Case Report Autologous Transplantation of CD133 Selected Hematopoietic Progenitor Cells in a Pediatric Patient with Relapsed Leukemia

Bone Marrow Transplantation (2002) 29, 927–930  2002 Nature Publishing Group All rights reserved 0268–3369/02 $25.00 www.nature.com/bmt Case report Autologous transplantation of CD133 selected hematopoietic progenitor cells in a pediatric patient with relapsed leukemia

U Koehl1, S Zimmermann1, R Esser1,JSo¨rensen1,HPGru¨ttner1, M Duchscherer2, E Seifried2, T Klingebiel1 and D Schwabe1

1University Hospital Frankfurt, Department of Pediatric Hematology and Oncology, Frankfurt, Germany; and 2Institute for Transfusion Medicine and Immunohematology, RCBDS Hessia, Frankfurt, Germany

Summary: stem cell glycoprotein antigen selectively expressed on CD34bright progenitor cells.5 Here, we report on the appli- A pediatric patient with very early meningeal relapse cation of this new selection strategy for purification of an of his CD34+ CD133؊ pre-B-ALL was transplanted with autologous graft and transplantation of a boy with relapsed .؋ 106/kg CD133 selected autologous progenitor cells. CD34-positive CD133-negative ALL 2.5 Enrichment of CD133+ cells resulted in a purity of CD133+. Hematopoietic engraftment with 3.5% ؎ 92.3 Ͼ1.0 ؋ 109/l neutrophils and Ͼ50 ؋ 109/l platelets was Patients and methods reached within 13 and 24 days, respectively. At a follow- 1 up of 112 months after autologous transplantation, the Case report patient is in complete remission. To our knowledge, the successful transplantation with a CD133 selected graft An 11-year-old boy with a pre-B-ALL suffered very early is the first one to be reported worldwide. CD133 selected isolated meningeal relapse. Initially, he had been treated cells may serve as an alternative in the case of CD34+ with polychemotherapy according to ALL-BFM 95 malignancy. (German/Austrian multicenter study) from July 1999 to Bone Marrow Transplantation (2002) 29, 927–930. DOI: February 2000. At that time, flow cytometric 10.1038/sj/bmt/1703558 immunophenotyping of the leukemic blasts showed positi- Keywords: progenitor cells; transplantation; CD133+ vity for antigenes CD19, CD20, CD24, CD22, cyCD22, positive selection CD10, CD34, HLA-DR but negative for CD133. Moreover, the blasts were negative for the fusion transcripts TEL- AML1, BCR-ABL and MLL–1 as determined by PCR. Fourteen months later, isolated meningeal relapse was con- Pediatric patients with high-risk of relapse of ALL are fre- firmed by magnetic resonance imaging and lumbar and quently treated with transplantation of allogeneic bone mar- bone marrow puncture. From October 2000, the boy was row (BM) or peripheral blood stem cells (PBSC). However, treated according to ALL-BFM-Rez 96 including intrathe- despite allogeneic transplantation, prognosis for patients cal application of methotrexate, cytosine arabinoside and with a very early meningeal relapse of leukemia remains prednisolone and intravenous application of dexame- poor. Few case reports suggest a better outcome in these thasone, methotrexate, vincristine and PEG-asparaginase. patients after autologous transplantation.1 It has been shown For the poor prognosis of both conventional chemotherapy that clonogenic tumor cells within the graft can cause and allogeneic transplantation in this case, we decided on relapse of the disease.2 To avoid relapse from reinfusion of high-dose chemotherapy and autologous transplantation. grafts, CD34 selection of stem cells is a well-established strategy to remove contaminating tumor cells from autografts in CD34-negative solid tumors of pediatric patients.3 Autologous stem cell harvesting To reach high purity of the stem cell transplant, refined Autologous peripheral blood stem cells (PBSC) were har- selection strategies are needed. In order to eliminate poten- vested after five courses of chemotherapy. Course 5 of the tially CD34-positive tumor cells from autologous grafts, we ALL-BFM-Rez 96 protocol consisted of oral application of established a clinical scale CD133 selection procedure for dexamethasone (20 mg/m2/day) and mercaptopurine 4 pediatric patients with neuroblastoma. CD133 is a novel (100 mg/m2/day) from day 1 to day 5; vincristine (i.v. 1.5 mg/m2/day) on days 1 and 6, methotrexate (1 g/m2;36h infusion) on day 1, cytosine arabinoside (2 g/m2/twice a Correspondence: U Koehl, University Hospital Frankfurt, Department of 2 Pediatric Hematology and Oncology, Theodor Stern Kai 7, 60596 Frank- day, 3 h infusion) on day 5, PEG-asaparaginase (500 U/m ; furt, Germany 1 h infusion) on day 6, and intrathecal application of metho- Received 28 August 2001; accepted 5 March 2002 trexate (12 mg) and cytosine arabinoside (30 mg) and Transplantation of CD133 selected hematopoietic progenitor cells U Koehl et al 928 prednisolone (10 mg) on day 1. From 2 days after the end the current patient status. For this, cells were of chemotherapy to the last day of harvest, stem cells were FITC/PE/ECD/PC5-labeled with CD45/CD4/CD8/CD3, mobilized with s.c. 5 ␮g/kg rhG-CSF (Neupogen; Amgen, CD45/CD56/CD19/CD3, HLA-DR/CD38/CD3/CD8, Munich, Germany). At the time of apheresis the patient HLA-DR/CD38/CD3/CD4, CD45RA/CD45RO/CD3/CD4, weighed 40 kg. PBSC I, II, III was collected on days 9, 10 CD45RA/CD45RO/CD3/CD8 and CD4/CD69/CD3/CD8. and 11 of mobilization using a Cobe Spectra (Cobe Labora- CD133–1-PE and CD133–2-PE were obtained from tories, Lakewood, CO, USA). Miltenyi/Biotec; all other antibodies were from Coulter Immunotech (Marseille, France). Dead cells were detected CD133+ selection and excluded from analysis by 7-AAD and propidium- iodide (PI) staining. Additionally, IgG1-FITC/IgG1- CD133+ cells were immunomagnetically selected from PE/IgG1-ECD/IgG1-PC5 and IgG1-FITC/IgG1-PE/CD45- aphereses using the automated CliniMacs system ECD/IgG1-PC5 staining served as controls. (Miltenyi/Biotec, Bergisch-Gladbach, Germany). Cells were processed as described for CD34 selection4,6 under GMP conditions. Briefly, both the pooled PBSC I+II and PBSC III in a separated run were washed twice with Clini- Results Macs PBS/EDTA buffer (Miltenyi/Biotec) supplemented with 0.4% human serum albumine (BSD Hessia, Frankfurt, + Germany) for platelet reduction and cells were resuspended Collection and positive selection of CD133 cells to a total volume of 80 ml. To avoid nonspecific binding, cells were incubated with 5 ml intraglobine (50 ␮g/ml; Three leukaphereses were performed. The number of leuko- 10 + Biotest, Frankfurt, Germany) for 5 min. Cells were then cytes collected was 4.66 ϫ 10 CD45 cells with a total 6 + 6 + labeled with anti-CD133 antibodies (CD133+ microbeads, number of 299 ϫ 10 CD34 cells (7.5 ϫ 10 CD34 /kg) ϫ 6 + ϫ 6 + kindly provided by Miltenyi/Biotec) for 30 min at 4°C. and 226 10 CD133 cells (5.7 10 CD133 /kg). After 6 + After incubation, cells were washed twice with CliniMacs cryopreservation of 80 ϫ 10 CD34 cells as an unselected PBS/EDTA buffer by centrifugation at 1100 g for 15 min. back up product, CD133 selection in two separate runs led ϫ 6 + ϫ 6 Finally, CD133 positive selection was performed using the to a total number of 105 10 CD34 cells (2.6 10 CD34+/kg) and 106 ϫ 106 CD133+ cells (2.6 ϫ 106 software programme ‘enrichment 2.1’ on the CliniMacs + + + system. CD133 /kg). The percentages of CD34 and CD133 cells in the different PBSCs prior to and after CD133 selection are shown in Table 1. Conditioning regimen and ASCR Enrichment of progenitor cells by CD133 selection from Ϫ Ϫ the pooled PBSC I+II and from PBSC III resulted in a The patient received TBI on days 7, to 4 (12 Gy total + + body, 18 Gy cranial) and etoposide on day Ϫ3 (60 mg/kg). purity of 92.3 Ϯ 3.5% CD133 and 91.3 Ϯ 4.6% CD34 Seventy-two hours after the completion of the radio-chemo- cells. Examplarily, the percentages of progenitor cells in therapy, cryopreserved stem cells were thawed rapidly and PBSC I and after CD133 selection in the pooled PBSC I a total of 2.5 ϫ 106 CD133+/CD34+ cells/kg were reinfused. and II are shown in Figure 1. Prior to selection, 71% of the CD34+ cells were also positive for CD133+. After CD133 enrichment Ͼ99% of Flow cytometric analysis the CD34+ cells and Ͼ97% of the CD133+ cells were CD34+CD133+ cells. Moreover, purification led to an Three- and four-color flow cytometric analyses were perfor- Ϫ + med on a Coulter Epics XL (Coulter, Krefeld, Germany) increased frequency of CD34 CD133 cells (0.9% and to determine the percentage of leukocytes expressing 2.6%). CD133+ and CD34+ in PBSC and CD133+ selected pre- Viability of the cells was Ͼ97% prior to and after purification. Recovery of CD133+ hematopoietic progenitor cells parations. Moreover, we searched for residual leukemic + blasts. After transplantation, we monitored the reconsti- resulted in 63.9 Ϯ 5.5% while the recovery of CD34 cells tution of immunological subtypes in the peripheral blood. was 48.1 Ϯ 4.2%, respectively (Table 1). Briefly, 2 ϫ 105–1 ϫ 106 cells were labeled for 20 min in separate tubes with CD34-FITC/CD133-PE/CD45-PC5 for determination of progenitor cells. Also, Detection of residual leukemic cells CD45FITC/CD34PE/7-AAD including stem-count beads (Stem-Kit/Coulter Immunotech and 7-AAD) was used to Due to nonspecific staining of membrane particles, which evaluate the absolute count of CD34+ cells according to a cannot be excluded from analysis by 7-AAD or propidium- modified ISHAGE protocol. For detection of residual leu- iodide, flow cytometric analysis did not allow for kemic cells prior to and after CD133 selection, cells were definitive exclusion of residual leukemic blasts labeled with CD19-FITC/CD10-PE/CD45-ECD/CD34- (CD45+CD34+CD19+CD10+) in apheresis products I, II, III PC5, CD19-FITC/CD133-PE/CD45-ECD/CD34-PC5 as and one selected graft (in both cases, Ͻ0.02%). The second well as CD19-FITC/CD34-PE/PI. Immune reconstitution purified graft stained negatively. Molecular genetic screen- was monitored in T cell subsets, NK cells and B cells in ing for blasts was not possible, since the leukemic blasts weekly peripheral blood samples during the first 100 days were negative for the fusion transcripts TEL-AML1, BCR- post transplant, then monthly samples or according to ABL and MLL–1 determined by PCR.

Bone Marrow Transplantation Transplantation of CD133 selected hematopoietic progenitor cells U Koehl et al 929 Table 1 Number of collected cells, percentage of CD34+ and CD133+ cells before and after CD133 selection and resulting recovery after puriﬁcation

Apheresis product No of CD34+ cells CD133+ cells CD34+CD133+ CD34+ cells CD133+ cells Recovery of Recovery of (PBSC) nucleated cells of total CD45+ of total CD45+ of total CD45+ (ϫ106/kg) (ϫ106/kg) CD34+ cellsb CD133+ cellsb (ϫ106/kg) (%) (%) (%) (%) (%)

Before enrichment PBSC I 485 0.69 0.52 0.50 3.4 2.6 PBSC II 393 0.65 0.49 0.46 2.6 1.9 PBSC III 288 0.52 0.40 0.35 1.5 1.2 After enrichment 1 + a 2 PBSC I PBSC II 1.98 94.5 94.8 93.9 1.87 1.87 45.1 60.0 PBSC III 0.87 88.0 89.8 87.2 0.76 0.78 51.1 67.8

Each value is the mean of three independent determinations as described in Patients and methods. a80 ϫ 106 CD34+ cells (2 ϫ 106 CD34+/kg) were taken from PBSC I for cryoconservation of an unselected back up product. The ﬁnal half part of PBSC I (9/20 of PBSC I) was pooled with PBSC II and selected by CD133 enrichment. PBSC III was puriﬁed in an independent run. bRecovery for CD133+ and CD34+ cells was determined dividing the absolute count of the respective cells after selection by that before selection.

AB were treated with rhG-CSF for 14–15 days after autolog- 1000 1 2 1000 1 2 ous transplantation. Monitoring of different immunological subtypes is illus- trated in Figure 3. CD19+ B cell recovery started as early as 1.5 months after transplantation. CD56+CD16+CD3Ϫ

A B natural killer cells (NK) increased early and reached normal

CD34 FITC CD34 CD34 FITC CD34 3 4 3 4 levels within 1 month. No marked decrease in the absolute + + 0.1 0.1 number of CD4 T helper cells was observed and CD8 0.1CD133 PE 1000 0.1 CD133 PE 1000 cytotoxic cells reconstituted fairly rapidly resulting in a Figure 1 Flow cytometric assessment of progenitor cells. The percentage slightly inversed CD4/CD8 ratio, only. CD45 isoforms con- of CD133+CD34+ and CD133+CD34Ϫ cells was determined in PBSC I ﬁrmed rapid recovery of memory cytotoxic T cells prior to (A) and after CD133 selection of pooled PBSC I and II (B). The (CD45RO+CD8+CD3+) and naive cytotoxic T cells lower ﬂuorescence intensity for CD133 after selection as compared to the (CD45RA+CD8+CD3+). By contrast, the naive helper T apheresis product is explained by the use of the same antibody for both + + + selection and phenotyping. cells subset (CD45RA CD4 CD3 ) was somewhat reduced (not shown in Figure 3). No striking difference to other patients was observed in the activation markers on the Hematopoietic recovery various T cell subsets. Engraftment was seen on days 13, 13 and 24 for leukocytes (Ͼ1.0 ϫ 109/l), neutrophils (Ͼ1.0 ϫ 109/l), and platelets Clinical status after transplantation (Ͼ50.0 ϫ 109/l), respectively. This is comparable with the + Infusion of CD133 selected progenitor cells was well toler- engraftment after reinfusion of CD34 selected cells as ated. The patient did not show any allergical reactions or shown for the absolute number of neutrophils count (ANC) side-effects due to the remaining CD133 microbeads on the of some of our pediatric patients (four neuroblastoma, one surface of the progenitor cells. No viral or fungal infection Ewing’s sarcoma, one osteosarcoma, one rhabdomyosar- or other kind of complication during and after hemato- coma; Figure 2). These patients were transplanted with a + poietic recovery was seen. No additional late toxicities have similar dose of autologous CD34 stem cells (1.8–4.0 ϫ 1 been observed. With a follow-up of 112 months after trans- 106/kg). Generally, to accelerate recovery, all our patients plantation, the boy is alive with a good performance status in continuous complete remission. 100 000 CD3+ 800 10 000 CD4+CD3+ 700 CD8+CD3+ /l CD8+CD3+CD45RA+ 6 1000 600 CD8+CD3+CD45RO+

/l 6

× CD8+CD3+HLA-DR+ 500

10 CD8+CD3+CD38+

100 × 400 CD8+CD3+CD69+ ANC CD56+CD3- CD133 selected cells 300 CD19+ 10 Cells CD34 selected cells 200 1 100 0 5 10 15 20 25 30 0.0 Time (days) 0 20 40 60 80 100 120 140 Time (days) Figure 2 Hematopoietic recovery of the absolute number of neutrophil counts (ANC) after transplantation with CD133 selected progenitor cells Figure 3 Reconstitution of immunological subtypes in the peripheral in comparison to CD34 selected cells. blood after transplantation with a CD133 selected autologous graft.

Bone Marrow Transplantation Transplantation of CD133 selected hematopoietic progenitor cells U Koehl et al 930 Discussion ible and achieves equivalent high purities of progenitor cells equivalent to that previously reported for CD34 selec- Very early leukemic relapse in the CNS continues to be a tion. Tumor and leukemic cell depletion is effective with difficult therapeutic challenge. Intensification of chemo- this method. CD133 selected grafts can lead to full hemato- therapy prior to allogeneic transplantation did not improve poietic reconstitution. Therefore, CD133 purification prognosis. Because a few cases are reported with similar or presents a favorable alternative to CD34 selection of auto- better outcome after autologous transplantation, high-dose logous grafts in patients with CD34-positive but CD133- chemotherapy with reinfusion of autografts seems to be an negative malignant cells. alternative worth exploring for these patients.1 For autologous grafts, a refined selection technique is indispensable to yield a high purity of progenitor cells and to avoid a relapse Acknowledgements due to residual leukemic contamination. For pediatric 2 This project was supported by ‘Hilfe fu¨r Krebskranke Kinder patients with neuroblastoma, Brenner et al showed direct Frankfurt eV’,by‘Frankfurter Stiftung fu¨r Krebskranke Kinder’ evidence that neuroblastoma cells present in patients’ bone and by ‘Paul und Ursula Klein-Stiftung’. We acknowledge the marrow or peripheral blood stem cell grafts can contribute excellent technical support of Andrea Brinkmann, Ilse Bu¨hler and to relapse. The commonly used CD34 selection is an excel- Stephanie Grohal. lent technique to achieve a high purity of CD34 cells.7 In various trials, clinical scale CD34 selection led to a median purity of 96–97% CD34+ cells.4,6,7 However, the use of a References CD34 enrichment technique is not to be recommended in cases of CD34 positive malignancies. Rather, enrichment 1 Messina C, Valsecchi MG, Arico M et al. Autologous bone with the new progenitor antigen CD133 may present a pref- marrow transplantation for treatment of isolated central ner- erable purification strategy in selected cases. Several stud- vous system relapse of childhood acute lymphoblastic leuke- ies demonstrated that the majority of ALLs are double posi- mia. Bone Marrow Transplant 1998; 21:9–14. 8,9 2 Brenner MK, Rill DR, Moen RC et al. Gene-marking to trace tive for CD133 and CD34. Therefore, purification with origin of relapse after autologous bone marrow transplan- CD133 can be used for a minority of ALLs, only. tation. Lancet 1993; 341:85–86. In our previous work, we compared large scale CD34 3 Handgretinger R, Lang P, Schumm P et al. Isolation and trans- and CD133 purification of autologous grafts of pediatric plantation of autologous peripheral CD34+ progenitor cells patients with neuroblastoma.4 Both selection strategies led highly purified by magnetic-activated cell sorting. Bone Mar- to similar high median purities of 98.0% and 97.3% CD34+ row Transplant 1998; 21: 987–993. 4 Koehl U, Esser R, Zimmermann S et al. Large scale purifi- cells. No contaminating neuroblastoma cells remained as + determined by RT-PCR for tyrosine hydroxylase. These cation of progenitor cells by AC133 selection. Bone Marrow patients were transplanted with CD34 selected progenitor Transplant 2001; 27: 296 (Abstr. P736). cells, only. Experimentally data demonstrate a higher proli- 5 Yin AH, Miraglia S, Zanjani ED et al. AC133, a novel marker for human hematopoietic stem and progenitor cells. Blood ferating and self-replicating capacity of CD133+CD34+ Ϫ + 1997; 12: 5002–5012. cells in comparison to CD133 CD34 cells. The latter con- 6 Koehl U, Gunkel M, Gruettner HP et al. Positive selection of tain committed progenitors, which do not have full reconsti- hematopoietic progenitor cells for autologous and allogeneic tuting capacity.5,10–12 An extremely rare population of cells transplantation in pediatric patients with solid tumors and leu- lacking CD34 and lineage commitment markers kemia. Transplant Hematol Oncol 1999; 1: 159–168. (CD34ϪlinϪCD38Ϫ) is positive for CD133 and shows a 7 Schumm M, Lang P, Taylor G et al. Isolation of highly pur- very high reconstituting ability.12 This gives rise to the ified autologous and allogeneic peripheral CD34+ cells using thought that CD133 selected cells might be superior for the CliniMACS device. J Hematother 1999; 6:5–11. stem cell transplantation to CD34 selected cells. We saw 8 Barsch G, Baumann M, Ritter J et al. Expression of AC133 that CD133 selection led to enrichment of this rare and CD117 on candidate normal stem cell populations in childhood B-cell precursor acute lymphoblastic leukemia. Br CD133+CD34Ϫ progenitor cell population together with the + + J Haematol 1999; 107: 572–580. CD133 CD34 double positive cells. Therefore, fewer pro- 9Bu¨hring HJ, Seiffert M, Marxer A et al. AC133 antigen is not genitor cells in the graft might be sufficient to reach full restricted to acute myeloid leukemic blasts but is also found reconstitution. However, hematopoietic recovery of our on acute lymphoid leukemia blasts and on subset of CD34+ patient reported here did not differ significantly from that of B-cell precursors. Blood 1999; 2: 832–833. our patients receiving the same number of conventionally 10 Kobari L, Giarratana MC, Pflumic F et al. CD133 cell selec- selected stem cells. Additionally, reconstitution of the tion is an alternative to CD34+ cell selection for ex vivo immunological subsets did not show remarkable differences expansion of hematopoietic stem cells. J Hematother Stem to patients treated with CD34-enriched grafts. To date, the Cell Res 2001; 10: 273–281. literature lacks any data about transplantation with CD133 11 Matsumoto K, Kazuta Y, Yamashita N et al. In vitro prolifer- ation potential of AC133 positive cells in peripheral blood. enriched progenitors. Therefore, to our knowledge, the suc- Stem Cells 2000; 18: 196–203. cessful transplantation with a CD133 selected graft of our + Ϫ 12 Gallacher L, Murdoch B, Wu DM et al. Isolation and charac- patient with a CD34 CD133 ALL is the first one to be terization of human CD34ϪLinϪ and CD34+LinϪ hemato- reported worldwide. poietic stem cells using cell surface markers AC133 and CD7. In summary, large scale CD133 positive selection is feas- Blood 2000; 9: 2813–2820.

Bone Marrow Transplantation