(Filgrastim) on Bone Marrow Progenitor Cells

Home , Filgrastim, Peripheral blood cell

Bone Marrow Transplantation, (1999) 23, 15–19  1999 Stockton Press All rights reserved 0268–3369/99 $12.00 http://www.stockton-press.co.uk/bmt Effects of short-term administration of G-CSF (ﬁlgrastim) on bone marrow progenitor cells: analysis of serial marrow samples from normal donors

C Martı´nez, A Urbano-Ispizua, M Rozman, M Rovira, P Marı´n, N Montfort, E Carreras and E Montserrat

Hematology Department, Postgraduate School of Hematology Farreras-Valentı´, Institut d’Investigacions Biome`diques August Pi i Sunyer, Hospital Clı´nic, University of Barcelona, Spain

Summary: cells.1–4 A large amount of information about the effects of cytokines on the mobilization of progenitor cells in periph- To determine the effect of G-CSF administration on eral blood is available.5–10 Thus, it is well established that both the total number of CD34؉ cells and the primitive 4 to 5 days of G-CSF administration to patients or normal -CD34؉ subsets in bone marrow (BM), we have analyzed donors effectively increases the absolute number of circul BM samples serially obtained from 10 normal donors ating CD34+ cells, allowing the collection of sufficient in steady-state and during G-CSF treatment. Filgrastim quantities of progenitor cells to ensure sustained was administered subcutaneously at a dosage of 10 engraftment after transplantation.6–9 Data concerning the ␮g/kg/day (n = 7) or 10 ␮g/kg/12 h (n = 3) for 4 consecu- effects of G-CSF on bone marrow (BM), however, are tive days. Peripheral blood sampling and BM aspirates scarce. In fact, it has not been elucidated whether the PBSC were performed on day 1 (just before G-CSF mobilization schedules currently in use result in the simul- administration), day 3 (after 2 days of G-CSF), and day taneous enrichment of stem and progenitor cells in BM, or 5 (after 4 days of G-CSF). During G-CSF adminis- simply in a shift of these cells from BM to peripheral blood. tration, a significant increase in the total number of BM Some clinical trials have recently been conducted to deter- nucleated cells was observed. The percentage (range) of mine the progenitor cell content and engraftment capability -CD34؉ cells decreased in BM from a median of 0.88 of BM harvested after G-CSF administration, with contro (0.47–1.44) on day 1 to 0.57 (0.32–1.87), and to 0.42 versial results.11–16 The main aim of this study was to ana- (0.16–0.87) on days 3 and 5, respectively. We observed lyze, in a population of normal donors, the effects of G- ,a slight increase in the total number of BM CD34؉ cells CSF administration on the number of BM progenitor cells on day 3 (0.66 × 109/l (0.13–0.77)), and a decrease on with special emphasis on the immature cell subsets. day 5 (0.23 × 109/l (0.06–1.23)) as compared with steady-state (0.40 × 109/l (0.06–1.68)). The proportion of primitive BM hematopoietic progenitor cells Materials and methods (CD34؉CD38؊, CD34؉HLA-DR؊, CD34؉CD117؊) decreased during G-CSF administration. In parallel, a Donors and G-CSF administration significant increase in the total number of CD34؉ cells in peripheral blood was observed, achieving the Peripheral blood and BM samples were obtained from 10 maximum value on day 5. These results suggest that in healthy donors included in a program of allogeneic PBSC normal subjects the administration of G-CSF for 5 days transplantation. Donors were thoroughly informed about may reduce the number of progenitor cells in BM, the investigative nature of the procedure and gave their particularly the most primitive ones. written consent. The protocol and consent forms were Keywords: G-CSF; bone marrow CD34+ cells; normal approved by the Hospital Clinic Ethics Committee and by donors the Spanish Ministry of Health. Median age of the donors was 48 years (range 20–57 years). Recombinant human G- CSF (Filgrastim; Amgen, Thousand Oaks, CA, USA) was administered subcutaneously at a dosage of 10 ␮g/kg/day Peripheral blood stem cells (PBSC) are being increasingly (n = 7) or 10 ␮g/kg/12 h (n = 3) for 4 consecutive days. used for both autologous and allogeneic transplantation. Bone marrow aspirates from the sternum were performed Numerous studies have shown that mobilized PBSC result under local anesthesia on day 1 (just before G-CSF in a more rapid hematopoietic reconstitution after mye- administration), on day 3 (after 2 days of G-CSF), and on loablative therapy as compared with bone marrow stem day 5 (after 4 days of G-CSF). The samples were collected with EDTA. In order to minimize peripheral blood cell con- Correspondence: Dr C Martıńez, Hematology Department, Hospital tamination, no more than 1–1.5 ml of BM were obtained Clinic, Villarroel 170, 08036 Barcelona, Spain from each aspiration. Peripheral blood samples were col- Received 17 June 1998; accepted 10 August 1998 lected by venipuncture from each donor on days 0, 3 and 5. Short-term administration of filgrastim on BMPC C Martıńez et al 16 Quantification and phenotyping of CD34+ cells median (range) of 1.7 (0.6–5.5) in steady-state to 1.9 (0.8– Bone marrow samples were collected with EDTA and 7.2), and 7.5 (3.2–9.3) on day 3 and 5, respectively. Abnor- diluted 1:2 in phosphate-buffered saline (PBS). Cell counts mal cytoplasmic vacuolization of promyelocytes was found were performed on the whole BM and peripheral blood on days 3 and 5 in seven cases. using an automated cell counter (Sysmex F-800; Baxter Fenwal, Deerfield, IL, USA). CD34 quantification was per- T lymphocytes and NK cells in bone marrow formed on unfractioned peripheral blood and BM using The relative numbers of CD3+ T lymphocytes and FACS Lysing Solution (Becton Dickinson Immunocytome- + + try Systems (BDIS), San Jose, CA, USA) and staining with CD3 CD56 NK cells decreased from a median (range) of the CD34-specific phycoerythrin (PE)-conjugated mono- 7.7% (4–21) and 1.65% (1.26–3.26) on day 1 to a median clonal antibody (MAb) 8G12 (HPCA-2-PE) (BDIS). The of 3.7% (2.4–6.5) and 0.9% (0.3–1.65) on day 5, respect- following MAbs were obtained from BDIS: FTIC-labeled ively. The total number of T and NK cells did not change. and PE-labeled 8G12 against CD34, FTIC-labeled MAbs against CD3 (Leu-4), and CD19 (Leu-12), PE-labeled Nucleated and CD34+ cells in bone marrow MAbs against CD56 (Leu-19), CD33 (anti-Leu-M9), CD38 (anti-Leu-17), and HLA-DR (anti-HLA-DR). The PE-lab- An increase in the total number of nucleated BM cells was observed during G-CSF administration, from a median eled MAb against CD117 was obtained from Immunotech × 9 = SA (Marseilles, France). The Simultest control (IgG1 FTIC 10 /l (range) of 41.5 (13.5–104) to 48 (35.5–120.8) (P 0.06) on day 3, and 56 (36.4–176) (P = 0.02) on day 5. At + IgG2a PE) was purchased from BDIS. Flow cytometry + was performed on a FACScan (BDIS) equipped with an the same points in time, the percentage (range) of CD34 air-cooled argon ion laser tuned to 488 nm. Lysis II cells decreased from a median of 0.88 (0.47–1.44) to 0.57 (0.32–1.87) (P = 0.09) and to 0.42 (0.16–0.87) (P = 0.01) software was used for data acquisition and analysis. For × 9 + the purpose of CD34+ cell analyses, a gate for viable cells (Table 2). The total number 10 /l (range) of BM CD34 excluding erythrocytes and debris was set according to for- cells increased slightly on day 3 (0.66 (0.13–0.77)), and ward (FSC) and side light scatter (SCC). The CD34+ cells decreased on day 5 (0.23 (0.06–1.23)) as compared with were analyzed in a fluorescence vs SCC plot. Thus, only basal values (0.40 (0.06–1.68)) (Figure 1). These changes cells with low SCC were counted as CD34+ cells. To obtain were not statistically significant, probably due to the small sufficient CD34+ cells for analysis of CD34+ subsets, an number of cases included in the study. acquisition gate was set according to SSC and fluorescence intensity range comprising the cells with positive CD34 Nucleated and CD34+ cells in peripheral blood fluorescence signals. A minimum of 300 000 events were × 9 run through the cytometer per test, and 3000–5000 CD34+ The peripheral WBC count increased from a median 10 /l (range) of 5.2 (3.2–13.6) before G-CSF administration to cells within the acquisition gate were stored in listmode = = data files for further analysis. 25.3 (23.8–37.8) (P 0.02) and 39.4 (21.0–69.2) (P 0.01) on days 3 and 5 of cytokine administration, respectively. The percentage of CD34+ cells rose from a median (range) Statistics of 0.002 (0.001–0.003) to 0.023 (0.007–0.030) (P = 0.06) Median values and ranges for each parameter were determ- on day 3, and 0.115 (0.028–1.62) (P = 0.01) on day 5 ined. The Wilcoxon test (paired, nonparametric) was (Table 2). This resulted in a significant increment in the performed to compare the results of the different total number of CD34+ cells in the peripheral blood, the subpopulations. maximum value being achieved on day 5 (Figure 1).

+ Results CD34 cell subsets We analyzed CD34+ cell subsets from the BM in steady- Overall bone marrow cellularity state and during G-CSF administration, and from the per- Bone marrow cellularity was highly enriched in myeloid ipheral blood on day 5. Since CD34+ cells were nearly cells during G-CSF administration. An increase in both undetectable in peripheral blood in steady-state, no accurate early (myeloblasts and promyelocytes) and late myeloid analysis of the changes in CD34+ cell subsets in peripheral precursors (myelocytes and metamyelocytes) was observed blood could be done. FACS analysis showed that more than (Table 1). The myeloid:erythroid ratio changed from a 90% of CD34+ cells in both compartments coexpressed the CD38 and HLA-DR antigens at baseline and during cyto- Table 1 Effect of G-CSF administration on bone marrow cellularity kine treatment. The proportion of primitive BM hematopoietic progenitors identified as CD34+CD38Ϫ, CD34+HLA- Promyelocytes Myelocytes + Neutrophils DRϪ and CD34+CD117Ϫ decreased during G-CSF adminis- metamyelocytes tration (Figure 2). No significant changes were observed in the proportion of BM CD34+CD33+ and CD34+CD19+ cells Day 1 6 (3–11) 20 (11–30) 17 (11–40) during cytokine administration. The mean fluorescence Day 3 21 (10–23) 16 (12–33) 17 (2–64) + Day 5 26 (21–39) 35 (24–54) 15 (7–28) intensity of CD34 cells, used as a parameter for antigen density, was higher in BM CD34+ cells in steady-state as Values are expressed as percentage of total cellularity. compared with those from days 3 and 5 (Figure 3). Short-term administration of filgrastim on BMPC C Martıńez et al 17 Table 2 Total nucleated cells, percentage of CD34+ cells and total CD34+ cells in bone marrow and peripheral blood before and during G-CSF administration

Total nucleated cells ×109/l % CD34+ CD34+ cells ×109/l

Bone marrow day 1 41.4 (13.5–104) 0.88 (0.47–1.44) 0.40 (0.06–1.68) day 3 48 (35.5–120.8) 0.57 (0.32–1.87) 0.66 (0.03–0.77) day 5 56 (36.4–176) 0.42 (0.16–0.87) 0.23 (0.06–1.23) Peripheral blood day 1 5.16 (3.2–13.6) 0.04 (0.01–0.07) 0.002 (0.001–0.003) day 3 25.3 (23.8–37.8) 0.08 (0.03–0.09) 0.023 (0.007–0.030) day 5 39.4 (21–69.2) 0.16 (0.07–1.62) 0.115 (0.028–1.620)

Values are expressed as median and range.

1.0 200 BM a 100 0.8 PB CD34 PB /l 9 150 60 10

× 0.6 3.0 + cells

cells 0.4 100 2.5 +

× 2.0 10

0.2 % 6

50 /l 1.5 BM CD34 0.0 1.0 0.5 –0.2 0 12345 0.0 day 1 day 3 day 5 Days of G-CSF administration CD34+CD38– Figure 1 Effect of G-CSF administration on bone marrow (BM) and peripheral blood (PB) CD34+ cells. Results are expressed as median Ϯ s.e.m. b 100

Discussion 60 4 Recent studies suggest that G-CSF administration might increase the content of progenitor cells in BM, and change 3 % the biological characteristics of bone marrow stem cells, 2 making them similar to PBSC.11,12,17 These observations are the background for some recent clinical trials aimed at 1 determining the engraftment capability of G-CSF-stimulated BM. Thus, it has been reported that the kinetics of 0 day 1 day 3 day 5 engraftment with G-CSF-stimulated BM could be as rapid as engraftment with PBSC, and superior to conventional CD34+HLADR– autologous bone marrow transplants;11,13–15 other authors, c however, have found no differences after primed BM when 100 compared with historical controls of unprimed BM.12,16 A 70 preliminary report of G-CSF-stimulated BM allogeneic 50 transplantation suggests that engraftment is rapid and stable.18 A high risk of chronic graft-versus-host disease 40 19,20 % has been reported after allogeneic PBSC transplants, 30 probably due to the large number of T lymphocytes con- tained in leukapheresis products. Therefore, the adminis- 20 tration of G-CSF before BM harvesting has been suggested as an alternative to increase the quantity of progenitor cells 10 without a corresponding increase in the number of lympho- 0 cytes. Nevertheless, the effects of G-CSF on the total num- day 1 day 3 day 5 + ber of both progenitor cells and immature CD34 subsets CD34+ckit– in bone marrow have been poorly investigated. Figure 2 Proportion of bone marrow primitive CD34+ subsets (a, In a recent study in mice, a great increase in the number CD34+CD38Ϫ; b, CD34+HLADRϪ; c, CD34+ckitϪ) in steady-state (day of primitive BM hematopoietic progenitor cells after 2 days 1) and during G-CSF administration (days 3 and 5). Data are presented of G-CSF was observed, declining afterwards, to reach as mean Ϯ s.e.m. values. Short-term administration of ﬁlgrastim on BMPC C Martı´nez et al 18 300 ing the optimal timing of G-CSF-stimulated BM collection to guarantee a sustained engraftment. To that end, our study 250 suggests that a brief course of G-CSF could be enough to improve the myeloid engrafting capacity of BM. This 200 would reduce the cytokine exposure time of normal donors and would avoid the possibility of immature progenitor 150 cells migrating to the peripheral blood on day 5. Further studies to establish the optimal timing of G-CSF-stimulated BM collection, and the quality of engraftment and long- 100 term hematopoietic reconstitution in allogeneic primed BM recipients are required. Mean fluorescence intensity 50

0 Acknowledgements day 1 day 3 day 5 Figure 3 Changes in mean fluorescence intensity of CD34+ cells in This work has been supported in part by grants SGR 1996/0068 steady-state (day 1) bone marrow (BM) and during G-CSF treatment from Comissionat per a Universitats i Recerca, Generalitat Catalu- Ϯ (days 3 and 5). Data are presented as mean s.e.m. values. nya, and FIS-98/995 from Fondo de Investigaciones Sanitarias de la Seguridad Social, Spanish Ministry of Health. basal level on days 5–6. Most of the reported progenitor cell expansion in BM occurred prior to the appearance of these cells in peripheral blood.21 We observed a slight References increase in the absolute number of BM CD34+ cells on day 3 of G-CSF administration, whereas CD34+ cells decreased 1 Bensinger W, Singer J, Appelbaum F et al. Autologous trans- below steady-state values on day 5. Although it has been plantation with peripheral blood mononuclear cell collected reported that 5 days of G-CSF administration results in a after administration of recombinant granulocyte stimulating factor. Blood 1993; 81: 3158–3162. higher number of BM CFU-GM and BFU-E, compared 2 Sheridan SP, Begley CG, Juttner CA et al. Effect of periph- with historical controls,11,12 in most studies no significant + eral-blood progenitor cells mobilized by G-CSF (G-CSF) on increase in BM CD34 cell content has been platelet recovery after high-dose chemotherapy. Lancet 1992; observed.15,22,23 Indeed, we found that the proportion of 339: 640–646. primitive BM hematopoietic progenitors (CD34+CD38Ϫ, 3 Bensinger W, Clift R, Martin P et al. Allogeneic peripheral CD34+HLADRϪ and CD34+CD117Ϫ cells) decreased dur- blood stem cell transplantation in patients with advanced hem- ing G-CSF administration, with the lowest value being atologic malignancies: a retrospective comparison with mar- reached on day 5. In this regard, several studies suggest row transplantation. Blood 1996; 88: 2794–2800. that cytokines can shift early repopulating stem cells out of 4 Korbling M, Przepiorka D, Huh YO et al. Allogeneic blood the marrow into the peripheral blood. Thus, Bodine et al24 stem cell transplantation for refractory leukemia and lymphoma: potential advantage of blood over marrow allografts. found that after 5 days of SCF/G-CSF, mice had only 25% Blood 1995; 85: 1659–1665. of the baseline long-term repopulating stem cells in the 5 Gianni AM, Siena S, Bregni M et al. Granulocyte–macro- marrow, and three times the amount of baseline stem cells phage colony-stimulating factor to harvest circulating haema- had shifted to peripheral blood. Mandalam et al25 in a pre- topoietic stem cells for autotransplantation. Lancet 1989; 2: liminary report on normal donors, suggest that the pro- 580–588. portion of CD34+ primitive subsets in BM is lower in 6Du¨hrsen U, Villeval JL, Boyd J et al. Effects of recombinant donors having received G-CSF for 3 days, than in those granulocyte colony-stimulating factor on hemopoietic progeni- treated for 2 days, and in those who received no treatment. tor cells in cancer patients. Blood 1988; 72: 2074–2081. During prolonged G-CSF administration to mice, de Haan 7 Dreger P, Haferlach T, Eckestein V et al. G-CSF-mobilized et al26 also showed a severe depletion of BM stem cells, peripheral blood progenitor cells for allogeneic transplantation: safety, kinetics of mobilization, and composition of the particularly the most primitive ones, that increased in the 27 graft. Br J Haematol 1994; 87: 609–613. spleen. In the clinical setting, Mavroudis et al have 8 Russell JA, Luider J, Weaver M et al. Collection of progenitor recently reported, on a group of 12 patients, three cases cells for allogeneic transplantation from peripheral blood of of incomplete engraftment and four cases that developed normal donors. Bone Marrow Transplant 1995; 15: 25–31. pancytopenia 2 to 3 months following T cell-depleted G- 9 Anderlini P, Przepiorka D, Champlin R et al. Biological and CSF-stimulated BM allogeneic transplant. This suggests clinical effects of granulocyte colony-stimulating factor in normal individuals. Blood 1996; 88: 2819–2825. that after 5 days of G-CSF administration, the BM provided + a sufficient number of engrafting cells which were, none- 10 Siena S, Bregni M, Brando B et al. Circulation of CD34 hem- theless, deficient in self-renewing, long-term repopulating atopoietic stem cells in the peripheral blood of high-dose cells. cyclophosphamide-treated patients: enhancement by intra- venous recombinant human granulocyte–macrophage colony- In conclusion, our study shows that in normal subjects stimulating factor. Blood 1989; 74: 1905–1914. the administration of G-CSF for 5 days may reduce the 11 Stoppa AM, Blaise D, Viens P et al. Phase I study of in vivo number of progenitor cells in BM, particularly the most lenograstim (rHuG-CSF) for stem cell collection demonstrates primitive ones, basically by increasing their mobilization improved neutrophil recovery after autologous BM transplan- into peripheral blood. This poses the question of determin- tation. Bone Marrow Transplant 1994; 13: 541–547. Short-term administration of filgrastim on BMPC C Martıńez et al 19 12 Hansen PB, Knudsen H, Gaarsdal E et al. Short-term in vivo dence of chronic GVHD after allogeneic peripheral blood pro- priming of bone marrow haematopoiesis with rhG-CSF, genitor cell transplantation from matched related donors. rhGM-CSF or rhIL-3 before marrow harvest expands myelo- Haematologica 1997; 82: 683–689. poiesis but does not improve engraftment capability. Bone 21 Morrison SJ, Wright DE, Weissman IL. Cyclophos- Marrow Transplant 1995; 16: 373–379. phamide/granulocyte colony-stimulating factor induces hema- 13 Janssen WE, Smilee RC, Elfenbein GJ. A prospective ran- topoietic stem cells to proliferate prior to mobilization. Proc domized trial comparing blood- and marrow-derived stem Natl Acad Sci USA 1997; 94: 1908–1913. cells for hematopoietic replacement following high-dose 22 Haas R, Mo¨hle R, Pfo¨rsich M et al. Blood-derived autografts chemotherapy. J Hematother 1995; 4: 139–140. collected during granulocyte colony-stimulating factor- 14 Damiani D, Fanin R, Silvestri F et al. Randomized trial of enhanced recovery are enriched with early Thy-1+ hematopoi- autologous G-CSF-primed bone marrow transplantation vs G- etic progenitor cells. Blood 1995; 85: 1936–1943. CSF-mobilized peripheral blood stem cell transplantation in 23 Atkinson K, Fay K, Nivison-Smith I, Downs K. Lenograstim lymphoma patients. Blood 1997; 90: 36–42. administration to HLA-identical donor–recipient pairs to 15 Dicke KA, Deborah LH, Hood L et al. Effects of short-term accelerate marrow recovery post-transplant. Bone Marrow in vivo administration of G-CSF on bone marrow prior to har- Transplant 1997; 19: 15–21. vesting. Exp Hematol 1997; 25: 34–38. 24 Bodine DM, Seidel EN, Orlic D. Bone marrow collected 14 16 Johnsen HE, Hansen PB, Plesner T et al. Increased yield of days after in vivo administration of granulocyte colony-stimul- myeloid progenitor cells in bone marrow harvested for auto- ating factor and stem cell factor to mice has 10-fold more logous transplantation by pretreatment with recombinant repopulating ability than untreated bone marrow. Blood 1996; human granulocyte colony-stimulating factor. Bone Marrow 88: 89–97. Transplant 1992; 10: 229–234. 25 Mandalam RK, Vento CA, Stiff PJ et al. Stem and progenitor 17 Lemoli RM, Tafuri A, Fortuna A et al. Cycling status of cell content and ex vivo expansion potential of human bone CD34+ cells mobilized into peripheral blood of healthy donors marrow primed in vivo with G-CSF. Blood 1997; 90 (Suppl. by recombinant human granulocyte colony-stimulating factor. 2): 344 (Abstr. 4295). Blood 1997; 89: 1189–1196. 26 de Haan G, Dontje B, Engel C et al. The kinetics of murine 18 Isola LM, Scigliano E, Skerrett D et al. A pilot study of allo- hematopoietic stem cells in vivo in response to prolonged geneic bone marrow transplantation using related donors increased mature blood cell production induced by granulo- stimulated with G-CSF. Bone Marrow Transplant 1997; 20: cyte colony-stimulating factor. Blood 1995; 86: 2986–2992. 1033–1037. 27 Mavroudis DA, Read EJ, Molldrem J et al. T cell-depleted 19 Storek J, Gooley T, Siadak M et al. Allogeneic peripheral granulocyte colony-stimulating factor (G-CSF) modified allo- blood stem cell transplantation may be associated with a high geneic bone marrow transplantation for hematological malig- risk of chronic graft-versus-host disease. Blood 1997; 90: nancy improves graft CD34+ cell content but is associated 4705–4709. with delayed pancytopenia. Bone Marrow Transplant 1998; 20 Urbano-Ispizua A, Garcıá-Conde J, Brunet S et al. High inci- 21: 431–440.