Search

Killer Ig-like receptors (KIR) and HLA class I ligands were studied in unrelated hemopoietic stem cell transplantation for chronic myeloid leukemia (n = 108). Significantly improved overall survival was observed in patients, which were homozygous for HLA-C-encoded group 1 (C1) ligands compared with those with group 2 (C2) ligands. Favorable outcome in the former patient group was an early effect that was highly significant in patients transplanted with G-CSF-mobilized peripheral blood and patients with advanced disease stages. In contrast, presence of C1 ligands in the donor was associated with significantly reduced patient survival. The differential roles of the two HLA-C ligands are explained in the context of a biased NK cell reconstitution, which is generally dominated by the presence of C1- but absence of C2-specific NK cells. The clinical observations are corroborated by in vitro experiments showing that NK cells derived from hemopoietic progenitor cells generally acquire the C1-specific inhibitory KIR2DL2/3 at earlier time points and with higher frequency than the C2-specific KIR2DL1. These findings define a novel determinant for understanding the role of NK cells in clinical hemopoietic stem cell transplantation.

Objective: In the hematopoietic stem cell transplantation setting, granulocyte colony-stimulating factor (G-CSF) administration can reduce donor cell reactivity in vitro, but the clinical significance of this phenomenon was only sparsely defined., Methods: We performed lymphocyte transformation tests in 28 related stem cell donors pre and 5 days post G-CSF treatment, respectively, and correlated proliferative responses of donor peripheral blood mononuclear cells with clinical parameters in the corresponding recipients., Results: In vitro reactions towards 4 mitogens and 12 recall antigens at day 5 post G-CSF administration were predictive for the occurrence of chronic graft-vs-host disease (cGVHD). Here, proliferative responses towards the mitogen anti-CD3 monoclonal antibody (OKT3) above median were most informative; this threshold could be determined by discrimination and receiver operating curve (ROC) analyses. In the whole cohort (18 human leukocyte antigen [HLA]-identical and 10 partially mismatched donor-recipient pairs), OKT3 responses predicted cGVHD with an odds ratio of 33.0, a sensitivity of 79%, and a specificity of 90%. A subgroup analysis of HLA-identical pairs even yielded an odds ratio of 85.0. Furthermore, bivariate analysis defined HLA compatibility and responses towards OKT3 as independent risk factors for cGVHD (p = 0.02 and p = 0.0007, respectively)., Conclusion: The proliferative capacity of G-CSF-mobilized donor cells appears as a graft factor that determines the future incidence of cGVHD in the corresponding recipient.

It remains controversial whether alloreactive donor-derived natural killer (NK) cells display graft-versus-leukemia reactions after unmodified allogeneic hematopoietic stem cell transplantation (HSCT). The present study evaluated the role of inhibitory killer immunoglobulin-like receptor (KIR) ligand incompatibility using a well-defined and uniform setting of unmodified allogeneic HSCT in 374 patients with myeloid leukemias. The most striking finding was a significant heterogeneity in the 5-year estimates of hematologic leukemic relapse after human leukocyte antigen (HLA)-identical (n = 237; 22%), HLA class I-disparate (n = 89; 18%), and KIR ligand-incompatible transplantations (n = 48; 5%) (P < .04). Multivariate analysis confirmed that the relative relapse risk (RR) was influenced by HLA class I disparity alone (RR 0.49), but was lowest after HLA class I-disparate, KIR ligand-incompatible transplantations (RR 0.24) (P < .008). The primary graft failure rates, however, increased from 0.4% after HLA class I-identical to 2.3% after HLA class I-disparate, and to 6.3% after KIR ligand-incompatible transplantations, respectively (P < .02). Unlike some other reports, no beneficial effect of KIR ligand incompatibility on other major endpoints of allogeneic HSCT (transplantation-related mortality, and overall and event-free survival) was detectable in the present study. In conclusion, unmodified allogeneic HSCT from KIR ligand-incompatible donors provides a superior long-term antileukemic efficacy in patients with myeloid malignancies.

Despite human leukocyte antigen (HLA) identity between donor and recipient, several patients develop acute graft-versus-host disease (aGVHD) after hematopoetic stem cell transplantation (HSCT) because of minor histocompatibility antigen (mHag) incompatibilities. The impact of multiple mHag disparities on the clinical outcome after HSCT still remains to be determined. We studied the genomic polymorphisms of HA-1, CD31, and CD49b and correlated mHag distribution with the occurrence of aGVHD after HSCT from HLA-matched sibling and unrelated donors. All 163 patients examined in our single-center study underwent HSCT for chronic myeloid leukemia in the first chronic phase. HA-1 and CD31 disparities are associated with increased aGVHD incidence in a subgroup of patients who test HLA-B44 supertype positive in univariate analysis. However, in a multivariate analysis, only increased patient age was confirmed as an independent aGVHD risk factor. Our findings indicate that the impact of mHag disparity on aGVHD development in HSCT from HLA-matched sibling and unrelated donors seems to be subordinated to classic aGVHD risk factors.

Allogeneic hematopoietic stem cell transplantation (HSCT) is a proven curative therapy for many hematologic malignancies. HSCT from HLA-identical sibling donors (ISDs) is still the golden standard. For the remaining 70% of the patients lacking an ISD, alternative (partially) HLA-matched family donors (MFDs) and HLA-matched unrelated donors (MUDs) are now widely accepted. However, it is presently unclear whether outcome after HSCT from an MFD or an MUD is superior. Thus, the classical clinical end points after HSCT from an ISD (n = 138), MFD (n = 86), and MUD (n = 101) were compared by means of univariate and multivariate statistical analyses. MFD transplantations with HLA class II (DRB1 +/- DQB1) mismatches in graft-versus-host (GVH) direction showed an increased risk of grades II to IV graft-versus-host disease, and MFD transplantations with more than a single HLA class I (A +/- B +/- C) mismatch in host-versus-graft (HVG) direction were associated with a higher risk of graft failure. However, no significant difference in overall survival was detectable among the 3 study groups after adjustment for the main predictors of transplantation outcome. Thus, for patients lacking an ISD, an already identified MFD with an HLA-DRB1 +/- DQB1 mismatch in GVH or a combined HLA-A +/- B +/- C mismatch in HVG direction should be accepted only in clinically urgent settings that leave no time to identify an MUD.

Background: Previous data indicate that a transfer of specific humoral and cellular immunity by way of allogeneic hematopoietic cell transplantation (HCT) should, in principle, be possible., Methods: In the HCT setting with a follow-up of up to 55 months, we studied the transfer of hepatitis B virus (HBV) specific immunity from electively immunized donors into HLA compatible recipients suffering from chronic myeloid leukemia (CML). After excluding preexisting HBV specific immunity in donor-recipient pairs, 27 prospective donors were vaccinated against HBV. In addition, on an average of 22 months postHCT, 8 of the 19 recipients were immunized once for HBV., Results: Donor vaccination resulted in detectable hepatitis B surface (HBs) antibodies in 85% of donors and specific cellular in vitro responses in 77%. Two weeks postHCT, 86 and 67% of the recipients displayed positive humoral and cellular HBV reactions, respectively, which then decreased. Afterwards, HBV immunity reappeared in 83% of the recipients without revaccination. Following a single vaccination in recipients, seven of eight displayed a typical memory response. An HBV specific response was already detectable 1 week after vaccination, approximately 1,300-fold (humoral) and 60-fold (cellular) higher than observed in the corresponding donors after a single immunization., Conclusions: The "spontaneous" recurrence of HBV immunity and the memory response in recipients give evidence for an elective immune transfer (e.g., for viral antigens) by way of allogeneic HCT.

Background: Evaluation of patient sera for complement-fixing anti-donor antibodies (serum crossmatch [XM]) before allogeneic blood stem cell transplantation (BSCT) is routine in most centers. However, in contrast to kidney transplantation, the predictive value of a positive XM for outcome of BSCT is still unclear, and a positive XM is presently not regarded as an absolute contraindication to proceed to transplant., Methods: To clarify the role of a positive XM as predictor for overall survival (OS) and graft failure (GF) after BSCT, a retrospective, single-center, matched-pair analysis was performed. Enrolled were all XM-positive BSCT performed at our institution from 1985 to 2000 (n=30). Controls (n=30) were matched for disease, disease stage, patient age, period of transplant, conditioning regimen, protocol for prevention of graft-versus-host disease, and type of donor (related vs. unrelated, HLA-identical vs. HLA-mismatched)., Results: Multivariate statistical analysis of all enrolled 60 transplants revealed GF as the all-dominating, independent risk factors for low OS (relative risk [RR]: 59.5, P<0.0001). Univariate (Kaplan-Meier) analysis could attribute inferior OS and high incidence of GF to the subgroup of HLA-mismatched, XM-positive transplants (P=0.01)., Conclusions: A XM should always be performed in patients awaiting a BSCT from HLA-mismatched donors, because a positive XM is a predictor for inferior OS due to GF in BSCT.

Outcomes after peripheral blood stem cell transplantation (PBSCT) for chronic phase chronic myeloid leukemia (n = 37) were compared with outcomes after bone marrow transplantation (BMT) (n = 54) in the HLA-compatible unrelated donor setting. Median follow-up was 17 months after PBSCT and 29 months after BMT. Both neutrophil and platelet recovery were faster after PBSCT (P <.05). PBSCT was associated with improved immune reconstitution, with higher peripheral blood naive (CD4(+)CD45RA(+)) and memory (CD4(+) CD45RO(+)) helper T cells at 3 months and 12 months after transplantation (P <.03). The cumulative incidence of acute (grades II-IV) and chronic graft-versus-host disease (GVHD) were similar, but BMT was associated with a higher cumulative incidence of severe, acute (grade III-IV) GVHD at 24% as compared with 8% with PBSCT (P <.05). Molecular relapse, defined by 2 consecutive positive polymerase chain reaction assays for bcr-abl within a 4-week interval, occurred in 12 of 45 evaluable patients (27%) after BMT and in 4 of 37 (11%) after PBSCT (not significant). Cytogenetic relapse occurred in 5 of 54 patients after BMT (9%) and in 1 of the 37 (3%) patients after PBSCT (not significant). Seventeen of the 54 patients died after BMT (31%), as compared with 2 of 37 patients after PBSCT (5%). Deaths in the BMT group were associated mainly with infections and severe, acute GVHD. The estimated probability of transplant-related mortality (TRM) and disease-free survival at 1000 days after receiving the transplant were 30% and 64% in the BMT group and 5% and 91% in the PBSCT group (P <.03). Overall survival 1000 days after receiving the transplant was 66% after BMT and 94% after PBSCT (P <.02). In the multivariate analysis, only acute GVHD significantly influenced TRM (P <.01).