Diagnosis and management of acquired aplastic anemia in childhood. Guidelines from the Marrow Failure Study Group of the Pediatric Haemato-Oncology Italian Association (AIEOP).

Acquired aplastic anemia (AA) is a rare heterogeneous disorder characterized by pancytopenia and hypoplastic bone marrow. The incidence is 2-3 per million population per year in the Western world, but 3 times higher in East Asia. Survival in severe aplastic anemia (SAA) has improved significantly due to advances in hematopoietic stem cell transplantation (HSCT), immunosuppressive therapy, biologic agents, and supportive care. In SAA, HSCT from a matched sibling donor (MSD) is the first-line treatment. If a MSD is not available, options include immunosuppressive therapy (IST), matched unrelated donor, or haploidentical HSCT. The purpose of this guideline is to provide health care professionals with clear guidance on the diagnosis and management of pediatric patients with AA. A preliminary evidence-based document prepared by a group of pediatric hematologists of the Bone Marrow Failure Study Group of the Italian Association of Pediatric Hemato-Oncology (AIEOP) was discussed, modified and approved during a series of consensus conferences that started online during COVID 19 and continued in the following years, according to procedures previously validated by the AIEOP Board of Directors.

Late-onset and long-lasting neutropenias in the young: A new entity anticipating immune-dysregulation disorders.

This study identifies a new chronic form of immune neutropenia in the young with or without detectable indirect anti-neutrophil antibodies, characterized by mild/moderate neutropenia low risk of severe infection (14%), tendency to develop autoimmune phenomena over the course of the disease (cumulative incidence of 58.6% after 20 years of disease duration), leukopenia, progressive reduction of absolute lymphocyte count and a T- and B-cell profile similar to autoimmune disorders like Sjogren syndrome, rheumatoid arthritis, and systemic lupus erythematosus (increased HLADR+ and CD3 + TCRγδ cells, reduced T regulatory cells, increased double-negative B and a tendency to reduced B memory cells). In a minority of patients, P/LP variants related to primary immuno-regulatory disorders were found. This new form may fit the group of "Likely acquired neutropenia," a provisional category included in the recent International Guidelines on Diagnosis and Management of Neutropenia of EHA and EUNET INNOCHRON ACTION 18233. The early recognition of this form of neutropenia would help clinicians to delineate better specific monitoring plans, genetic counseling, and potentially targeted therapies.

Establishment of a reliable experimental procedure for bone marrow purging with mafosfamide (ASTA Z 7557).

Bone marrow purging with cyclophosphamide derivatives (Mafosfamide) requires the establishment of a defined experimental procedure for reliable leukemic cell destruction while sparing normal hematopoietic stem cells to ensure engraftment. We previously defined the granulocyte-macrophage colony-forming unit (CFU-GM) LD95 as being the maximum tolerable dose of drug to use. We now report, in 20 patients with acute non-lymphoblastic leukemia (n = 5), acute lymphoblastic leukemia (n = 5), chronic myelogenous leukemia (n = 5), and non-Hodgkin's lymphoma (n = 5), that the nature of the cells treated (i.e., buffy coat cells or mononuclear cells) significantly influences the accuracy of the LD95 determination, whereas other parameters such as hematocrit or nucleated cell concentration do not. We subsequently define the most reliable experimental procedure for in vitro purging with Mafosfamide: incubation of 2 x 10(7) buffy coat cells/ml with a hematocrit of 5%. We show that the wide individual susceptibility to the drug is not related to any incubation procedure. In a series of 163 patients with hematological malignancies, we confirm the large variation of sensitivity to the drug according to patient susceptibility and diagnosis. These data favor the adjustment of the dose of Mafosfamide on an individual basis, prior to bone marrow purging for autologous bone marrow transplantation.

Presence of primitive lymphoid progenitors with NK or B potential in ex vivo expanded bone marrow cell cultures.

OBJECTIVE: In previous work, we showed that CD34+ bone marrow cells can be successfully expanded along the myeloid pathway in stroma- and serum-free conditions in the presence of SCF+IL-3+IL-6+Flt3-l+G-CSF+MGDF. Due to the lack of phenotypically detectable lymphoid cells, it was necessary to address the question of the lymphoid potential of the expanded populations under these conditions. MATERIALS AND METHODS: The present report describes a long-term culture system that supports human B- and NK-cell differentiation from the day 14 fraction without further selection of the more primitive cells. In NK proliferation assays, the cells were maintained over stroma cells in the presence of IL-2 for 4-5 weeks. NK initiating cells (NK-IC) were determined by a limiting dilution assay. In B-cell cultures, the expanded cells were maintained over MS5 in the presence of Flt3-l for 4-8 weeks. RESULTS: NK cells rose from 0.2%+/-0.04% at culture initiation to 71%+/-6% at week 5. These cells displayed cytolytic activity. NK-IC evaluation showed a mean 18-fold expansion in the day 14 expanded fraction as compared to the initial day 0 fraction. Similarly, CD19+ cells rose from 0.1% at culture initiation to 30%+/-1% at week 6. Cells produced under these B-LTC conditions were CD34-CD19+CD10+. We also demonstrated that the CD34+/Lin- sorted cells from the day 14 fraction gave rise to NK and B cells. CONCLUSION: This culture system permits the revelation of a population that, although poorly represented in terms of phenotypically detectable cells, nevertheless retains high levels of lymphoid NK and B potential after 14 days expansion. Such data suggest the persistence, or expansion, of lymphoid progenitors and, hence, the multipotentiality of the expanded progenitor/stem cells.

Recovery of CFU-GM from cryopreserved marrow and in vivo evaluation after autologous bone marrow transplantation are predictive of engraftment.

In 18 patients with non-Hodgkin's lymphomas or solid tumors treated with intensive chemotherapy and/or total-body irradiation followed by autologous bone marrow transplantation (ABMT), we assessed the value of granulocyte-macrophage progenitor cells (CFU-GM) monitoring to predict engraftment. We studied CFU-GM in cryopreserved marrow and attempted to settle whether detection of CFU-GM in vivo after ABMT has a predictive value on engraftment. Our data showed: The absence of linear correlation linking recovery of hematopoiesis to the dose of CFU-GM/kg infused. The existence of a CFU-GM threshold in respect to engraftment. Patients receiving doses of CFU-GM greater than 10(3)/kg had significantly faster recovery kinetics for hematopoiesis than did patients receiving doses below this threshold, with median recoveries to 0.5 and 1.0 X 10(9) neutrophils/liter, respectively, on days 14 and 15 versus days 29 and 31.5 (p less than 0.05 and p less than 0.02) and median recoveries to 1.0 and 2.5 X 10(9) leukocytes/liter respectively, on days 12.5 and 16 versus days 28 and 30.5 (p less than 0.05 and p less than 0.02). Considering the entire course of events during the first four weeks, we were able to show that white blood cell recovery was significantly faster in the group of patients receiving doses of CFU-GM greater than 10(3)/kg (p less than 0.001). Sequential studies of the reappearance of CFU-GM in marrow and peripheral blood indicated that the kinetics of CFU-GM recovery in vivo after ABMT predict engraftment. By day 7 after the graft, CFU-GM were already detectable in the marrow at a level of 10% of the dose infused for patients with optimal engraftment--median time to recovery to 1.0 and 2.5 X 10(9) leukocytes/liter and 1.0 X 10(9) neutrophils/liter on days 11, 15, and 14.5 versus days 18, 23, and 23 (p less than 0.02, less than 0.05, and less than 0.05), respectively after. On day 10 after ABMT, a 15% CFU-GM level in bone marrow confirmed engraftment, with a significant correlation of all parameters studied--1.0 and 2.5 X 10(9) leukocytes/liter (p less than 0.02 and less than 0.01), 0.5 and 1.0 X 10(9) neutrophils/liter (p less than 0.05), 50.0 and 100.0 X 10(9) platelets/liter (p less than 0.05). On day 14, a 50% CFU-GM level was reached in all patients with optimal engraftment; p less than 0.01 on 1.0, and 2.5 X 10(9) leukocytes on 0.5 and 1.0 X 10(9) neutrophils/liter. The detection of circulating CFU-GM in the blood by day 10 or 14 indicated engraftment.(ABSTRACT TRUNCATED AT 400 WORDS)

Amifostine (WR-2721) protects normal haematopoietic stem cells against cyclophosphamide derivatives' toxicity without compromising their antileukaemic effects.

We compared the effects of amifostine (WR-2721) on the cytotoxicity of mafosfamide or 4-hydroperoxycyclophosphamide (4-HC) in normal marrow progenitor cells (CFU-GM) and leukaemic progenitor cells (CFU-L) during ex vivo purging for autologous bone marrow transplantation (ABMT). Mononuclear cells (MNC) were incubated with amifostine 3 mg/ml for 15 min, washed, and subsequently tested for their sensitivity to mafosfamide or 4-HC (20-200 micrograms/ 10(7) MNC/ml). The LD95 was significantly higher among amifostine-treated cells for PCM-CFU-GM in 6 of 13 patients and for 5R-CFU-GM in 4 of 10 patients (P < 0.05). In contrast, amifostine exhibited no protective effects upon CFU-L. The results of this study will show that amifostine protects normal late and early progenitor cells for the toxic effects of cyclophosphamide derivatives while preserving their antileukaemic effects. These results suggest that amifostine has therapeutic value as a protective agent for normal marrow progenitor cells during ex-vivo purging of bone marrow for ABMT.

Comparative effects of amifostine (Ethyol) on normal hematopoietic stem cells versus human leukemic cells during ex vivo purging in autologous bone marrow transplants.

The protective action of amifostine (Ethyol, US Bioscience, Inc. West Conshohocken, PA) against the toxic effects of cyclophosphamide derivatives on the normal progenitor/stem cell pool was investigated. Early and late normal progenitor/stem cells were studied in the presence of placenta-conditioned medium (placenta-conditioned medium-granulocyte-macrophage colony-forming units); in the presence of granulocyte colony-stimulating factor, granulocyte-macrophage colony-stimulating factor, interleukin-3, erythropoietin, stem cell factor (5R-CFU-GM); stimulated granulocyte-macrophage colony-forming and in a long-term culture-initiating cell assay. The preservation of an antileukemic effect was investigated by growing leukemic progenitor cells in the presence of phytohemagglutinin and leukocyte feeder layer with or without 25u interleukin-2. In 10 of 13 cases, a statistically significant (P < .05) protective effect was found on PCM-granulocyte-macrophage colony-forming units, in four of 10 cases on factor-stimulated granulocyte-macrophage colony-forming units, and in two of six cases on long-term culture-initiating cell. In contrast, amifostine exhibited no protective effect (none of nine cases) on leukemic progenitor cells. From the experimental data, it seems that amifostine is able to protect human normal progenitor/stem cells from cyclophosphamide derivative toxicity, while preserving their antileukemic effects. The therapeutic usefulness of such protection in autologous bone marrow transplantation with purged marrow is obvious.

Blood and spleen haematopoiesis in patients with myelofibrosis.

Blood nucleated cells collected by leukapheresis and spleen cell suspension from patients with myelofibrosis with myeloid metaplasia (MMM) were studied for their haematopoietic capacity. Using committed progenitor cell assays (CFU-GM, BFU-e) and a one-stage long-term liquid stem cell system, we have shown: (1) a preferential expansion of the circulating committed progenitor cell pool above the more primitive stem cell compartment; (2) the absence of any development of a stromal adherent layer in long-term cultures of peripheral blood nucleated cells suggesting the self-sustaining capacity of the circulating primitive stem cells; (3) that the spleen is only a production site of committed progenitor cells but does not generate primitive stem cells; (4) the presence, in the spleen, of stromal progenitor cells. We conclude that the peripheral blood primitive stem cells in patients with MMM are not of splenic origin.

Characterization of late and early hematopoietic progenitor/stem cell sensitivity to mafosfamide.

The preservation of the hematopoietic value of the graft is essential in the context of bone marrow purging for autologous transplantation. In the present study we have investigated mafosfamide toxicity to transplantable hematopoietic stem cells using combinations of recombinant growth factors (GF) in semi-solid assays and by measurement of Long-Term Culture Initiating Cells (LTC-IC). Our data show that various subtypes of progenitor/stem cells were inhibited in a dose-dependent manner by mafosfamide. A hierarchy appeared clearly regarding sensitivity to the drug in the following order of increasing resistance: PCM CFU-GM (grown in presence of placenta-conditioned medium), 4R CFU-GM (grown in presence of GM-CSF + IL-3 + G-CSF + EPO), 5R CFU-GM (grown in presence GM-CSF + IL-3 + G-CSF + EPO + SCF) and LTC-IC with respective lethal dose 95 (LD95) levels of 40 micrograms, 55 micrograms, 90 micrograms and 140 micrograms/10(7) MNC (P = 0.05 to P = 0.018). Even the primitive stem cells treated with very high doses of mafosfamide (2 to 4 times the usually recommended dose) responded to a combination of SCF + GM-CSF + G-CSF + IL-3 in liquid marrow culture, suggesting that they were functionally spared by the treatment. We also observed a higher sensitivity of 5R CFU-GM and LTC-IC from patients with hematological malignancies, compared with normal donors, when marrow was treated with the dose chosen for clinical purging.(ABSTRACT TRUNCATED AT 250 WORDS)

Flt 3 ligand, MGDF, Epo and G-CSF enhance ex vivo expansion of hematopoietic cell compartments in the presence of SCF, IL-3 and IL-6.

The aim of the study is to define the ability of Flt3 ligand, MGDF, Epo and G-CSF to modulate the expansion of different hematopoietic compartments in association with a basic cocktail of SCF + IL-3 + IL-6 (S36). CD34+ cells from normal bone marrow were cultured in stroma-free, serum-free medium for 10 days. Using various concentrations of cytokines, total cells could be expanded up to 5200-fold, CD34+ cells up to 78-fold, CFU-GM up to 143-fold, BFU-E up to 46-fold, CFU-MK up to six-fold and LTC-IC up to four-fold. The results were assessed by multiparametric analysis of variance. Three factors had a significant stimulatory effect on the late precursor compartment: Epo (P < 10(-5)), G-CSF (P=5 x 10(-3)) and FL (P=10(-5)). Two were critical for CD34+ cell expansion: FL (P=4 x 10(-5)) and Epo (P=6 x 10(-5)), while two were critical for BFU-E expansion: MGDF (P=8 x 10(-4)) and FL (P=0.017). FL strongly stimulated CFU-GM expansion (P < 10(-5)), whereas none of the growth factors studied had any effect on CFU-MK. FL (P=10(-4)) and MGDF (P=0.002) were essential to obtain high levels of expansion of LTC-IC as determined in limiting dilution assays. In the light of the above results showing a preferential effect on the expansion of precursor cells (3080-fold), CD34+ cells (53-fold), CFU-GM (134-fold), BFU-E (46-fold) and LTC-IC (five-fold), the combination SCF, IL-3, IL-6, FL, MGDF, Epo and G-CSF was chosen as a putative cytokine cocktail for further studies on long-term culture. Sustained production of precursor cells, progenitor cells, LTC-IC and E-LTC-IC for up to 100 days reflects the persistence of very primitive stem cells. This suggests that these populations are probably able to undergo self-renewal divisions. The above combination of cytokines meets the required criterion for potential clinical application, which may be defined as an effective capacity to expand all cell compartments, using as the starting material high concentrations of low purity CD34+ cells.

Comparison of CD34+ bone marrow cells purified by immunomagnetic and immunoadsorption cell separation techniques.

We tested two positive selection techniques for separation of CD34+ cells from bone marrow and analyzed the yields of CD34+ cells, BFU-E, CFU-GM, CFU-MK and LTC-IC after selection and expansion. An immunoadsorption procedure (CellPro) and an immunomagnetic (Baxter) CD34+ cell separation method were employed to purify the same bone marrow samples from seven normal subjects. Mean yields of CFU-GM and CFU-MK and absolute numbers of LTC-ICs were not different in the two purified cell populations. In contrast, the mean recovery of BFU-E was significantly lower for the immunoadsorption (21 +/- 14%) than for the immunomagnetic technique (44 +/- 27%). After separation, CD34+ cells were evaluated in 10-day liquid cultures for their expansion capacity in terms of total cells and progenitors. The expansion capacity of progenitors such as CFU-GM, CFU-MK and especially BFU-E selected by immunoadsorption was higher than the capacity of progenitors obtained by immunomagnetism, although final total and progenitor cell numbers are similar. Our results suggest that the populations separated by the two techniques differ mainly in the expansion capacity of progenitors and in the recovery of BFU-E after the selection procedure. These differences between two methods, which already are widely employed in research and in clinical transplantation, should be taken into account when considering the aims of the experiments.

Cell culture bags allow a large extent of ex vivo expansion of LTC-IC and functional mature cells which can subsequently be frozen: interest for a large-scale clinical applications.

The aim of this study was to evaluate the ex vivo expansion of normal CD34+ cells in gas-permeable polypropylene bags suitable for clinical use. Cells were cultured for 14 days in serum-free medium supplemented with SCF, IL3, IL6, FLT3-1, G-CSF + MGDF or Epo. The bags supported the expansion of hematopoietic cells in a similar manner to small scale well or flask systems, allowing mean expansions of up to 2193-fold for total nucleated cells, 140-fold for CFU-GM and 66-fold for LTC-IC. Increasing the initial cell concentration from 5 x 10(3) to 1 x 10(5)CD34+ cells/ml induced the production of granulocytic cells with terminal differentiation while simultaneously decreasing the overall extent of expansion of the white blood cells produced. We tested the phagocytic activity and oxidative metabolism of the white blood cells produced. The percentage of phagocytic cells was 39+/-0.5% in expanded cultures derived from fractions initiated at 5 x 10(3), 10(4) or 10(5) cells/ml and 45+/-6% in cultured cells obtained from starting fractions containing 5 x 10(4) cells/ml, as compared to 58+/-4% in normal controls. A study of the potential for oxygen-dependent microbe killing showed that the expanded cells produced H2O2, although in lesser quantities than control cells. We subsequently investigated the possibility of freezing expanded cells. Total cell recovery after thawing was 45+/-4%, while recoveries of progenitors and stem cells ranged from 65 to 90%, without any influence of the initial cell concentration. This new approach could be of major interest for clinical practice, as it would allow evaluation of the quality of a graft prior to its infusion and employs experimental conditions which meet the criteria for potential clinical use.

Ex vivo expansion of autologous PB CD34+ cells provides a purging effect in children with neuroblastoma.

Peripheral blood CD34+ cell samples from eight children with advanced neuroblastoma and from 10 healthy adult donors were seeded at 5 x 10(4) cells/ml in stroma-free, serum-free medium with FL, SCF, MGDF (100 ng/ml each), G-CSF, IL6 (10 ng/ml each) and IL3 (5 ng/ml), and incubated for 10 days. The levels of expansion of PBCD34+ cells observed in neuroblastoma patients, with up to 214-fold expansion for total nucleated cells, 39-fold for CD34+ cells, 79-fold for CFU-GM and nine-fold for LTC-IC were identical to those obtained with PBCD34+ cells of healthy donors (P>/=0.5). All samples from patients with neuroblastoma and five donor's PBCD34+ cell samples contaminated with IMR-32 neuroblasts, were screened for the number of tyrosine hydroxylase (TH) mRNA transcript using LightCycler software. In all samples, progressive 1.9-4.4 log decreases in the number of TH transcripts were observed between days 0 and 10 of expansion. Our results show that in extensively pretreated children with neuroblastoma, the culture conditions that were effective for BM and CB cell expansion can generate an expansion of PBCD34+ cells and provide a purge of tumour cells.

Polymerase chain reaction: a method for monitoring tumor cell purge by long-term culture in BCR/ABL positive acute lymphoblastic leukemia.

We report the successful purging of leukemia cells bearing the Philadelphia chromosome and BCR/ABL transcripts by long-term marrow culture (LTC), and subsequent grafting of the purged marrow in a case of refractory acute lymphoblastic leukemia. The efficiency of the purge was evaluated by polymerase chain reaction (PCR) for BCR/ABL transcripts. In two LTCs initiated in the blastic stage, we demonstrated the selective effect of three culture media (serum dependent, serum-free (SF) supplemented or not with IL3 and GM-CSF) on the proliferative potential of normal hematopoietic (CFU-GM/BFU-E) and leukemic progenitors (CFU-ALL). BCR/ABL positive cells disappeared after 3 to 4 weeks of culture. The addition of IL3 and GM-CSF to the SF medium enhanced the growth of CFU-GM/BFU-E and shortened the purging period. We therefore carried out a LTC in the presence of IL3 and GM-CSF with marrow harvested in morphological remission. BCR/ABL positivity was detected at the outset, although no leukemia cells could be identified. The BCR/ABL was no longer found by PCR in the 7 and 14 day LTCs. The patient, consolidated by high dose polychemotherapy and total body irradiation, was infused with the 14 day LTC. This study indicates that PCR is a useful and sensitive technique for monitoring tumor cell reduction after LTC prior to autografting.

Plasma interacts with mafosfamide toxicity to normal haematopoietic progenitor cells: impact on in vitro marrow purging.

Bone marrow purging with cyclophosphamide derivatives in autologous bone marrow transplantation has demonstrated that the killing of leukemic cells and simultaneous preservation of normal progenitor cells depends on a number of parameters, in particular the haematocrit, nucleated cell concentration and nature of the cells. We have previously described a reliable experimental procedure for in vitro bone marrow treatment, based on individual adjustment of the drug dosage. The present study reveals an inhibitory action of plasma on the toxicity of mafosfamide to normal haematopoietic progenitor cells. In an initial series of 42 successive patients, determination of the CFU-GM lethal dose 95% (CFU-GM LD 95) showed this parameter to be inversely correlated to the nucleated cell concentration (NCC) (p < 0.001). Assuming the plasma content of the buffy coat cells (BC) to be higher in the less rich marrow samples, we then investigated the role of plasma in progenitor cell sensitivity to the drug. Results were as follows: (1) in the presence of 60% autologous or allogeneic plasma, CFU-GM LD 95 was increased by a factor of 2.18 +/- 0.35 or 1.98 +/- 0.23 respectively as compared to controls in a solution of 2% bovine serum albumin (p = 0.014), (2) this observation remained valid whatever the origin of the plasma and (3) the same was true whatever the nature of the cells, derived from normal donors or patients with haematological malignancies. These data suggest that plasma contains an inhibitor(s) of mafosfamide.(ABSTRACT TRUNCATED AT 250 WORDS)

The ex vivo expansion capacity of normal human bone marrow cells is dependent on experimental conditions: role of the cell concentration, serum and CD34+ cell selection in stroma-free cultures.

The present study was conducted to establish defined culture conditions for ex vivo expansion of normal human bone marrow cells. We investigated the role of three experimental expansion parameters: the cell concentration in the initial culture medium, the role of animal serum, human plasma and serum-free substitute, and the expansion potential of mononucleated cells (MNC) versus CD34+ cells. Cells were cultured in suspension with stem cell factor (SCF), IL3, IL6 and Erythropoietin (Epo) for 10 days. 1) Reducing the cell concentration from 3 x 10(4) to 1.5 x 10(3)/ml increased total cell expansion almost 20 fold, progenitor expansion more than 3 fold, and the maintenance of long term culture-initiating cells (LTC-IC). 2) In medium containing a serum-free substitute, total and CD34+ cell expansion was 3 times greater than in medium containing 1-10% human AB plasma or 25% animal serum. 3) The expansion potential of selected CD34+ cells was significantly greater than that of the total MNC population. However, taking into account the cell loss due to CD34+ selection, the overall results for quantitative expansion in relation to the initial number of MNCS favor the use of non-selected MNCS. 4) SCF + IL3 + IL6 was clearly the best combination of early cytokines for LTC-IC maintenance, with or without lineage-restricted cytokines, whereas the presence of IL1 beta in any combination augmented the decrease in LTC-IC. Addition of G-CSF to the medium resulted in 1 log increase in total cell expansion and a 2-fold increase in CFU-GM expansion. Addition of Epo always induced a dramatic proliferation of erythroid cells (up to 2000 fold) as well as of CFU-GM (up to 4 fold), without exhausting the LTC-IC pool. We concluded that the expansion of hemopoietic cells for clinical purposes needs establishment of controlled, reproducible and reliable culture conditions.

The role of recombinant haematopoietic growth factors in human long-term bone marrow culture in serum-free medium.

We have previously reported prolonged in vitro maintenance of human bone marrow progenitor cells using a serum-free (SF) liquid culture system. The present study was undertaken to determine recombinant growth factor (rGF) requirement of long-term marrow culture (LTMC) in absence of exogenous serum, to avoid interference of any undefined components. Our data clearly show that the presence of serum is a major obstacle to the correct evaluation of rGF activity. However, in SF conditions the sequential analysis of these rGFs, alone or in combination, clearly showed a stimulating activity of interleukin 3 (IL3), granulocyte-macrophage colony-stimulating factor (GM-CSF) and erythropoietin (EPO) on granulopoiesis and erythropoiesis. Indeed, the cumulative number of progenitors recovered during an 8-week period exceeded the initial input by a factor of 1.7 for granulocyte-macrophage (CFU-GM), of 3 for erythroid blast-forming units (BFU-E) and of 5.45 for CFU-E when EPO, GM-CSF and IL3 were combined. This study has confirmed that the system is able to sustain haematopoiesis for 8 weeks in a way similar to that in serum-dependent LTMC, despite diminished stromal adherent layer development which never covered more than 50% of the flask surface. We conclude that this defined SF-LTMC system provides a reproducible technique for studying human haematopoiesis.