Increased Infection-Related Mortality in... : Transplantation (original) (raw)
Natural killer (NK) cells interact with major histocompatibility complex class I molecules on target cells (1,2). In humans, NK cells are negatively regulated by killer cell immunoglobulin-like receptors (KIRs) recognizing groups of human leukocyte antigen (HLA) class I alleles. Transplantation across HLA barriers may trigger donor NK-cell alloreactivity if the recipient lacks KIR ligands that are present in the donor (KIR-ligand mismatch) (3,4). A remarkable role for such KIR-ligand mismatches was recently demonstrated in patients with acute myeloid leukemia (AML), receiving haploidentical T-cell–depleted allogeneic hematopoietic stem-cell transplantation (ASCT) (5). KIR-ligand mismatch was associated with significantly increased overall survival, better engraftment, and reduced incidence of graft-versus-host disease (GVHD) (5). This correlated with the possibility to expand donor-derived NK-cell clones that were capable of killing recipient tumor cells, forming an indirect evidence for the involvement of alloreactive NK cells. However, the impact of KIR-ligand mismatch has been more controversial in unrelated HLA-mismatched ASCT. In one report, the overall survival also was significantly better for patients with KIR-ligand–incompatible donors in unrelated ASCTs (6). In contrast, Davies et al. could not demonstrate any survival advantage for patients receiving KIR-ligand–mismatched grafts (7). Several explanations for this discrepancy have been suggested, including the stem-cell dose, T-cell depletion of the graft, and the type and intensity of posttransplantation immunosuppression. In particular, the use of antithymocyte globulin (ATG) during conditioning and as GVHD prophylaxis has been emphasized because it leads to in vivo depletion of donor T cells during the early posttransplantation period, thus allowing NK-cell alloreactivity (6). In this study we retrospectively analyzed the role of KIR-ligand-mismatch in patients receiving ATG as GVHD prophylaxis for hematopoietic grafts from unrelated donors displaying various grades of HLA disparity.
Study Design
Patients
A total of 190 patients with hematologic malignancies undergoing unrelated donor ASCT between 1988 and 2003 at Karolinska University Hospital Huddinge were included in the study. Details of the patient and donor characteristics are shown in Table 1. The study was approved by the ethics committee at Huddinge University Hospital.
Patient and donor characteristics
Continued
Transplantation Procedure
Details of the transplantation procedure, including stem-cell source, GVHD prophylaxis, type of ATG, and conditioning, are presented in Table 1 and reported elsewhere (8–11). All patients received ATG during the conditioning (12).
Definitions
Early disease was defined as all patients in first complete remission and chronic phase. Acute GVHD was diagnosed and graded according to previously reported criteria (13). Transplantation-related mortality (TRM) was defined as death from all causes in the absence of relapse. KIR-ligand mismatch was defined according to the algorithm used by Ruggeri et al. (5). In brief, the KIR-ligand–mismatched group involved pairs in which the recipient lacks KIR ligands present in the donors. The following KIR ligands were considered: HLA–Cw3-related alleles defined by Ser 77 and Asn 80 (e.g., Cw1, Cw3, Cw7, and Cw8); HLA–Cw4-related alleles defined by Asn77 and Lys 80 (e.g., Cw2, Cw4, Cw5, and Cw6). No donor–recipient pairs were mismatched for the other two major KIR ligands HLA-Bw4–related alleles and HLA-A3, A11.
Statistical Analysis
Data were analyzed as of December, 2003. Overall survival, disease-free survival, relapse, TRM, and GVHD were estimated by using the Kaplan-Meier method and compared with the log-rank test. In the univariate and multivariate analysis the Cox regression analysis was used. Factors included were HLA disparity, patient and donor age, patient and donor sex, cytomegalovirus status, diagnosis, disease stage, type of conditioning, nucleated cell dose, and stem-cell source.
RESULTS AND DISCUSSION
KIR-ligand match or mismatch was deduced from the high-resolution molecular typing of HLA of donors and recipients according to the algorithm used by Ruggeri et al. (5). In contrast with previously reported studies, but corroborating our preceding report (14), we observed inferior overall survival in patients receiving KIR-ligand–mismatched ASCT (Fig. 1A, _P_=0.01). The same results were obtained when analyzing only patients with myeloid leukemia (Fig. 1B, _P_=0.03). A multivariate statistical analysis revealed that these results could not be explained by any imbalance in the patient groups with respect to HLA disparity, patient and donor age, patient and donor sex, cytomegalovirus status, diagnosis, disease stage, type of conditioning, nucleated cell dose, and stem-cell source (Table 2). Nevertheless, we wanted to isolate the role of KIR-ligand incompatibility as much as possible and exclude the possibility that fully matched donors contributed to the better survival in the KIR-ligand–matched group. To that end we analyzed patients matched for HLA-A, -B, -DRB1 displaying only HLA-C mismatches. This narrow analysis only strengthened the observation that KIR-ligand mismatch was associated with inferior survival (Fig. 1C, _P_=0.03). There was no significant difference in relapse rate or engraftment in the two groups whether we analyzed the heterogeneous material or the donor–patient pairs with only HLA-C mismatches and with only myeloid leukemias (data not shown). The role of HLA-DQ mismatches could not be evaluated because most donor–recipient pairs were identical at this locus. HLA-DP mismatches were more common, and we found a higher frequency of double HLA-DP mismatches in the KIR-ligand–matched group. Our previous results, showing an association of mismatch at HLA-DP and increased TRM, suggest that such imbalance would favor a high TRM in the KIR-ligand–matched transplantations (14,15). We conclude that the DQ and DP mismatches did not influence our results that KIR-ligand mismatch is associated with increased TRM and infection-related mortality.
Multivariate analysis of causes of death
The reason for the inferior survival in patients receiving KIR-ligand–mismatched ASCT, as seen in this study, stemmed from their significantly higher TRM (Fig. 1D, _P_=0.02). The higher TRM was noted early posttransplantation, and all deaths occurred within the first 6 months in the KIR-ligand–mismatched group (Fig. 1D). Nevertheless, TRM remained significantly higher in KIR-ligand–mismatched transplantations even if deaths occurring before engraftment and within the first 30 days were excluded (_P_=0.015). The major causes for TRM are GVHD and opportunistic infections. We observed no significant differences in the development of acute GVHD between KIR-ligand–mismatched and matched donor–recipient pairs (Table 3). Instead, we observed that KIR-ligand mismatch correlated with a higher rate of infection-related mortality (35% vs. 16%, _P_=0.02), significantly more fungal infections (26% vs. 10%, _P_=0.03), and a trend for more septicemia (61% vs. 43%, _P_=0.097). Notably, KIR-ligand mismatch was associated with an increased infection-related mortality even when patients receiving reduced intensity conditioning were excluded from the analysis. Furthermore, a multivariate analysis revealed that KIR-ligand mismatch was an independent risk factor for TRM and infection-related mortality (Table 2).
Acute and chronic graft-versus-host disease
Although our results were obtained in a limited number of patients, the observation of an increased infection-related mortality in patients receiving KIR-ligand–mismatched ASCT at our center warrants further investigation. One may speculate that alloreactive NK cells interfered with effective immunity to infection by eliminating dendritic cells (DCs) in the recipient. Elimination of host DCs by alloreactive NK cells was suggested to be the mechanism behind the reduced risk of GVHD in haploidentical ASCT (5), because host DCs are known to be important for the induction of GVHD (16,17).
Significant differences in survival between KIR-ligand–matched and mismatched groups were not detected when only patients with AML were analyzed. Although TRM was significantly higher in these patients (57% vs. 23%, _P_=0.04), this was counteracted by the tendency for decreased relapse rate in patients with AML receiving KIR-ligand–mismatched grafts (0% vs. 28%, not statistically significant). It is possible that NK-mediated graft-versus-leukemia effects may have been operating in patients with AML at our center, but that such effects would need a larger number of patients to be become statistically significant.
CONCLUSION
Our report suggests that alloreactive NK cells, developing in KIR-ligand–mismatched ASCT, may interfere with immunity to infection, thus counteracting the beneficial graft-versus-leukemia effect and the survival advantage otherwise associated with KIR-ligand mismatch. Second, we observed inferior survival in KIR-ligand–mismatched ASCT despite the use of ATG during conditioning. This indicates that ATG may not be a key factor for preserving the NK cell-mediated graft-versus-leukemia effect. Prospective studies are needed to elucidate what details in the different transplantation protocols that determine the potential of NK cell-mediated alloreactivity against residual tumor cells and NK cell interference with immunity to infection.
REFERENCES
1. Karre K, Ljunggren HG, Piontek G, et al. Selective rejection of H-2-deficient lymphoma variants suggests alternative immune defense strategy. Nature 1986; 319(6055): 675.
2. Ljunggren HG, Karre K. In search of the ‘missing self': MHC molecules and NK cell recognition [see comments]. Immunol Today 1990; 11(7): 237.
3. Parham P, McQueen K. Alloreactive killer cells: hindrance and help for haematopoietic transplants. Nat Rev Immunol 2003; 3(2): 108.
4. Farag S, Fehniger T, Ruggeri L, et al. Natural killer cell receptors: new biology and insights into the graft-versus-leukemia effect. Blood 2002; 100(6): 1935.
5. Ruggeri L, Capanni M, Urbani E, et al. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science 2002; 295(5562): 2097.
6. Giebel S, Locatelli F, Lamparelli T, et al. Survival advantage with KIR ligand incompatibility in hematopoietic stem cell transplantation from unrelated donors. Blood 2003; 102(3): 814.
7. Davies SM, Ruggieri L, DeFor T, et al. Evaluation of KIR ligand incompatibility in mismatched unrelated donor hematopoietic transplants. Killer immunoglobulin-like receptor. Blood 2002; 100(10): 3825.
8. Ringden O, Ruutu T, Remberger M, et al. A randomized trial comparing busulphan with total body irradiation as conditioning in allogeneic marrow transplant recipients with leukemia. A report from the Nordic Bone Marrow Transplantation Group. Blood 1994; 83: 2723.
9. Ringden O, Remberger M, Runde V, et al. Peripheral blood stem cell (PBMC) transplantation from unrelated donors: a comparison with marrow transplantation. Blood 1999; 94(2): 455.
10. Slavin S, Nagler A, Naparstek E. Non myeloablative stem cell transplantation and cell therapy as an alternative to conventional bone marrow transplantation with lethal cytoreduction for the treatment of malignant and nonmalignant hematological disease. Blood 1998; 91: 756.
11. Hentschke P, Barkholt L, Uzunel M, et al. Low-intensity conditioning and hematopoietic stem cell transplantation in patients with renal and colon carcinoma. Bone Marrow Transplant 2003; 31: 253.
12. Remberger M, Svahn B-M, Hentschke P, et al. Effect on cytokine release and graft-versus-host-disease of different anti-T-cell antibodies during conditioning for unrelated haematopoietic stem cell transplantation. Bone Marrow Transplant 1999; 24: 823.
13. Glucksberg H, Storb R, Fefer A, et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HLA-matched sibling donors. Transplantation 1974; 18(4): 295.
14. Schaffer M, Aldener-Cannava A, Remberger M, et al. Roles of HLA-B, HLA-C and HLA-DPA1 incompatibilities in the outcome of unrelated stem-cell transplantation. Tissue Antigens 2003; 62(3): 243.
15. Ringden O, Schaffer M, Le Blanc K, et al. Which donor should be chosen for hematopoietic stem cell transplantation among unrelated HLA-A, -B, and -DRB1 genomically identical volunteers? Biol Blood Marrow Transplant 2004; 10(2): 128.
16. Kosaka H, Surh CD, Sprent J. Stimulation of mature unprimed CD8+ T cells by semiprofessional antigen-presenting cells in vivo. J Exp Med 1992; 176(5): 1291.
17. Shlomchik WD, Couzens MS, Tang CB, et al. Prevention of graft versus host disease by inactivation of host antigen-presenting cells. Science 1999; 285(5426): 412.
Keywords:
Killer cell immunoglobulin-like receptor; Stem-cell transplantation; Natural killer cell; Transplantation-related mortality; Opportunistic infections
© 2004 Lippincott Williams & Wilkins, Inc.