HUMAN IMMUNOLOGY doi: 10.1111/sji.12244 ..................................................................................................................................................................
T cell Reconstitution in Allogeneic Haematopoietic Stem Cell Transplantation: Prognostic Significance of Plasma Interleukin-7 K. Kielsen*†, K. K. Jordan*†, H. H. Uhlving*†, P. L. Pontoppidan*‡, Z. Shamim*†, M. Ifversen†, C. Heilmann†, C. H. Nielsen*, H. Sengeløv§, L. P. Ryder¶ & K. G. M€ uller*†
Abstract *Institute for Inflammation Research, Department of Infectious Diseases and Rheumatology, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark; †Department of Pediatrics and Adolescent Medicine, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark; ‡Department of Nutrition, Exercise and Sports, University of Copenhagen, Copenhagen, Denmark; §Department of Hematology, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark; and ¶The Tissue Typing Laboratory, Department of Clinical Immunology, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark
Received 1 August 2014; Accepted in revised form 18 September 2014 Correspondence to: K. G. M€uller, MD, PhD, DMSc, DepartmentofPediatricsandAdolescentMedicine, Copenhagen University Hospital Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark. E-mail: [email protected]
Infections and acute graft-versus-host disease (aGVHD) are major causes of treatment-related mortality and morbidity following allogeneic haematopoietic stem cell transplantation (HSCT). Both complications depend on reconstitution of the T-lymphocyte population based on donor T cells. Although it is well established that Interleukin-7 (IL-7) is a cytokine essential for de novo T cell development in the thymus and homoeostatic peripheral expansion of T cells, associations between circulating levels of IL-7 and T cell reconstitution following HSCT have not been investigated previously. We prospectively measured IL-7 levels in 81 patients undergoing myeloablative HSCT with either sibling donor or an unrelated donor. Plasma IL-7 levels peaked at day +7 post-transplant (1.3– 82.4 pg/ml), at the time of maximal lymphopaenia. In multivariate analysis, peak levels of IL-7 were significantly higher in patients treated with antithymocyte globulin (ATG) compared with those not treated with ATG (P = 0.0079). IL-7 levels at day +7 were negatively associated with T cell counts at day +30 to +60 (at day +60: CD3+: b = 10.6 9 106 cells/l, P = 0.0030; CD8+: b = 8.4 9 106 cells/l, P = 0.061; CD4+: 6 b = 2.1 9 10 cells/l, P = 0.062) in multivariate analyses. In adults, high IL-7 levels were associated with increased risk of grade II-IV aGVHD (OR = 5.4, P = 0.036) and reduced overall survival (P = 0.046). The present data indicate that high plasma levels of IL-7 in the early post-transplant period are predictive for slow T cell reconstitution, increased risk of aGVHD and increased mortality following HSCT.
Introduction Allogeneic myeloablative haematopoietic stem cell transplantation (HSCT) is challenged by long-lasting immunodeficiency and acute graft-versus-host disease (aGVHD), both contributing significantly to morbidity and transplant-related mortality [1–3]. Although effective T cell reconstitution reduces the risk of infections, this also carries the risk for expansion of alloreactive T cells exacerbating aGVHD [4]. More detailed insights into the process of immune reconstitution after HSCT is essential to limit infections and in the same time allow effective immune surveillance of cancer cells without increasing the risk of aGVHD. Interleukin-7 (IL-7) is a cytokine essential for de novo T cell development in the thymus and homoeostatic peripheral expansion of T cells [5–9]. Accordingly, recombinant
72
IL-7 has been suggested as a potential therapeutic agent in patients with lymphopaenia [10–12]. IL-7 signals through the IL-7 receptor (IL-7R), a heterodimer consisting of the common cytokine receptor c-chain (CD132) and the high-affinity IL-7 receptor a-chain (IL-7Ra, CD127), which is also used by the thymic stromal lymphopoietin receptor [13]. IL-7Ra is expressed only on lymphocytes, being strictly regulated during different developmental stages of T cells and is downregulated in response to IL-7 stimulation [7, 14, 15]. IL-7 is produced constitutively by non-haematopoietic stromal and epithelial cells in lymphoid organs, intestine, skin and liver. According to the current view, plasma levels of IL-7 are determined primarily by the size of the T cell population through consumption by binding to IL-7R, rather than by regulation of IL-7 production [6].
Ó 2014 John Wiley & Sons Ltd
K. Kielsen et al. IL-7 and T cell Reconstitution in Allogeneic HSCT 73 ..................................................................................................................................................................
A role of IL-7 in HSCT has been suggested by previous studies reporting an association between high levels of IL-7 and the risk of aGVHD [16–18]. Moreover, studies from our group indicate that single nucleotide polymorphisms of the donor in the gene encoding IL-7Ra are associated with increased risk of transplant-related mortality and aGVHD [19–21]. Additionally, studies in mice indicate that treatment with IL-7 enhances both thymus-dependent T cell generation and peripheral proliferation of mature donor-derived T cells following HSCT [22]. These studies suggest that plasma levels of IL-7 may be associated with T cell reconstitution. We investigated associations between plasma IL-7 levels in the early post-transplant phase and lymphocyte reconstitution, aGVHD and mortality in patients undergoing myeloablative allogeneic HSCT.
Materials and methods Patients. In this prospective study, we consecutively recruited 81 patients undergoing stem cell transplantation. Patients undergoing their first myeloablative allogeneic HSCT at Copenhagen University Hospital Rigshospitalet, Denmark, from June 2010 to January 2013 and older than 1 year, were eligible for inclusion. Written informed consent was obtained from 81 patients and/or their legal guardians of a total of 154 patients matching the inclusion criteria during the study period, while non-participating patients were excluded due to lack of consent. The study protocol was approved by the local ethics committee (H-12010-009). Clinical characteristics of patients are listed in Table 1. The study included 39 children (median age 7.8 years (1.1–15.8)) and 42 adults (40.2 years (16.5–55.4)). Diagnoses were ALL (n = 26), AML (n = 19), myelodysplastic syndrome (n = 10), other haematological malignancies (n = 8) and non-malignant diseases (n = 18). Donors were either fully human leucocyte antigen A-, B-, C-, DR- and DQ-matched siblings (n = 18), matched unrelated donors (n = 55) (bone marrow or peripheral blood stem cell grafts) or HLA-mismatched umbilical cord blood grafts (n = 8). All patients were pretreated with a myeloablative conditioning regimen, consisting of total body irradiation (TBI) plus cyclophosphamide or etoposide (n = 46) or chemotherapy-based conditioning without TBI (n = 35). Antithymocyte globulin (ATG) was given to 37 patients. The indication for ATG was transplantation with an unrelated donor in children and the use of an unrelated donor with at least one antigen mismatch in adults. Allografts were nonT cell depleted and capped based on counts of total nucleated cells (bone marrow grafts) or CD34+ cells (peripheral blood stem cell grafts). T cell counts in the grafts were not available. Diagnosis, donor age, sex mismatch, pretransplant Karnofsky/Lansky performance score, cell dose/kg, graft type and cytomegalovirus (CMV) antibody status for
Ó 2014 John Wiley & Sons Ltd
donors and recipients were similar in participants and non-participants. Included patients were younger than non-included patients (mean age 22.9 versus 32.2 years, P < 0.0001). Fewer included patients received TBI as part of the conditioning (57% versus 70%, P = 0.021), while a higher number of included patients were treated with ATG (46% versus 23%, P = 0.0073) compared with nonincluded patients. Fourteen of the included patients were not available for the whole set of blood samples, due to relapse of leukaemia (n = 6), graft failure (n = 1), death (n = 4), withdrawal of consent (n = 1) and transfer to other hospital for clinical follow-up (n = 2). Healthy controls. Blood samples from 36 healthy, unrelated controls were collected from laboratory staff and students at the University of Copenhagen. All controls were Scandinavian Caucasians with a median age of 26.0 years (21.0–41.6). Quantification of plasma IL-7. Patient samples were collected before the start of conditioning, at the day of transplantation before graft infusion, and at day +7 (3), day +14 (3), day +21 (3), day +28 (3), day +60 (only paediatric patients) and day +90 post-transplantation. EDTA-anticoagulated blood was centrifuged shortly after collection and plasma was stored at 80 °C. Plasma IL-7 concentrations were measured in duplicate by high-sensitivity enzyme-linked immunosorbent assay (Quantikine HS ELISA; R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s instructions. The lower limit of detection was 0.1 pg/ml. T, B and NK cells in peripheral blood. T cell, B cell and NK cell counts were determined at day +30 (day +21 to +44, n = 71), +60 (day +45 to +70, n = 71), +90 (day +77 to +105, n = 47) and +180 (day +161 to +197, n = 51). Peripheral blood was analyzed using Trucount Tubes (Becton Dickinson, Albertslund, Denmark) for quantification of lymphocyte subsets and a panel of conjugated monoclonal antibodies: CD3-PerCP, CD3FITC, CD4-FITC, CD8-PE, CD45-PerCP, CD16/56-PE, CD20-FITC and CD19-PE (Becton Dickinson) on a FC500 flow cytometer (Beckman Coulter, Copenhagen, Denmark). T cells were determined as CD3+ T cells, CD3+CD4+ T cells and CD3+CD8+ T cells. NK cells were numbered based on a CD3 CD45+CD16+CD56+ phenotype. B cells were characterized as total B cells (CD45+CD19+), mature B cells (CD45+CD19+CD20+) and immature B cells (CD45+CD19+CD20 ) (Fig. S1). Statistical analysis. Mann–Whitney U-test, Wilcoxon rank sum test or Kruskal–Wallis test were used for continuous variables, and Fisher’s exact test was applied for categorical variables. Correlation analyses were performed using Spearman nonparametric analysis. Clinical characteristics associated with high IL-7 levels day +7 were analyzed using a multiple linear regression model. Linear and logistic multiple regression analyses were used to determine vari-
74 IL-7 and T cell Reconstitution in Allogeneic HSCT K. Kielsen et al. .................................................................................................................................................................. Table 1 Patient characteristics and transplantation modalities, n = 81. Characteristics Age at transplantation (years), mean (range) Recipients Donors Disease at transplantation, no. of patients (%) Acute lymphoblastic leukaemia Acute myeoloid leukaemia Other leukaemia Myelodysplastic syndrome Non-Hodgkin’s lymphoma Multiple myelomas Severe aplastic anaemia Immunodeficiency Other non-malignant diseases Donor type, no. (%) HLA-identical. siblings HLA-matched unrelated donors (9/10 or 10/10 match) HLA-mismatched umbilical cord blood donors (4/6 or 5/6 mismatch) Stem cell source, no. (%) Bone marrow stem cells Peripheral blood stem cells, G-CSF mobilized Umbilical cord blood, 1–2 donors per recipient Total nucleated cell dose infused 9106/kg recipient wt, median (range)a Conditioning regimen, no. of patients (%) TBIb + CY or Etoposide BU-CY Other high-dose chemotherapy-based conditioning regimens Fludarabine-based conditioning Anti-thymocyte globulin as part of conditioning regimen, no. (%) GVHD prophylaxis, no. (%) CsA + MTX CsA + corticosteroids CsA alone Sex mismatch (female donor to male recipient), no. (%)c CMV IgG mismatch, no. (%) Seronegative donor to seropositive recipient Seropositive donor to seronegative recipient
Values
17.6 (1.1–55.4) 30.0 (0.0–60.8) 26 19 5 10 2 1 6 7 5
(32.1) (23.5) (6.2) (12.3) (2.5) (1.2) (7.4) (8.6) (6.2)
18 (22.2) 55 (67.9) 8 (9.9)
57 16 8 3.8
(70.4) (19.8) (9.9) (0.003–16.3)
46 (56.8) 17 (21.0) 6 (7.4) 12 (14.8) 37 (45.7)
70 3 8 8
(86.4) (3.7) (9.9) (9.9)
children and adults, with a stratification of patients based on the median IL-7 level in each age group. All P-values reported are two sided, and P < 0.05 was considered statistically significant. All statistical analyses were performed using R statistical software version 2.15.3 (R Foundation for Statistical Computing, Vienna, Austria) and RStudio (RStudio, Boston, MA, USA). The package cmprsk was used for cumulative incidence analysis.
Results Transplant outcome
Mean follow-up time was 17.9 months from transplantation (4.7–36.0 months). Seventy-seven of 81 (91.4%) patients engrafted. Four patients were retransplanted (day +51, +80, +157 and +245) due to insufficient engraftment, post-transplant lymphoproliferative disorder, relapse of leukaemia and bone marrow aplasia, respectively. Twentyone per cent of the patients (n = 17) died at median 99 days (9–716 days) following transplantation, of those, 14 (82%) were adults. Of 63 patients transplanted due to a malignant disease, nine had relapse post-HSCT, and nine patients died in continuous remission. Causes of death were recurrence of primary malignant disease (n = 7), infections (n = 4), GVHD [aGVHD (n = 2), cGVHD (n = 1)], organ failure (n = 2) and haemorrhage (n = 1). Thirty-three patients (40.7%) developed grade II–IV acute GVHD day +20 (+9 to +106) post-transplant; of those, 23 (70%) were adults. In five patients, aGVHD occurred before day +14. One patient was not included in aGVHD analysis due to death from infection 9 days after HSCT. Patients who relapsed or were retransplanted were excluded from analysis from the date of the event. Plasma IL-7 levels during HSCT
24 (29.6) 9 (11.1)
TBI, Total body irradiation; CY, cyclophosphamide; BU, busulphan; CsA, cyclosporine A; MTX, methotrexate; CMV, cytomegalovirus. a Available in 76 transplantations (94%). b TBI was given as either a single dose (2 Gy, n = 4), fractionated TBI (4 Gy for 3 days, n = 22) or hyperfractionated TBI (2 9 2 Gy for 3 days, n = 20). c Umbilical cord blood graft not included in sex-mismatch frequencies due to 2 donors and low risk of alloreactivity.
ables associated with lymphocyte reconstitution and aGVHD, respectively, including variables based on their statistical significance in univariate analyses. Kaplan–Meier estimates with log-rank test for overall survival and cumulative incidence analysis for transplantrelated mortality and relapse frequencies were applied. Due to differences in conditioning and frequencies of aGVHD and mortality between children (
T cell Reconstitution in Allogeneic Haematopoietic Stem Cell Transplantation: Prognostic Significance of Plasma Interleukin-7 K. Kielsen*†, K. K. Jordan*†, H. H. Uhlving*†, P. L. Pontoppidan*‡, Z. Shamim*†, M. Ifversen†, C. Heilmann†, C. H. Nielsen*, H. Sengeløv§, L. P. Ryder¶ & K. G. M€ uller*†
Abstract *Institute for Inflammation Research, Department of Infectious Diseases and Rheumatology, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark; †Department of Pediatrics and Adolescent Medicine, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark; ‡Department of Nutrition, Exercise and Sports, University of Copenhagen, Copenhagen, Denmark; §Department of Hematology, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark; and ¶The Tissue Typing Laboratory, Department of Clinical Immunology, Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark
Received 1 August 2014; Accepted in revised form 18 September 2014 Correspondence to: K. G. M€uller, MD, PhD, DMSc, DepartmentofPediatricsandAdolescentMedicine, Copenhagen University Hospital Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark. E-mail: [email protected]
Infections and acute graft-versus-host disease (aGVHD) are major causes of treatment-related mortality and morbidity following allogeneic haematopoietic stem cell transplantation (HSCT). Both complications depend on reconstitution of the T-lymphocyte population based on donor T cells. Although it is well established that Interleukin-7 (IL-7) is a cytokine essential for de novo T cell development in the thymus and homoeostatic peripheral expansion of T cells, associations between circulating levels of IL-7 and T cell reconstitution following HSCT have not been investigated previously. We prospectively measured IL-7 levels in 81 patients undergoing myeloablative HSCT with either sibling donor or an unrelated donor. Plasma IL-7 levels peaked at day +7 post-transplant (1.3– 82.4 pg/ml), at the time of maximal lymphopaenia. In multivariate analysis, peak levels of IL-7 were significantly higher in patients treated with antithymocyte globulin (ATG) compared with those not treated with ATG (P = 0.0079). IL-7 levels at day +7 were negatively associated with T cell counts at day +30 to +60 (at day +60: CD3+: b = 10.6 9 106 cells/l, P = 0.0030; CD8+: b = 8.4 9 106 cells/l, P = 0.061; CD4+: 6 b = 2.1 9 10 cells/l, P = 0.062) in multivariate analyses. In adults, high IL-7 levels were associated with increased risk of grade II-IV aGVHD (OR = 5.4, P = 0.036) and reduced overall survival (P = 0.046). The present data indicate that high plasma levels of IL-7 in the early post-transplant period are predictive for slow T cell reconstitution, increased risk of aGVHD and increased mortality following HSCT.
Introduction Allogeneic myeloablative haematopoietic stem cell transplantation (HSCT) is challenged by long-lasting immunodeficiency and acute graft-versus-host disease (aGVHD), both contributing significantly to morbidity and transplant-related mortality [1–3]. Although effective T cell reconstitution reduces the risk of infections, this also carries the risk for expansion of alloreactive T cells exacerbating aGVHD [4]. More detailed insights into the process of immune reconstitution after HSCT is essential to limit infections and in the same time allow effective immune surveillance of cancer cells without increasing the risk of aGVHD. Interleukin-7 (IL-7) is a cytokine essential for de novo T cell development in the thymus and homoeostatic peripheral expansion of T cells [5–9]. Accordingly, recombinant
72
IL-7 has been suggested as a potential therapeutic agent in patients with lymphopaenia [10–12]. IL-7 signals through the IL-7 receptor (IL-7R), a heterodimer consisting of the common cytokine receptor c-chain (CD132) and the high-affinity IL-7 receptor a-chain (IL-7Ra, CD127), which is also used by the thymic stromal lymphopoietin receptor [13]. IL-7Ra is expressed only on lymphocytes, being strictly regulated during different developmental stages of T cells and is downregulated in response to IL-7 stimulation [7, 14, 15]. IL-7 is produced constitutively by non-haematopoietic stromal and epithelial cells in lymphoid organs, intestine, skin and liver. According to the current view, plasma levels of IL-7 are determined primarily by the size of the T cell population through consumption by binding to IL-7R, rather than by regulation of IL-7 production [6].
Ó 2014 John Wiley & Sons Ltd
K. Kielsen et al. IL-7 and T cell Reconstitution in Allogeneic HSCT 73 ..................................................................................................................................................................
A role of IL-7 in HSCT has been suggested by previous studies reporting an association between high levels of IL-7 and the risk of aGVHD [16–18]. Moreover, studies from our group indicate that single nucleotide polymorphisms of the donor in the gene encoding IL-7Ra are associated with increased risk of transplant-related mortality and aGVHD [19–21]. Additionally, studies in mice indicate that treatment with IL-7 enhances both thymus-dependent T cell generation and peripheral proliferation of mature donor-derived T cells following HSCT [22]. These studies suggest that plasma levels of IL-7 may be associated with T cell reconstitution. We investigated associations between plasma IL-7 levels in the early post-transplant phase and lymphocyte reconstitution, aGVHD and mortality in patients undergoing myeloablative allogeneic HSCT.
Materials and methods Patients. In this prospective study, we consecutively recruited 81 patients undergoing stem cell transplantation. Patients undergoing their first myeloablative allogeneic HSCT at Copenhagen University Hospital Rigshospitalet, Denmark, from June 2010 to January 2013 and older than 1 year, were eligible for inclusion. Written informed consent was obtained from 81 patients and/or their legal guardians of a total of 154 patients matching the inclusion criteria during the study period, while non-participating patients were excluded due to lack of consent. The study protocol was approved by the local ethics committee (H-12010-009). Clinical characteristics of patients are listed in Table 1. The study included 39 children (median age 7.8 years (1.1–15.8)) and 42 adults (40.2 years (16.5–55.4)). Diagnoses were ALL (n = 26), AML (n = 19), myelodysplastic syndrome (n = 10), other haematological malignancies (n = 8) and non-malignant diseases (n = 18). Donors were either fully human leucocyte antigen A-, B-, C-, DR- and DQ-matched siblings (n = 18), matched unrelated donors (n = 55) (bone marrow or peripheral blood stem cell grafts) or HLA-mismatched umbilical cord blood grafts (n = 8). All patients were pretreated with a myeloablative conditioning regimen, consisting of total body irradiation (TBI) plus cyclophosphamide or etoposide (n = 46) or chemotherapy-based conditioning without TBI (n = 35). Antithymocyte globulin (ATG) was given to 37 patients. The indication for ATG was transplantation with an unrelated donor in children and the use of an unrelated donor with at least one antigen mismatch in adults. Allografts were nonT cell depleted and capped based on counts of total nucleated cells (bone marrow grafts) or CD34+ cells (peripheral blood stem cell grafts). T cell counts in the grafts were not available. Diagnosis, donor age, sex mismatch, pretransplant Karnofsky/Lansky performance score, cell dose/kg, graft type and cytomegalovirus (CMV) antibody status for
Ó 2014 John Wiley & Sons Ltd
donors and recipients were similar in participants and non-participants. Included patients were younger than non-included patients (mean age 22.9 versus 32.2 years, P < 0.0001). Fewer included patients received TBI as part of the conditioning (57% versus 70%, P = 0.021), while a higher number of included patients were treated with ATG (46% versus 23%, P = 0.0073) compared with nonincluded patients. Fourteen of the included patients were not available for the whole set of blood samples, due to relapse of leukaemia (n = 6), graft failure (n = 1), death (n = 4), withdrawal of consent (n = 1) and transfer to other hospital for clinical follow-up (n = 2). Healthy controls. Blood samples from 36 healthy, unrelated controls were collected from laboratory staff and students at the University of Copenhagen. All controls were Scandinavian Caucasians with a median age of 26.0 years (21.0–41.6). Quantification of plasma IL-7. Patient samples were collected before the start of conditioning, at the day of transplantation before graft infusion, and at day +7 (3), day +14 (3), day +21 (3), day +28 (3), day +60 (only paediatric patients) and day +90 post-transplantation. EDTA-anticoagulated blood was centrifuged shortly after collection and plasma was stored at 80 °C. Plasma IL-7 concentrations were measured in duplicate by high-sensitivity enzyme-linked immunosorbent assay (Quantikine HS ELISA; R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s instructions. The lower limit of detection was 0.1 pg/ml. T, B and NK cells in peripheral blood. T cell, B cell and NK cell counts were determined at day +30 (day +21 to +44, n = 71), +60 (day +45 to +70, n = 71), +90 (day +77 to +105, n = 47) and +180 (day +161 to +197, n = 51). Peripheral blood was analyzed using Trucount Tubes (Becton Dickinson, Albertslund, Denmark) for quantification of lymphocyte subsets and a panel of conjugated monoclonal antibodies: CD3-PerCP, CD3FITC, CD4-FITC, CD8-PE, CD45-PerCP, CD16/56-PE, CD20-FITC and CD19-PE (Becton Dickinson) on a FC500 flow cytometer (Beckman Coulter, Copenhagen, Denmark). T cells were determined as CD3+ T cells, CD3+CD4+ T cells and CD3+CD8+ T cells. NK cells were numbered based on a CD3 CD45+CD16+CD56+ phenotype. B cells were characterized as total B cells (CD45+CD19+), mature B cells (CD45+CD19+CD20+) and immature B cells (CD45+CD19+CD20 ) (Fig. S1). Statistical analysis. Mann–Whitney U-test, Wilcoxon rank sum test or Kruskal–Wallis test were used for continuous variables, and Fisher’s exact test was applied for categorical variables. Correlation analyses were performed using Spearman nonparametric analysis. Clinical characteristics associated with high IL-7 levels day +7 were analyzed using a multiple linear regression model. Linear and logistic multiple regression analyses were used to determine vari-
74 IL-7 and T cell Reconstitution in Allogeneic HSCT K. Kielsen et al. .................................................................................................................................................................. Table 1 Patient characteristics and transplantation modalities, n = 81. Characteristics Age at transplantation (years), mean (range) Recipients Donors Disease at transplantation, no. of patients (%) Acute lymphoblastic leukaemia Acute myeoloid leukaemia Other leukaemia Myelodysplastic syndrome Non-Hodgkin’s lymphoma Multiple myelomas Severe aplastic anaemia Immunodeficiency Other non-malignant diseases Donor type, no. (%) HLA-identical. siblings HLA-matched unrelated donors (9/10 or 10/10 match) HLA-mismatched umbilical cord blood donors (4/6 or 5/6 mismatch) Stem cell source, no. (%) Bone marrow stem cells Peripheral blood stem cells, G-CSF mobilized Umbilical cord blood, 1–2 donors per recipient Total nucleated cell dose infused 9106/kg recipient wt, median (range)a Conditioning regimen, no. of patients (%) TBIb + CY or Etoposide BU-CY Other high-dose chemotherapy-based conditioning regimens Fludarabine-based conditioning Anti-thymocyte globulin as part of conditioning regimen, no. (%) GVHD prophylaxis, no. (%) CsA + MTX CsA + corticosteroids CsA alone Sex mismatch (female donor to male recipient), no. (%)c CMV IgG mismatch, no. (%) Seronegative donor to seropositive recipient Seropositive donor to seronegative recipient
Values
17.6 (1.1–55.4) 30.0 (0.0–60.8) 26 19 5 10 2 1 6 7 5
(32.1) (23.5) (6.2) (12.3) (2.5) (1.2) (7.4) (8.6) (6.2)
18 (22.2) 55 (67.9) 8 (9.9)
57 16 8 3.8
(70.4) (19.8) (9.9) (0.003–16.3)
46 (56.8) 17 (21.0) 6 (7.4) 12 (14.8) 37 (45.7)
70 3 8 8
(86.4) (3.7) (9.9) (9.9)
children and adults, with a stratification of patients based on the median IL-7 level in each age group. All P-values reported are two sided, and P < 0.05 was considered statistically significant. All statistical analyses were performed using R statistical software version 2.15.3 (R Foundation for Statistical Computing, Vienna, Austria) and RStudio (RStudio, Boston, MA, USA). The package cmprsk was used for cumulative incidence analysis.
Results Transplant outcome
Mean follow-up time was 17.9 months from transplantation (4.7–36.0 months). Seventy-seven of 81 (91.4%) patients engrafted. Four patients were retransplanted (day +51, +80, +157 and +245) due to insufficient engraftment, post-transplant lymphoproliferative disorder, relapse of leukaemia and bone marrow aplasia, respectively. Twentyone per cent of the patients (n = 17) died at median 99 days (9–716 days) following transplantation, of those, 14 (82%) were adults. Of 63 patients transplanted due to a malignant disease, nine had relapse post-HSCT, and nine patients died in continuous remission. Causes of death were recurrence of primary malignant disease (n = 7), infections (n = 4), GVHD [aGVHD (n = 2), cGVHD (n = 1)], organ failure (n = 2) and haemorrhage (n = 1). Thirty-three patients (40.7%) developed grade II–IV acute GVHD day +20 (+9 to +106) post-transplant; of those, 23 (70%) were adults. In five patients, aGVHD occurred before day +14. One patient was not included in aGVHD analysis due to death from infection 9 days after HSCT. Patients who relapsed or were retransplanted were excluded from analysis from the date of the event. Plasma IL-7 levels during HSCT
24 (29.6) 9 (11.1)
TBI, Total body irradiation; CY, cyclophosphamide; BU, busulphan; CsA, cyclosporine A; MTX, methotrexate; CMV, cytomegalovirus. a Available in 76 transplantations (94%). b TBI was given as either a single dose (2 Gy, n = 4), fractionated TBI (4 Gy for 3 days, n = 22) or hyperfractionated TBI (2 9 2 Gy for 3 days, n = 20). c Umbilical cord blood graft not included in sex-mismatch frequencies due to 2 donors and low risk of alloreactivity.
ables associated with lymphocyte reconstitution and aGVHD, respectively, including variables based on their statistical significance in univariate analyses. Kaplan–Meier estimates with log-rank test for overall survival and cumulative incidence analysis for transplantrelated mortality and relapse frequencies were applied. Due to differences in conditioning and frequencies of aGVHD and mortality between children (
