relapsed or refractory (rel/ref) cohort. Other covariates were arbitrarily set as follows: CMV: “other than D-/R-”; graft source: “BM”; timing of allo-HSCT: “>6 months”; total body irradiation (TBI): “yes”; in vivo T-cell depletion: “yes”. The complete tables of 5-year predicted probabilities for all covariate combinations are provided in the Online Supplementary Tables S3 and S4. MRD: matched related donor; GF: graft failure; allo: allogeneic; aGvHD: acute graft-versus-host disease; cGvHD: chronic GvHD; UD: unrelated donor.