High-risk patients with hematological malignancies and solid tumors may benefit from dose-escalated cytotoxic chemotherapy, provided that the tumor cells are chemosensitive as reflected by a dose-response relationship between the cytotoxic drugs administered and the degree of cell kill achieved. Since the first transplantation of peripheral blood stem cells (PBSC) in 1985, mobilized peripheral blood has gradually replaced the bone marrow as a source of hematopoietic stem cells (Körbling et al. 1986; Reiffers et al. 1986; Kessinger et al. 1988; see also Mahon et al., Gisselbrecht, and Goldschmidt et al., this volume). The faster hematological reconstitution after transplantation of PBSC compared to bone marrow ameliorates the dose-limiting myelotoxicity of high-dose regimens and results in a significant reduction of nonhematological toxicity (To et al. 1992; Schmitz et al. 1995). The use of hematopoietic growth factors is essential for efficient mobilization of PBSC (Haas et al. 1990; Hohaus et al. 1993; Haas and Murea 1995; Hohaus et al. 1997a). In this setting, flow cytometry analysis with CD 34 staining permits counting of PBSC in a rapid and reliable fashion (Siena et al. 1991; Haas et al. 1994). It has become apparent that a minimum or threshold number of transplanted CD 34+ cells are needed for rapid and sustained hematological recovery (Hohaus et al. 1993; Haas et al. 1994; Schwartzberg et al. 1993; Weaver et al. 1995; Bensinger et al. 1993).

Contamination of transplants with tumour cells may contribute to relapse after peripheral blood stem cell transplantation (PBSCT). We studied the feasibility of CD34+ cell selection from blood-derived autografts obtained following G-CSF-supported cytotoxic chemotherapy in a group of 25 patients with breast cancer (10 with high-risk stage II/III and 15 with stage IV without bone or bone marrow involvement). Using immunomagnetic beads (Isolex 300 SA. Baxter) CD34+ cells were enriched and released by chymopapain resulting in a median purity of 95% (range 82-99%) and a median recovery of 80% (range 27-132%). The enrichment procedure did not change the proportion of CD34+ subsets coexpressing HLA-DR, CD38 and Thy-1, while L-selectin was removed from the cell surface following selection. Using a sensitive immunocytological technique with a cocktail of epithelial-specific antibodies (anti-cytokeratin 8, 18 and 19; HEA125; BM7 and BM8), five leukaphereses products contained epithelial cells, whereas the selected CD34+ cell fraction was free of tumour cells. A neutrophil count of 0.5 x 10(9)/l and a platelet count of 20 x 10(9)/l was reached after a median time of 14 and 10d following 40 high-dose chemotherapy (HDC) cycles. Our results indicate that immunomagnetic selection of CD34+ cells yields highly purified autografts devoid of tumour cells whereas the engraftment ability of the progenitor and stem cells is fully retained.

We evaluated the therapeutic efficacy and toxicity of a tandem high-dose therapy with peripheral blood stem cell (PBSC) support in 40 patients with high-risk, primary breast cancer (stage II–III) and involvement of ten or more positive axillary lymph nodes. Their median age was 44 years (range 23–56). Two cycles of cytotoxic chemotherapy with ifosfamide (10000 mg/m2) and epirubicin (100 mg/m2) were administered. Granulocyte colony-stimulating factor (G-CSF) was given to hasten neutrophil reconstitution and to mobilise PBSC during marrow recovery. Leukaphereses were performed following the first and/or second cycle. Tandem high-dose therapy consisted of two cycles with ifosfamide (15 or 12 g/m2) and epirubicin (150 mg/m2), while carboplatin (900 mg/m2) was added for the last 24 patients included. Using an immunocytochemical method, two of 11 patients had cytokeratin-positive tumour cells in three leukapheresis products that were collected following the first G-CSF-supported cycle with ifosfamide and epirubicin, whereas only two harvests obtained following the second cycle in 26 patients contained cytokeratin-positive tumour cells. The number of CD34+ cells/kg re-infused following both high-dose cycles was similar (4.20 ± 0.29 × 106, first cycle and 5.25 ± 0.63 × 106, second cycle), and no notable difference was noted in the speed of haematological reconstitution. An absolute neutrophil count (ANC) of 0.5 × 109/l was reached after a median time of 13 days, while an unsupported platelet count of 20.0 × 109/1 was achieved after a median time of 8 (first cycle) and 9 (second cycle) days post-transplantation. Patients autografted with more than 7.5 × 106 CD34+ cells/kg had platelet counts above 20 × 109/1 within less than 10 days. 6 patients relapsed between 7 and 11 months (median 8 months) post-transplantation. 37 patients are alive and in remission with a median follow-up time of 11 months (range 1–38). This translates into a probability of disease-free survival (DFS) of 77% (95% CI 32–95%) at 38 months. The probability of overall survival is 85%, since 3 patients with local relapse achieved a second complete remission following surgery and involved-field radiotherapy. In conclusion, a sequential high-dose therapy including ifosfamide, epirubicin, carboplatin and PBSC support is well tolerated and effective in patients with high-risk primary breast cancer. Involved-field irradiation should be performed post-transplantation to reduce the risk of local relapse.

Patienten mit indolenten Non-Hodgkin-Lymphomen (NHL) in fortgeschrittenen Stadien sind mit konventioneller Therapie nicht heilbar. Obwohl eine komplette Remission mit Standardbehandlung häufig erreicht werden kann, ist deren Dauer mit nur 1-3 Jahren gering. Die hohe Rezidivrate beschränkt die krankheitsfreie Überlebenswahrscheinlichkeit auf 30% nach 5 Jahren. Die Intensivierung einer zytotoxischen Behandlung ist durch Transplantation blutbildender Vorläuferzellen nach hoch dosierter Chemotherapie mit oder ohne Ganzkörperbestrahlung möglich. Durch Transplantation von Knochenmark oder peripherer Blutstammzellen von dem Patienten selbst oder einem gesunden Spender ist die Myelotoxizität nicht mehr dosislimitierend. Darüber hinaus hat der Einsatz rekombinanter hämatopoetischer Wachstumsfaktoren die Entwicklung der Stammzelltransplantation gefördert und die behandlungsassoziierte Mortalität verringert. Im Vergleich zur Gesamtzahl der transplantierten Patienten bilden Patienten mit indolenten Lymphomen noch eine kleine Gruppe. Für diese Patientengruppe werden wir versuchen, die Frage nach dem therapeutischen Nutzen der Hochdosistherapie mit Stammzelltransplantation zu beantworten, insbesondere unter Berücksichtigung des Zeitpunkts der Transplantation innerhalb des Krankheitsverlaufs, der Art der Transplantation (autolog oder allogen) und der Quelle der Stammzellen (Knochenmark oder peripheres Blut). Für die Gruppe der Patienten mit follikulären Lymphomen werden wir auf die Bedeutung des Nachweises von residualen Tumorzellen im Autotransplantat sowie im Knochenmark nach der Transplantation eingehen.

Zentroblastisch-zentrozytische (CB-CC-)Non-Hodgkin-Lymphome (NHL) werden nach der Kiel-Klassifikation [9] aufgrund pathomorphologischer Kriterien und ihres klinischen Verlaufes als niedrigmaligne eingestuft [1,11]. Die mittlere Überlebensdauer von Patienten im Stadium IV liegt zwischen 4 und 6 Jahren. Eine Heilung der Patienten durch konventionelle Therapie wird nur selten erreicht, da selbst Patienten, die nach Polychemotherapie mit Cyclophosphamid, Vincristin, Adriamycin und Prednison (CHOP) in kompletter Remission sind, fast immer rezidivieren [2]. Angesichts dieser schlechten Langzeitergebnisse behandelten wir Patienten mit einem CB-CC-NHL ungeachtet der indolenten Biologie dieses histologischen Subtyps mit einer Hochdosistherapie und Blutstammzelltransplantation [3,5, 6, 8,10]. Bei diesem Therapieverfahren wird die dosislimitierende Nebenwirkung der Myelotoxizität durch Transplantation hämatopoetischer Stammzellen umgangen. Noch bis Ende der 8oer Jahre diente v.a. autologes Knochenmark als Quelle hämatopoetischer Stammzellen, bis es durch die Verfügbarkeit hämatopoetischer Wachstumsfaktoren möglich wurde, ausreichende Mengen blutbildender Stammzellen ins periphere Blut zu mobilisieren. In Heidelberg wurde 1991 eine Studie zur sequentiellen Hochdosistherapie mit Blutstammzelltransplantation für Patienten mit CB-CC-NHL begonnen, deren Ergebnisse in dieser Arbeit vorgestellt werden.

It was the objective of our study to evaluate the efficacy of a sequential high-dose therapy with peripheral blood progenitor cell (PBPC) support in patients with low-grade non-Hodgkin's lymphoma (NHL). Since July 1991, 48 patients (23 male/25 female) with a median age of 43 years (range 26-55) were included in the study. At the time of entry, 28 patients were in first and seven in second or higher remission. Twelve patients had relapse of disease and one patient had tumor progression. PBPC were collected during granulocyte colony-stimulating factor (G-CSF)-enhanced leukocyte recovery following treatment with high-dose cytarabine and mitoxantrone (HAM). A median of two leukaphereses (range 2-7) resulted in 6.9 x 10(6) CD34+ cells/kg (median, range 2.1 x 10(6)-38.8 x 10(6)). A comparison was made between the harvests obtained from patients in first remission and those from patients in second remission, in relapse or progressive disease. Patients mobilized in first remission tended to have a greater collection efficiency for CD34+ cells comprising a significantly greater proportion of more primitive CD34+/Thy-1+ progenitor cells. Conversely, leukapheresis (LP) products collected during first remission contained a significantly smaller proportion of CD34+/CD45RA+ cells and CD34+/c-kit+ cells, subsets which reflect a more differentiated progenitor cell stage. Following high-dose therapy and PBPC autografting, the median time to reach platelets > or = 20 x 10(9)/l and neutrophils > or = 0.5 x 10(9)/l and 12 and 13 days, respectively. Two patients died of treatment-related toxic organ failure. Thirty-nine patients are alive in remission after a median follow-up time of 15 months (range 1-31), while seven patients relapsed between 5 and 29 months post-transplantation. Except for one patient autografted in first remission, the patients with relapse had a history of previous relapse or progressive disease. Since the probability of disease-free survival appears to be related to the disease status at the time of autografting, PBPC-supported high-dose therapy including total body irradiation should be investigated further for patients with low-grade NHL while they are in first remission.

In order to study the effect of high-dose therapy with peripheral blood stem cell transplantation (PBSCT) on the haemopoietic reserve in man, the number and composition of bone marrow (BM) and peripheral blood (PB)-derived progenitor cells were examined in 137 cancer patients. In 45 patients, paired samples from BM and PB were obtained before PBSC mobilization and 6-27 months after transplantation. Following PBSCT. the proportion of CD34+ cells was significantly smaller than before mobilization (BM 1.99 +/- 0.24 versus 0.8 +/- 0.09, P < 0.001), and no change was observed at several follow-up visits thereafter. The reduction was most pronounced for the primitive BM progenitor subsets such as the CD34+/DR- and CD34+/ Thy-1+ cells. The impairment of hematopoiesis was also reflected by a significant reduction in the plating efficiency of BM and PB samples. No relationship was found between the decrease in the proportion of CD34+ cells and any particular patient characteristics, kind of high-dose therapy or the CD34+ cell content in the autograft. In conclusion, high-dose therapy with PBSC transplantation is associated with a long-term impairment of the haemopoietic system. The reduction in the number of haemopoietic progenitor cells is not associated with a functional deficit, as peripheral blood counts post-transplantation were normal in the majority of patients.

It was the aim of our study to determine the collection efficiency and yield of CD34+ cells in 88 cancer patients (pts, 44 males/44 females) who underwent 154 large-volume leukaphereses (LV-LPs). The diagnoses were as follows: 18 patients had Non-Hodgkin's lymphoma, 9 Hodgkin's disease, 24 multiple myeloma, 6 acute leukemia, 27 breast cancer, and 4 patients had solid tumors of different types. During the course of LV-LPs, 20 liters (l) of blood were processed at a median flow-rate of 85 ml/min (CS 3000 Baxter) and 130 ml/min (COBE Spectra), respectively. Peripheral blood stem cells (PBSC) were collected following granulocyte colony-stimulating factor (G-CSF)-supported cytotoxic chemotherapy. A 31% and 21% mean decrease in the platelet and white blood count was noted at the end of the LV-LPs when compared with the pre-leukapheresis values. The aphereses were well tolerated without adverse effects. The level of circulating CD34+ cells was closely related to the number of CD34+ cells contained in the respective leukapheresis product (R = 0.89, P < 0.001). Compared with 270 patients who underwent 838 regular 101 LPs, the yield of CD34+ cells/kg was almost two-fold greater (4.84 ± 0.63 × 106 [Mean ± SEM] vs 2.60 ± 0.16 × 106, P < 0.001). The antigenic profile of CD34+ cells was assessed in 54 separate products collected on the occasion of 27 LV-LPs following the processing of 101 and 201, respectively. The intra-individual comparison included differentiation as well as lineage-associated markers (CD38, Thy-1, c-kit, CD33, CD45RA). No difference in the subset composition was observed between the first and second product, arguing against a preferential release of particular CD34+ cell subsets during the procedure. As shown by molecular biological or immunocytochemical examination, the likelihood of harvesting malignant cells using large-volume aphereses was not increased in comparison with regular leukaphereses. Single harvests of ≥2.5 × 106 CD34+ cells/kg could be obtained in 74% of the patients, compared with 52% in case of regular LPs. As the majority of patients were autografted with more than 2.5 × 106 CD34+ cells/kg following high-dose therapy, hematological recovery in general was rapid and not related to the type of apheresis product used. Considering patient comfort and savings in resource utilization, large-volume leukaphereses have become the standard procedure for PBSC collection in our center. © 1996 Wiley-Liss, Inc.

Long-term disease-free survival following conventional cytotoxic therapy is extremely rare in patients with advanced-stage mantle cell lymphoma (MCL). High-dose conditioning therapy consisting of hyperfractionated total body irradiation (TBI, 14.4 Gy) and cyclophosphamide (200 mg/kg) was therefore offered to 13 patients (four females/nine males) with advanced-stage MCL. The patients were relatively young with a median age of 49 years (range 30-60). High-dose cytarabine and mitoxantrone with granulocyte colony-stimulating factor (G-CSF) support were given for second-line therapy and mobilization of peripheral blood stem cells (PBSC). During cytokine-stimulated marrow recovery, a median of two leukaphereses (range 1-4) were performed. Using direct immunofluorescence analysis including two-color staining, the proportion of CD19+ B cells and CD34+/CD19+ B lymphoid progenitor cells was found to be extremely low with quantities below detection limit in approximately 50% of the autografts. At the time of autografting, nine patients (pts) were in first partial (five pts) or complete (four pts) remission, while four patients had achieved a second complete remission. Following myeloablative therapy a median number of 7.5 x 10(6) CD34+ cells/kg were autografted. The median time for neutrophil (> or = 0.5 x 10(9)/l) and platelet recovery (> or = 20 x 10(9)/l) was 13 and 10 days, respectively. Hematological recovery was delayed in a patient who received 5.8 x 10(6) positively selected CD34+ cells/kg. There was one toxic death 17 days post-transplantation because of overwhelming interstitial pneumonia. Two patients with a history of previous treatment failure relapsed 10 and 11 months post-transplantation, respectively, at sites of previous disease. Ten patients are disease-free with a median follow-up time of 18 months (range 10-47). The results presented here suggest that PBSC-supported high-dose therapy including TBI may provide long-term disease-free survival for patients with advanced-stage mantle cell lymphoma.

To assess whether measurement of CD34+ cells in the peripheral blood allows one to estimate the progenitor cell yields of subsequent leukapheresis procedures, 733 corresponding blood and leukapheresis samples were analyzed. Peripheral blood progenitor cells of cancer patients were mobilized with hematopoietic growth factors alone or postchemotherapy, and harvested processing 10 liters of blood for each leukapheresis product. The CD34+ cell count (CD34+ cells/μl blood) correlated most closely with the progenitor cell yield in the corresponding leukapheresis product (CD34+ cells/kg bodyweight, r = 0.80), while the proportion of circulating CD34+ cells to the white blood and mononuclear cells predicted the yield less reliably (r = 0.74 and r = 0.60). The CD34+ cell yield was independent of the white blood count (r = 0.04), whereas a weak correlation was found between the mononuclear cell count and the number of CD34+ cells/kg collected (r = 0.42). It was unlikely to obtain the threshold quantity of 2.5 × 10 6 CD34+ cells/kg required for rapid engraftment when counts below 10 CD34+ cells/μl blood were detected. At levels between 10 and 30 CD34+ cells/μl sufficient autografts could be harvested, whereas 30–100 CD34+ cells/μl were required to achieve this by a single leukapheresis. A surplus of CD34+ cells was likely above 100 CD34+ cells/μl which could be useful for progenitor cell enrichment techniques. The correlation between the CD34+ cell count and progenitor cell yield was independent of the mobilizing regimen and whether leukaphereses had been performed previously. In conclusion, the number of CD34+ cells/μl blood allows a reliable prediction of the CD34+ progenitor cell yield in subsequent leukapheresis procedures. However, rare cases of unexpectedly sufficient progenitor cell yields may be observed even at CD34+ cell levels below detection limit.

The purpose of this study was to compare two different in vitro culture conditions for the preservation of human granulocytes. These cells could be used in patients with severe neutropenia following cytotoxic chemotherapy if the functional capacity was retained, and autologous transfusions of granulocytes would circumvent the risk of alloimmunization. Granulocytes were obtained from the peripheral blood of healthy donors and patients with hematologic malignancies who received cytotoxic chemotherapy supported by recombinant human granulocyte colony-stimulating factor (R-metHuG-CSF, 300 micrograms/day, s.c.). Granulocytes were either cultured for 72 h at 4 degrees C in the presence of 100 ng/ml G-CSF or cryopreserved at -196 degrees C. The viability, surface antigen expression, and function of the granulocytes were assessed. Since effective microbial killing involves the attachment of granulocytes to blood vessel walls, transmigration into tissues, chemotaxis, and phagocytosis, the surface expression of the adhesion molecules LFA-1 (CD11a/CD18) and gp 150,95 (CD11c/CD18) was measured. In addition, the IgG receptors Fc gamma RI (CD64), Fc gamma RII (CD32), and Fc gamma RIII (CD16), as well as the complement receptor CR3 (CD11b/CD18), were assessed. Dynamic superoxide anion release served as a measure of the metabolic pathway of the oxidative burst after f-Met-Leu-Phe (fMLP) and phorbol-12-myristate-13-acetate (PMA) stimulation. Substantial differences in the preservation of granulocyte integrity and function were observed between the two storage conditions. Cryopreservation abolished reactivity to extracellular stimuli and severely affected the cell phenotype. On the other hand, functional activity could be maintained for up to 72 h when in vivo primed granulocytes of patients were incubated at 4 degrees C in the presence of G-CSF. This storage modality may permit the use of granulocyte autotransfusion to reduce the risk of neutropenic fever.

Peripheral blood progenitor cells (PBPC) can be mobilized using cytotoxic chemotherapy and cytokines. There is a substantial variability in the yield of hematopoietic progenitor cells between patients. We were looking for predictive parameters indicating a patient's response to a given mobilization regimen. Multiparameter flow-cytometry analysis and clonogenic assays were used to examine the hematopoietic progenitor cells in bone marrow (BM) and peripheral blood (PB) before filgrastim (R-metHuG-CSF; Amgen, Thousand Oaks, CA)-supported chemotherapy and in PB and leukapheresis products (LPs) in the recovery phase. Fifteen patients (four with high-grade non-Hodgkin's lymphoma [NHL], two with low-grade NHL, two with Hodgkin's disease, two with multiple myeloma, three with breast cancer, one with ovarian cancer, and one with germ cell tumor) were included in this study. The comparison of immunofluorescence plots showed a homogenous population of strongly CD34+ cells in steady-state and mobilized PB whereas in steady-state BM, the CD34+ cells ranged from strongly positive with continuous transition to the CD34- population. Consistent with the similarity in CD34 antigen expression, a correlation analysis showed steady-state PB CD34+ cells (r = .81, P < .001) and colony-forming cells (CFCs; r = .69, P < .01) to be a measure of a patient's mobilizable CD34+ cell pool. Individual estimates of progenitor cell yields could be calculated. With a probability of 95%, eg, 0.4 steady-state PB CD34+ cells x 10(6)/L allowed to collect in six LPs 2.5 x 10(6) CD34+ cells/kg, the reported threshold-dose of progenitor cells required for rapid and sustained engraftment after high-dose therapy. For the total steady-state BM CD34+ cell population, a weak correlation (r = .57, P < .05) with the mobilized CD34+ cells only became apparent when an outlier was removed from the analysis. Neither the CD34+ immunologic subgroups defined by the coexpression of the myeloid lineage-associated antigens CD33 or CD45-RA or the phenotypically primitive CD34+/HLA-DR-subset nor the BM CFC count had a predictive value for the mobilization outcome. This may be caused by the additional presence of maturing progenitor cells in BM, which express lower levels of the CD34 antigen and do not circulate. Our results permit us to recognize patients who are at risk to collect low numbers of progenitor cells and those who are likely to achieve sufficient or high progenitor cell yields even before mobilization chemotherapy is administered.

Between September 1991 and April 1995, high-dose therapy with peripheral blood progenitor cell (PBPC) support was administered to 105 patients with non-Hodgkin's lymphoma (NHL). Thirty-three patients had high-grade NHL, while 72 patients had different forms of low- or intermediate-grade NHL. Except for three patients who received G-CSF during steady-state hematopoiesis, PBPCs were collected following cytokine-supported cytotoxic chemotherapy. This included G-CSF or the sequential administration of interleukin 3 (IL-3) and GM-CSF. Assessing bone marrow (BM) samples before the start of chemotherapy and leukapheresis (LP) products collected during cytokine-enhanced marrow recovery, a 2.3-fold greater mean concentration of CD34- cells was found in peripheral blood (p < 0.005). The blood-derived progenitor cells were enriched with a particular subset of more primitive progenitors, as the mean proportion of CD34+/Thy-1+ cells in LP products was three-fold greater in comparison to premobilization BM samples, respectively (p < 0.001). In contrast, the mean proportion of CD34+/CD19+ and CD19+ cells in LP products was 8.8- and 80-fold smaller compared to BM samples, respectively (p < 0.001). Following high-dose conditioning therapy including TBI in 74 patients, reinfusion of PBPC resulted in rapid and sustained engraftment in the majority of patients, while in seven patients an unsubstituted platelet count of greater than 20 x 10(9)/l was reached between 31 and 51 days. Five patients died of treatment-related complications between 13 and 188 days following transplantation. The probability of long-term disease-free survival at 30 months in patients autografted while they were in first remission was 70% in high-grade and 83% in low-grade NHL, respectively. The data may provide the rationale for the use of PBPC-supported high-dose regimens as first-line treatment for patients at high risk of treatment failure.

The article provides a review of the role of granulocyte colony-stimulating factor (G-CSF) for mobilization and transplantation of peripheral blood progenitor and stem cells. Recombinant gene technology has permitted the production of highly purified material for therapeutic use in humans. Progenitor cells can be assessed using semisolid and liquid culture assays or direct immunofluorescence analysis of cells expressing CD34. This antigen is found on lineage-determined hematopoietic progenitor cells as well as on more primitive stem cells with extensive self-renewal capacity. Administration of G-CSF during steady-state hematopoiesis or following cytotoxic chemotherapy leads to an increase of hematopoietic progenitor cells in the peripheral blood. The level of circulating CD34+ cells post-chemotherapy is greater compared with G-CSF administration during steady state. On the other hand, CD34+ cells harvested post-chemotherapy contain a smaller proportion of more primitive progenitor cells (CD34+/HLA-DR- or CD34+/CD38-) compared with G-CSF treatment alone. Independent of the mobilization modality, the amount of previous cytotoxic chemo- and radiotherapy adversely affects the yield of hematopoietic progenitor cells. While continuous subcutaneous administration of G-CSF between 5 and 16 micrograms/kg bodyweight is preferred, additional dose-finding studies may be helpful to optimize current dose schedules. Adhesion molecules like L-selectin, VLA (very late antigen)-4 and LFA (leukocyte function antigen)-1 are likely to play a role in mobilization, since these antigens are expressed on CD34+ cells from bone marrow in different densities compared with blood-derived CD34+ cells collected following G-CSF-supported cytotoxic chemotherapy. It is also relevant for transplantation that during G-CSF-enhanced recovery post-chemotherapy, peripheral blood is enriched with a greater proportion of CD34+ cells expressing Thy-1 in comparison with CD34+ cells from bone marrow samples obtained on the same day or before the mobilization therapy was started. The early nature of the CD34+/Thy-1+ cells is very likely since this phenotype has been found on stem cells from human fetal liver and bone marrow and on cord blood cells. As a result, G-CSF-mobilized blood stem cells provide rapid and sustained engraftment following high-dose therapy, including myeloablative regimens. Positive selection of CD34+ cells as well as ex vivo expansion using different cytokines are currently being investigated for purging and improvement of short-term recovery post-transplantation. Future developments include the use of blood-derived hematopoietic stem cells for somatic gene therapy. The availability of growth factors has been an important prerequisite for the development of these new avenues for cell therapy.