¹Department of Pediatrics at University of Michigan, Ann Arbor, MI, USA

²University of North Carolina, Chapel Hill, NC, USA

³Northwestern University, Chicago, IL, USA

⁴University of Pittsburgh, PA, USA

Correspondence to: Dr J E Levine, CCGC B1-207, University of Michigan Cancer Center, 1500 E Medical Center Drive, Ann Arbor, MI 48109-0914, USA

Between October 1995 and October 1998, 24 children aged 9 months to 17 years (median 11 years) underwent cytokine-mobilized allogeneic peripheral blood stem cell (PBSC) transplantation for treatment of hematological disorders. All of the transplants were the first allogeneic transplant for the recipient. Twenty patients were transplanted for hematological malignancies (ALL = 8, AML = 6, CML = 4, MDS = 2) and four patients were transplanted for non-malignant disease (thalassemia major = 2, Wiskott-Aldrich syndrome = 1, Kostmann's syndrome = 1). Nineteen donors were HLA-identical siblings, four were HLA-matched or single antigen mismatched parents, and one was a syngeneic transplant. Donors aged 8 to 38 years (median 15 years, 14 donors <18 years) received g-csf 10 g/kg/day subcutaneously beginning 4 days before pbsc collection and were submitted to one to three leukapheresis collections. the median cd34⁺ cell yield was 7.8 ´ 10⁶ cells/kg recipient body weight. all patients achieved an anc >0.5 ´ 10⁹/l after a median of 13 days (range 10-21). Twenty-three patients eventually achieved platelet transfusion independence. One patient died on day 63 without ever achieving platelet transfusion independence. Four patients received platelet transfusions to maintain a platelet count well above 20 ´ 10⁹/l due to bleeding complications. Of the 19 evaluable patients, the median time to a non-transfused platelet count of 20 ´ 10⁹/l was 12 days (range 0-44). Ten of 23 at-risk patients developed acute GVHD grades II to IV, with grades III to IV in four patients. Twelve of 19 patients followed for at least 100 days have developed chronic GVHD (extensive = 2, limited = 10) with an actuarial risk of chronic GVHD of 75% at 1 year. The Kaplan-Meier estimate of event-free survival is 65% at 2 years. Four patients died (GVHD = 3, VOD = 1), three patients relapsed, and one patient with thalassemia major had a late graft failure with autologous recovery. Based upon our experience, allogeneic PBSCT is safe for both pediatric donors and recipients and engraftment of neutrophils and platelets is rapid. Bone Marrow Transplantation (2000) 25, 13-18.

peripheral blood stem cells; allogeneic transplantation; GVHD; childhood

Allogeneic bone marrow transplantation has been employed as a curative approach in a number of childhood disease states including aplastic anemia, leukemia, bone marrow failure states and other disorders. Although there have been advances in supportive care, including the introduction of cytokine support to enhance engraftment, transplant-related morbidity and mortality remains a significant problem for recipients of bone marrow transplants. The use of cytokine mobilized peripheral blood stem cell transplants, pioneered in the adult and pediatric autologous settings,^1,2,3,4 have become commonplace in the adult allogeneic setting.^5,6,7 Because time to neutrophil and platelet engraftment is more rapid when cytokine-mobilized peripheral blood stem cells are employed,^8,9 investigators at each of our institutions independently became interested in exploring this source of stem cells in the allogeneic transplantation setting. Because these pilot explorations were undertaken independently at four separate institutions, there were differences in the approaches used with regard to mobilization and supportive care issues. We summarize in this paper our joint experience with cytokine-mobilized peripheral blood stem cell transplants in pediatric recipients.

Patients and methods

Patient accrual and categorization

Patients receiving a first allogeneic transplant according to each individual institutional protocol were eligible for inclusion. At three of the centers (Ann Arbor, Chapel Hill, Pittsburgh) all allogeneic transplant recipients not already enrolled on cooperative group protocols were offered the option of participating in their institutional PBSC protocols. At one center (Chicago), peripheral blood stem cell collection was performed either because of refusal to permit bone marrow collection (1) or to permit a single collection from one donor for two siblings with thalassemia. All patients were under the age of 18 years. Patients had to have a condition for which bone marrow transplant was indicated and an HLA-identical family member (except in one case in which a single antigen mismatch was permitted). Patient and donor characteristics are listed in Table 1. Patients with ALL were transplanted in CR1 because of the presence of the Philadelphia chromosome (patient No. 6) or in CR2 or greater. Patients with AML were transplanted in CR1 or greater. Patient No. 10 developed AML following treatment for non-Hodgkin's lymphoma. Patient No. 12 also had monosomy 7. Patient No. 20 suffered from neurofibromatosis type 1 and developed monosomy 7 associated MDS while on treatment for ALL.

Informed consent

Patients and donors or their parents/legal guardians signed an informed consent indicating their voluntary participation in an experimental protocol. All informed consents were approved by the Institutional Review Board of the local institution.

Donors

All PBSC donors were given G-CSF (Amgen, Thousand Oaks, CA, USA) 10 g/kg/day subcutaneously once daily beginning 4 days prior to PBSC collection and continuing daily until a sufficient number of CD34⁺ cells were obtained. At one institution, a minimum of 10 ´ 10⁶ CD34⁺ cells were required for transplant. Twenty-one donors were pheresed once, two donors were pheresed twice, and one donor required three phereses to achieve a sufficient number of CD34⁺ cells. Donor venous access was obtained through venipuncture in both arms in 22 donors. Two donors had central venous catheters placed for leukapheresis. Apheresis was performed with the use of a continuous flow blood separator (Cobe Laboratories, Lakewood, CO, USA) according to the manufacturer's manual. In 20 cases, peripheral blood stem cells were infused on the day of collection into the recipient. In three cases cells were collected in advance, cryopreserved, and later infused. In the remaining case, stem cells were collected from one donor for transplant into two siblings with thalassemia major. The collection was divided into two aliquots. One child received an aliquot of stem cells on the day of collection. The other aliquot was cryopreserved and later used for transplant for the other affected sibling. Donors were followed for complications at least until the day of the last stem cell collection. Follow-up examinations and blood counts were not obtained in most donors after the last stem cell collection. At one center (n = 3) donors were seen within 1 week from the collection and contacted at 6 months, 1 year, and yearly thereafter for follow-up.

Supportive care

Twenty-one recipients received G-CSF starting day 6 or 7 post transplant to enhance neutrophil engraftment. Three patients received no cytokine support post transplant. GVHD prophylaxis varied according to institutional practice. Seventeen patients received tacrolimus-based GVHD prophylaxis (tacrolimus alone = 5, tacrolimus and steroids = 2, tacrolimus and short-course methotrexate = 10). The remaining six patients received cyclosporine and short-course methotrexate. Short course methotrexate consisted of methotrexate 15 mg/m² on day +1 and 10 mg/m² on days 3, 6 and 11. In six patients the day 11 dose was omitted according to institutional protocol. GVHD prophylaxis was discontinued by day 180 unless the patient was being treated for GVHD. No GVHD prophylaxis was administered in the case of the syngeneic transplant. All of the grafts were unmanipulated. No T cell depletion was employed to prevent GVHD.

Evaluation and definitions

PBSC were analyzed for CD34⁺ cells by flow cytometry. Engraftment was documented by increasing neutrophil count and platelet count unsupported by transfusions and by either analysis of restriction fragment length polymorphisms or of variable number of tandem repeats. Neutrophil engraftment was defined as the first of 3 days that the absolute neutrophil count exceeded 0.5 ´ 10⁹/l following the post-transplant nadir. Platelet engraftment was defined as the first of 7 consecutive days with an untransfused platelet count exceeding 20 ´ 10⁹/l. Acute GVHD was graded as previously described.¹⁰ Chronic GVHD was assessed in all patients surviving to day 100. Limited chronic GVHD was defined as GVHD occurring after day 100 that involved localized skin and/or mouth and/or liver. Extensive chronic GVHD was defined as GVHD occurring after day 100 that did not meet the definition of limited chronic GVHD.

Patients that died post transplant after relapse were cat- egorized as dying of relapse regardless of the proximate cause. GVHD was listed as the cause of death in patients who died with grade 3 or 4 GVHD regardless of the proximate cause.

Results

Collections

Twenty one donors underwent a single leukapheresis. Two donors required two phereses. One donor required three phereses. In most cases the donors were older and weighed more than their respective recipients. Two donors required central venous catheters for the purpose of stem cell collection. There were no catheter-related complications. Six donors (one 9 years, the other five 14-22 years) complained of bone pain or headache relieved by acetaminophen. No donor developed fever or bleeding complications. No other donor complications were observed, however, formal follow-up of most donors was limited to the last day of stem cell collection. Follow-up blood counts were not obtained on donors. A median of 7.8 ´ 10⁶ CD34⁺ cells/kg recipient weight (range 1.6-15.5 ´ 10⁶, mean 7.9 ´ 10⁶) were transplanted.

Engraftment characteristics

All patients engrafted neutrophils as defined above (Table 1). The range in days to an absolute neutrophil count of 0.5 ´ 10⁹/l was 10-31 days with a median of 13 days. Nineteen patients were evaluable for platelet engraftment. Five patients received platelet transfusions to maintain a platelet count well above 20 ´ 10⁹/l due to bleeding complications. One of these patients died without ever having been transfusion independent. Three eventually became platelet transfusion independent. Of the 19 evaluable patients, the range in days to a nontransfused platelet count of 20 ´ 10⁹/l was zero (never fell below 20 ´ 10⁹/l)-44 days with a median of 12 days. The median time to neutrophil engraftment was 13 days in patients who did or did not receive methotrexate as part of their GVHD prophylaxis. The median time to platelet engraftment was 15 days in children who received methotrexate as part of their GVHD prophylaxis and 12 days in children who did not receive methotrexate.

All patients were found to be full donor chimeras by either restriction fragment length polymorphisms or variable number of tandem repeats analysis. One child with -thalassemia major later had autologous reconstitution.

Survival

There were four treatment-related deaths. Three patients died of GVHD or related complications on days 60, 63 and 315. One patient died of veno-occclusive disease on day 48. Three patients relapsed with their original disease. One patient with -thalassemia major had autologous reconstitution. The Kaplan-Meier event-free survival estimate is 65% at 2 years (Figure 1). Median follow-up for the cohort is 21 months.

Risk of relapse

Three of 20 patients transplanted for malignant conditions have relapsed. The patients who relapsed were a 2-year-old girl with acute megakaryocytic leukemia complicated by monosomy 7 in first complete remission, an 8-year-old boy with monosomy 7 associated myelodysplastic syndrome arising during treatment for acute lymphoblastic leukemia, and a 15-year-old girl with acute myeloid leukemia in first complete remission who received a syngeneic transplant. The actuarial risk of relapse at 2 years is 19%.

Graft-versus-host disease

Twenty three patients were evaluable for acute GVHD (Table 2). One patient was not at risk due to syngeneic transplantation. Ten of 23 patients developed acute GVHD, grades 2-4. There were 13 cases of grade 0 GVHD, six cases of grade II GVHD, one case of grade III GVHD and three cases of grade IV GVHD. GVHD was the cause of death in three patients. Patients who received methotrexate as part of the GVHD prophylaxis were less likely to develop acute GVHD (P = 0.0186, Fisher's exact test). Sixteen patients received methotrexate in combination with either cyclosporine or tacrolimus. Four of these patients developed acute GVHD, three cases of grade II GVHD and one case of grade III GVHD. Two of the cases developed in recipients of maternal transplants and in one of these cases there was also a single antigen class II mismatch. Seven patients received either tacrolimus alone or in combination with steroids. Six of the seven patients developed acute GVHD, three cases of grade II GVHD and three cases of grade IV GVHD. All three GVHD related deaths occurred in the group treated with single agent tacrolimus prophylaxis. For the entire group of 23 at-risk patients the actuarial risk of developing acute GVHD grades II-IV by day 100 was 45%.

Twelve out of 20 patients followed for at least 100 days have developed chronic GVHD. Onset of chronic GVHD occurred between days 100 and 273. The syngeneic recipient was not at risk of chronic GVHD. Three patients died before day 100 and one patient had autologous reconstitution at day 63. Ten of the 12 cases were limited chronic GVHD of the skin, liver, and/or mouth. Treatment of these 10 patients varied but generally included steroids, tacrolimus or cyclosporine and occasionally mycophenolate. Seven of the 10 patients are either off all immunosuppressives or undergoing a slow taper for quiescent chronic GVHD. Three patients have manifestations of active skin or oral involvement and remain on immunosuppressive therapy (tacrolimus) at treatment dosages. Two additional patients developed extensive chronic GVHD. One patient developed bronchiolitis obliterans and remains on immunosuppressive therapy consisting of clofazamine and inhaled steroids. He has a Karnofsky perfomance scale of 90%. One patient developed a constellation of GVHD symptoms resembling acute involvement of the skin, gut, and liver with an onset on day +273. This patient had received a maternal transplant and had developed clinical grade II acute GVHD. Two months after the cessation of treatment for the acute GVHD, she developed florid GVHD symptoms that were unresponsive to immunosuppression and proved to be lethal. Thus, of the 12 patients that developed chronic GVHD, one patient died, seven are either off immunosuppression or on a slow taper with quiescent disease, and four are still receiving treatment for active chronic GVHD with follow-up ranging from 8 months to 3.5 years (median 1.5 years). Seventy-five percent of the patients at risk developed chronic GVHD by 1 year post transplant. So far, four patients followed from 673 to 1012 days post transplant have not developed chronic GVHD. No statistically significant difference in the probability of developing chronic GVHD was observed between patients who received methotrexate as part of their GVHD prophylaxis and those who did not receive methotrexate.

Discussion

In our series of 24 pediatric recipients of allogeneic cytokine-mobilized peripheral blood stem cells, all patients engrafted. Rapid hematopoietic recovery was observed with median times to a neutrophil count >0.5 ´ 10⁹/l and a platelet count >20 ´ 10⁹/l of 13 days and 12 days, respectively. Time to platelet engraftment was less consistently rapid than time to neutrophil engraftment. Our results confirm and expand upon the similar findings of Minero and colleagues¹¹ who reported in 11 pediatric patients (seven first allogeneic transplant recipients, four second allogeneic transplant recipients) a median time to a neutrophil count >0.5 ´ 10⁹/l of 12 days and a median time to a platelet count >50 ´ 10⁹/l of 15 days. Diaz and colleagues¹² reported that in five pediatric recipients of allogeneic peri- pheral stem cell transplants median times to a neutrophil count >0.5 ´ 10⁹/l and a platelet count >20 ´ 10⁹/l were 15 days and 14 days, respectively. By means of historical comparison, sibling bone marrow transplant patients given G-CSF to enhance neutrophil engraftment and methotrexate as part of GVHD prophylaxis took a median of 16 days to reach a neutrophil count >0.5 ´ 10⁹/l and a median of 27 days to reach a platelet count >20 ´ 10⁹/l. When methotrexate was not given to sibling BMT recipients, the median time to a neutrophil count >0.5 ´ 10⁹/l was 10 days and the median time to reach a platelet count >20 ´ 10⁹/l was 23 days.¹³ Thus, our series of patients, whose median time to platelet engraftment was 15 and 12 days, respectively, depending on whether or not they received methotrexate, appears to have a more rapid platelet engraftment when compared to a group of sibling allogeneic marrow transplant recipients reported in the literature. Time to neutrophil engraftment in the patients reported here is similar to the reported experience.

Our experience with pediatric donors was favorable. Fourteen of the donors were children: eight adolescents and six preteens. The youngest donors were 8 years old. Cells from most donors could be collected with a single leukapheresis and by using peripheral venous access. Complications were limited to mild complaints attributable to the G-CSF injections. In this series, most donors were older and weighed more than the recipients. Presumably this weight difference would decrease the need for longer or multiple leukapheresis collections. Moreover, when one considers the relative ease of collection, peripheral stem cell collection from healthy children, especially larger donors, should merit routine consideration.

Acute graft-versus-host disease was observed in 46% of the patients. There was an increased risk of developing acute GVHD when methotrexate was not part of the GVHD prophylaxis regimen. When methotrexate was part of the prophylaxis regimen the incidence of acute GVHD (25%) was similar to that expected in a pediatric population undergoing bone marrow transplant (33%).¹⁴ In contrast, patients who received tacrolimus alone or with steroids appeared more likely to develop acute GVHD with the GVHD likely to be more severe. When tacrolimus or cyclosporine is used in combination with short course methotrexate, allogeneic peripheral blood stem cell transplant recipients appear to be at no increased risk of developing acute GVHD when compared to bone marrow transplant recipients.

The incidence of chronic GVHD was 75% at 1 year. By comparison, children who receive bone marrow transplants have an approximately 30-35% probability of developing chronic GVHD by 2 years.¹⁴ Studies in adult recipients of allogeneic PBSCT have demonstrated an increased risk of chronic GVHD in up to 80% of recipients.^15,16 In this series, most of the chronic GVHD was limited to the mouth, skin or liver, but two patients developed extensive chronic GVHD that was fatal in one case. The majority of patients diagnosed with chronic GVHD are either off immunosuppressive therapy or on tapering doses with quiescent disease. Whilst our number of evaluable patients at risk for chronic GVHD was small, the risk does appear to be increased. Further experience will permit a more accurate assessment of the true risk. Because chronic GVHD is associated with a reduced risk of leukemic relapse, an increased risk of chronic GVHD may be an acceptable trade off for a high risk population.¹⁷

In conclusion, allogeneic PBSCT is safe for pediatric donors and recipients. Recipients achieved rapid neutrophil and platelet engraftment. In particular, platelet engraftment appears more rapid than reported experience with bone marrow transplant. The incidence of acute GVHD is higher than anticipated, but this is mainly seen in patients who receive a GVHD prophylaxis regimen that lacks short course methotrexate. The chronic GVHD rate appears high when compared with that following bone marrow transplantation, but in this series of patients chronic GVHD was generally limited and manageable.

Acknowledgements

The authors gratefully acknowledge the excellent technical expertise and clinical care provided by the apheresis and BMT unit staffs at our respective hospitals.

1 Henon PhR, Liang H, Beck-Wirth G et al. Comparison of hematopoietic and immune recovery after autologous bone marrow or blood stem cell transplants. Bone Marrow Transplant 1992; 9: 285-291, MEDLINE

2 To LB, Roberts MM, Haylock DN et al. Comparison of hematological recovery times and supportive care requirements of autologous recovery phase peripheral blood stem cell transplants, autologous bone marrow transplants and allogeneic bone marrow transplants. Bone Marrow Transplant 1992; 9: 277-284, MEDLINE

3 Huan S, Hester J, Spitzer G et al. Influence of mobilized peri-pheral blood cells on the hematopoietic recovery of autologous marrow and recombinant human granulocyte-macrophage colony-stimulating factor after high-dose cyclophosphamide, etoposide, and cisplatin. Blood 1992; 79: 3388-3393, MEDLINE

4 Leibundgut K, Hirt A, Ridolfi Lüthy A et al. Autotransplants with peripheral blood stem cells and clinical results obtained in children: a review. Eur J Pediatr 1993; 152: 546-554, MEDLINE

5 Lane TA, Law P, Maruyama M et al. Harvesting and enrichment of hematopoietic progenitor cells mobilized into the peri-pheral blood of normal donors by granulocyte-macrophage colony-stimulating factor (GM-CSF) or G-CSF: potential role in allogeneic marrow transplantation. Blood 1995; 85: 275-282, MEDLINE

6 Bensinger WI, Clift R, Martin P et al. Allogeneic peripheral blood stem cell transplantation in patients with advancedhematologic malignancies: a retrospective comparison withmarrow transplantation. Blood 1996; 88: 2794-2800, MEDLINE

7 Champlin R, Schmitz N, Horowitz M. Allogeneic blood stem cell (BSC) vs. bone marrow transplantation (BMT): improved leukemia-free survival in high risk but not low risk patients. Blood 1998; 92: (Suppl.1) 2710a,

8 Lickliter JD, DeFor T, Wagner JE et al. Matched-pair analysis of peripheral blood stem cells compared to marrow for allo-geneic transplantation. Blood 1998; 92: (Suppl.1) 561a,

9 Ottinger HD, Beelen DW, Sheulen B et al. Improved immune reconstitution after allotransplantation of peripheral blood stem cells instead of bone marrow. Blood 1996; 88: 2775-2779, MEDLINE

10 Thomas ED, Storb R, Clift RA et al. Bone marrow transplantation. New Engl J Med 1975; 292: 895-902, MEDLINE

11 Minero R, Busca A, Bonetti F et al. Allogeneic transplantation of peripheral blood progenitor cells in children: experience of two pediatric centers. Bone Marrow Transplant 1998; 22: (Suppl.5) S33-S36, MEDLINE

12 Diaz MA, Alegre A, Villa M et al. Allogeneic peripheral blood progenitor cell (PBPC) transplantation in children with haematological malignancies. Br J Haematol 1997; 96: 161-164, MEDLINE

13 Schriber JR, Chao NJ, Long GD et al. Granulocyte colony-stimulating factor after allogeneic bone marrow transplantation. Blood 1994; 84: 1680-1684, MEDLINE

14 Ringden O, Horowitz MM, Sondel P et al. Methotrexate, cyclosporine or both to prevent graft-versus-host disease after HLA-identical sibling bone marrow transplants for early leukemia? Blood 1993; 81: 1094-1101, MEDLINE

15 Urbano-Ispizua A, Garcia-Conde J, Brunet S et al. High incidence of chronic graft-versus-host disease after allogeneic peri-pheral blood progenitor cell transplantation. The Spanish Group of allo-PBPCT. Haematologica 1997; 82: 683-689, MEDLINE

16 Storek J, Gooley T, Siadak M et al. Allogeneic peripheral blood stem cell transplantation may be associated with a high risk of chronic graft-versus-host disease. Blood 1997; 90: 4705-4709, MEDLINE

17 Horowitz MM, Gale RP, Sondel PM et al. Graft-versus-leukemia reactions after bone marrow transplantation. Blood 1990; 75: 555-562, MEDLINE

Figure 1 Probability of event-free survival.

Table 1  Demographics and engraftment

Table 2  GVHD and outcome