Reduced Risk of Residual Molecular and Hematological Relapse in Patients with Cml after KIRir-Ligand Mismatched Hematpoietic Stem Cell Transplantation.

We compared the incidence of molecular and hematological relapse in 236 CML patients (pts) after HLA-identical (n=158) [group 1], HLA class I antigen mismatched and KIR-ligand compatible (n=49) [group 2], and HLA class I antigen mismatched and KIR-ligand incompatible (n=29) [group 3] hematopoietic stem cell transplantation. In group 1 133/158 (84%) pts were in first chronic phase of CML, the corresponding figures were 33/49 (67%) pts in group 2, and 19/29 (64%) in group 3 (p<0.05). Accordingly, 25/158 (16%) pts in group 1, 16/49 (33%) pts in group 2, and 10/29 (34%) pts in group 3 were in advanced disease phases of CML (accelerated phase, 2^nd chronic phase, or blastic phase).

Stem cell grafts were from sibling donors in 97% of pts in group 1, 37% of pts in group 2, and in 59% of pts in group 3 (p<0.01 group 1 versus groups 2+3). Median age and gender constellations were not different between the three groups (age medians between 37–41 years, range 16–63).

Acute GVHD grade 2–4 occurred in 62/158 (39%) pts in group 1, in 24/49 (49%) pts in group 2, and in 15/29 (52%) pts in group 3 (NS). Estimates of chronic GVHD did not differ significantly between the study groups with 94% + 2% in group 1, 95% + 4% in group 2, and 91% ± 6% in group 3.

Molecular relapse as detected by real-time RT-PCR occurred in 1/29 (3%) pts of group 3 compared to 62/158 (39%) of pts in group 1, and in 11/49 (22%) pts in group 2 (p<0.001). A hematological relapse developed in 20/158 (13%) pts in group 1, in 2/49 (4%) pts in group 2 and in 0/29 (0%) pts in group 3 (p<0.05). Multivariate analysis on the risk molecular relapse included possible influencial covariates like disease stage, age (> 40 years or < 40 years), gender matching of donor and recipient, occurrence of acute and chronic GVHD, HLA matching between donor and recipient, donor source [sibling or unrelated], and graft type [bone marrow or PBSCs]. This analysis confirmed that KIR-mismatches are a strong independent predictor for the occurrence of bcr-abl transcripts after transplantation (p<0.02). The 8-year overall survival estimates were not different between the three groups (67% group 1, 52% group 2 66% group 3). In conclusion, KIR-ligand incompatibility is an important prognostic factor for the occurrence of molecular and hematological relapse after allogeneic transplantation for CML.