Epsilon globin gene expression in developing human fetal tissues.

OBJECTIVE: The discovery of free fetal DNA in plasma of pregnant women has opened a new avenue for non-invasive prenatal diagnosis. We hypothesized that epsilon (ɛ)-globin gene expression could serve as a positive control for the presence of fetal nucleic acid. STUDY DESIGN: We measured ɛ-globin mRNA in human fetal tissues and compared concentrations with that measured in adult non-pregnant and pregnant samples. Total RNA was isolated from fetal marrow, liver, blood, and placenta (10-24 weeks gestation), from adult peripheral blood mononuclear cells, and from maternal plasma. RNA was reverse transcribed and quantitative polymerase chain reaction performed for ɛ-globin expression. RESULTS: ɛ-globin gene expression was detected in all fetal samples, was detected in plasma of pregnant women, but was negligible in non-pregnant samples. Relative ɛ-globin gene expression was significantly greater in fetal blood compared to fetal liver, and was minimally expressed in placenta. ɛ-globin gene expression decreased at the highest gestational ages in fetal blood, while expression was greatest at 15-19 weeks in fetal marrow. CONCLUSION: Fetal ɛ-globin gene expression is significantly greater than adult expression and is increased in maternal plasma compared to non-pregnant samples. ɛ-globin gene expression might serve as a positive control when determining the presence of fetal nucleic acid in total nucleic acid isolated from maternal plasma.

Urinary excretion of darbepoetin after intravenous vs subcutaneous administration to preterm neonates.

OBJECTIVE: Previous studies suggest that darbepoetin might stimulate erythropoiesis in preterm neonates more effectively if injected subcutaneously (s.c.) than if infused intravenously (i.v.). It has been postulated that this is because very high plasma concentrations after i.v. dosing result in urinary loss of the drug. However, this theory has not been tested systematically, and no direct comparisons have been made between s.c. and i.v. dosing of darbepoetin in preterm neonates. STUDY DESIGN: Preterm neonates were eligible for this pilot study if they were born at < or =32 weeks gestation with a weight of < or =1500 g, and had a hemoglobin < or =10.5 g/dl. The darbepoetin was given (4 microg/kg) i.v., over 4 h, if an i.v. was already in place and s.c. if no i.v. was in place. Urine was collected for drug quantification before dosing and for 48 h after. Blood was obtained for immature reticulocyte fraction (IRF), absolute reticulocyte count (ARC) and reticulocyte % before and 96 h after dosing. RESULTS: Ten preterm neonates were studied: five received i.v. and five received s.c. darbepoetin. No adverse effects of the administrations were detected. IRF, ARC and reticulocyte % increased in the i.v. and s.c. recipients, and no difference in magnitude of increase was apparent between the groups. Before the darbepoetin was administered, none of the patients had any erythropoietin (Epo) detected in their urine. After i.v. dosing, no darbepoetin was detected in any of the samples, on any of the subjects, over the subsequent 48 h. After s.c. dosing, three of the patients had minimal urinary Epo detected. The patient with the largest urinary loss of drug had only 0.13% of the administered dose detected in the urine. Thus, essentially no urinary loss of drug was observed following either i.v. or s.c. darbepoetin dosing. CONCLUSION: Darbepoetin loss into the urine was below detectable limits among seven patients, while three had minimally detectable urinary losses. i.v. and s.c. dosing resulted in approximately equivalent increases in reticulocyte response when measured 96 h after dosing. On this basis, we speculate that if a patient who is to receive darbepoetin has an i.v. in place, the drug can be given i.v., thus avoiding any discomfort associated with an s.c. injection.

Intravenous administration of darbepoetin to NICU patients.

OBJECTIVE: Darbepoetin can be administered either intravenously (i.v.) or subcutaneously (s.c.). However, no information is available regarding pharmacokinetics and pharmacodynamics following its i.v. administration to neonates. STUDY DESIGN: We administered a single i.v. dose (4 microg/kg) of darbepoetin to 10 neonates who had a hemoglobin < or =10.5 g/dl. Blood was obtained for immature reticulocyte fraction (IRF) and absolute reticulocyte count (ARC), before and 48 h following the dose. Blood was also drawn for pharmacokinetic analysis once before and at four pre-set intervals after dosing. RESULTS: The study subjects ranged from 704 to 3025 g (median, 1128 g) birth weight, and were 26.0-40.0 weeks (median, 29.2 weeks) gestation at delivery. When the darbepoetin dose was given, ages ranged from 3 to 28 days (median, 8.5 days) with a hemoglobin of 9.8 +/- 0.7 g/dl (mean +/- s.d.). Doses were administered by i.v. infusion over 4 h. No adverse effects of the infusions were detected. The half-life of darbepoetin (t (1/2)) was 10.1 h (range 9.0-22.7 h), the volume of distribution was 0.77 l/kg (range 0.18-3.05 l/kg) and the clearance was 52.8 ml/h/kg (range 22.4-158.0 ml/h/kg). In the preterm neonates, there was no significant correlation between gestational age, or age at darbepoetin administration, and pharmacokinetic parameters. However, in the term and near-term neonates, volume of distribution correlated significantly with both gestational and age at darbepoetin administration (P < 0.05). Forty-eight hours after dosing, the IRFs and ARCs were elevated in six subjects and not in four. Those with predosing reticulocyte counts >200,000/microl did not have an increase in reticulocytes by 48 h (P < 0.05). CONCLUSIONS: Darbepoetin administered i.v. to neonates had a shorter t (1/2), a larger volume of distribution and more rapid clearance than reported in children. We observed a significantly shorter t (1/2) and a less consistent rise in IRF and ARC after i.v. dosing than we previously reported following 4 microg/kg administered SC.

Erythropoietin and hypoxia inducible factor-1 expression in the mid-trimester human fetus.

AIM: Infants born prematurely lack a normal response to anemia and fail to increase erythropoietin (Epo) production despite an apparent need for improved tissue oxygenation. This anemia may involve a deficiency in the fetal and premature kidney to produce Epo. To evaluate fetal Epo production, Epo and hypoxia inducible factor-1 (HIF) mRNA expression was measured in the mid-trimester human fetus. METHODS: Fetal liver and kidney samples were obtained at 11-22 wk of gestation. RNA was isolated and reverse transcribed from snap-frozen specimens. Epo and HIF cDNA concentrations were determined using real-time polymerase chain reaction (PRISM). Epo cDNA concentrations were standardized to HIF concentrations present in each sample. RESULTS: HIF concentrations remained constant during gestation in kidney and liver samples. Epo cDNA concentrations in kidney did not change from 12 to 22 wk (8.4 +/- 3.4 fg Epo pg(-1) HIF cDNA, 4.8 +/- 1.4, 2.6 +/- 0.4, and 4.2 +/- 1.8 at 11-14, 15-16, 17-19, and 20-22 wk of gestation, respectively), while Epo cDNA concentrations in liver increased with gestation (74.5 +/- 31.9 fg pg(-1) HIF, 23.8 +/- 6.5, 96.4 +/- 19.2 and 276.1 +/- 28.5 at 12-14, 15-16, 17-19 and 20-22 wk of gestation, respectively, p < 0.05, 20-22 wk of gestation liver samples vs all other gestations). Concentrations were 5-20-fold higher in liver than in kidney in each gestational group (p < 0.01, liver vs kidney). CONCLUSION: HIF concentrations did not change with gestation in liver or kidney. The human fetal kidney produced approximately 5% of the total Epo mRNA measured during the second trimester. It remains to be determined how Epo production by these tissues is affected by premature birth.

Effects of early erythropoietin therapy on the transfusion requirements of preterm infants below 1250 grams birth weight: a multicenter, randomized, controlled trial.

OBJECTIVES: Infants of

Effects of granulocyte colony-stimulating factor on neutrophil adhesive molecules in neonates.

BACKGROUND: Term and preterm neonates experience quantitative and qualitative neutrophil deficiencies resulting in part from decreased production of granulocyte colony-stimulating factor (G-CSF). In adults, G-CSF improves neutrophil function by up-regulating adhesion molecules. PATIENTS AND METHODS: To evaluate the effects of G-CSF on neonatal neutrophil adhesive phenotypes, cord blood samples were incubated with G-CSF or phosphate-buffered saline and stimulated with N-formyl-methionyl-leucyl-phenylalanine (FMLP), and adhesion molecules were evaluated by flow cytometry. RESULTS: In term and preterm neutrophils, G-CSF incubation increased beta2-integrin expression significantly compared with baseline and to a greater extent than observed in adult neutrophils. With FMLP stimulation, beta2-integrin expression increased even more in the G-CSF group. L-selectin expression decreased after G-CSF incubation and decreased even more with FMLP stimulation in the G-CSF group compared with the phosphate-buffered saline group in term and preterm samples, but not in adult samples. CONCLUSIONS: The data show that G-CSF increases expression of beta2-integrin and decreases expression of L-selectin on unstimulated and stimulated term and preterm neonatal neutrophils in vitro. Further study is required to determine whether G-CSF improves neonatal neutrophil function.

The use of erythropoietin in neonates.

Although much information has been accumulated about the clinical use of Epo in preterm infants, many questions remain unanswered. The evolution of clinical practice in the care of extremely ill, preterm infants continues to affect the number of transfusions required during hospitalization. Decreasing phlebotomy losses and instituting standardized transfusion guidelines have both been shown significantly to decrease the transfusion requirements of preterm infants. The administration of Epo likely decreases transfusion need even further; however, the direct impact of each of these actions has not been studied prospectively. It is likely that the combination of instituting rigorous and standardized transfusion guidelines, decreasing phlebotomy losses, and the appropriate use of Epo will have the greatest impact in decreasing transfusion requirements in all preterm and term neonates, regardless of the cause of their anemia.

Endothelin concentrations in preterm infants treated with human recombinant erythropoietin.

BACKGROUND: Increased endothelin-1 (ET-1) production following recombinant erythropoietin (Epo) administration is a presumed etiology for the hypertension reported in some adults. It is unknown whether Epo has similar effects in preterm infants. METHODS: Serum ET-1 and Epo concentrations were measured prior to study, and following the second and third doses in 20 preterm infants receiving intravenous (IV) or subcutaneous (SC) Epo. Blood pressures were monitored prior to Epo administration and during the first, second, and third dose. RESULTS: Infants (963 +/- 54 g birth weight, 27.4 +/- 0.6 weeks gestational age, 18 +/- 3 days of life; mean +/- SEM) had baseline Epo concentrations of 5.5 +/- 1.3 mU/ml and ET-1 concentrations below the lower limits of detection (<1 pg/ml). Epo concentrations were 1,848 +/- 274 and 1,672 +/- 295 mU/ml following the second and third IV dose, respectively, while Epo concentrations were 420 +/- 92 and 290 +/- 35 mU/ml after the second and third SC dose, respectively (p < 0.005, SC versus IV). ET-1 concentrations remained below the limits of detection in all but 6 infants, whose concentrations were <3.3 pg/ml. Blood pressures did not increase above baseline in either group during the study period. CONCLUSION: Despite the wide range of Epo concentrations measured, no correlation was observed between Epo concentrations, ET-1 concentrations, and blood pressure during the 1-week study period. The long-term effects of Epo on ET-1 concentrations and blood pressure in preterm infants require further study.

Erythropoietin to prevent and treat the anemia of prematurity.

Human recombinant erythropoietin was first cloned in 1985, and is currently available for clinical use for a variety of anemias. Following successful clinical trials using erythropoietin to treat adults with the anemia of end-stage renal disease, the first clinical trial evaluating the use of erythropoietin in preterm infants to treat anemia was published in 1990. Since that initial report, numerous clinical trials have reported various levels of success in the treatment of this anemia. Most recently, erythropoietin has been used in the first weeks of life in an attempt to prevent the anemia of prematurity. This review describes mechanisms of erythropoiesis in the fetus and neonate, and focuses on recent clinical trials evaluating the use of erythropoietin to prevent and treat anemia in preterm infants.

The effect of erythropoietin on the transfusion requirements of preterm infants weighing 750 grams or less: a randomized, double-blind, placebo-controlled study.

BACKGROUND: Clinical trials of erythropoietin (EPO) administration to preterm infants have not focused on infants weighing 750 gm or less, the population most likely to receive multiple transfusions because of large phlebotomy losses. It is unknown whether preterm infants weighing 750 gm or less will respond to EPO by accelerating erythropoiesis, or whether EPO administered to this population will decrease blood transfusions. METHODS: We randomly assigned 28 extremely low birth weight preterm infants (mean +/- SEM: 24.7 +/- 0.3 weeks' gestation, 662 +/- 14 gm birth weight), in the first 72 hours of life, to receive either EPO (200 U/kg/day) or placebo for 14 days and administered transfusions only according to protocol over a 21-day study period. All infants received 1 mg/kg/day iron dextran in their total parenteral nutrition solution during the 14-day treatment period. RESULTS: During the 21-day study period, a lower number and volume of transfusions were received by the EPO recipients (4.7 +/- 0.7 transfusions per patient and 70 +/- 11 ml/kg per patient) than by the placebo recipients (7.5 +/- 1.1 transfusions per patient and 112 +/- 17 ml/kg per patient; p < 0.05, EPO vs placebo), whereas hematocrits remained similar in the two groups. Reticulocyte counts were similar in both groups on day 1 but were greater in the EPO recipients on day 14 (EPO day 1, 351 +/- 53; EPO day 14, 359 +/- 40 x 10(3)/microl; placebo day 1, 334 +/- 64; placebo day 14, 120 +/- 10 x 10(3)/microl; p < 0.01, EPO vs placebo). Serum ferritin concentrations were similar in both groups at the beginning of the study but were greater in the placebo recipients by day 14 (EPO, 262 +/- 44 microg/L; placebo, 593 +/- 92 microg/L; p < 0.01). No adverse effects of EPO or iron were noted. CONCLUSION: The combination of EPO and parenteral iron stimulates erythropoiesis in preterm infants weighing 750 gm or less and results in fewer transfusions during their first 3 weeks of life.

Serum erythropoietin concentrations fail to increase after significant phlebotomy losses in ill preterm infants.

UNLABELLED: OBJECTIVE/STUDY DESIGN: After blood loss, production of erythropoietin in adults increases, which accelerates erythropoiesis and restores the erythroid mass. It is unclear whether preterm infants with large phlebotomy losses have a similar response. We therefore measured serum erythropoietin concentrations in 11 ill preterm infants (1057 +/- 167 gm) as their phlebotomy losses accumulated. RESULTS: Before the first transfusion, erythropoietin concentrations were 68.9 +/- 36.2 mU/ml (range 0 to 205 mU/ml) at 5 ml/kg blood out, 17.4 +/- 8.9 mU/ml at 10 ml/kg, and 4.8 +/- 2.6 mU/ml at 15 ml/kg. Erythropoietin concentrations did not increase in any patients despite increasing phlebotomy losses. CONCLUSION: Serum erythropoietin concentrations in ill preterm infants do not increase in the face of significant blood loss. Although the mechanistic explanation for this failure is unclear, it likely contributes to the transfusion requirements of this population.

Stability of human recombinant epoetin alfa in commonly used neonatal intravenous solutions.

OBJECTIVE: To measure epoetin alfa concentrations after adding it to a variety of commonly used neonatal intravenous fluids to determine the stability of epoetin alfa over time. DESIGN: Epoetin alfa was added to the following fluids: sterile water; NaCl 0.9%; dextrose 10% in water; dextrose 10% with albumin at concentrations of 0.01%, 0.05%, and 0.1%; and total parenteral nutrition solution containing either 0.5% or 2.25% amino acids. The fluid was administered through intravenous tubing, a T-connector, and catheter, and samples were collected at 0, 2, 4, 8, and 24 hours. MAIN OUTCOME MEASURES: Epoetin alfa concentrations were compared with the measured original preinfusion concentration and recorded as the percentage recovered. RESULTS: Concentrations declined significantly in all fluids containing less than 0.05% protein, but remained stable over 24 hours in fluids containing 0.05% or more protein. CONCLUSIONS: We conclude that epoetin alfa should be mixed in intravenous fluids containing at least 0.05% protein.

Pharmacokinetics and effectiveness of recombinant erythropoietin administered to preterm infants by continuous infusion in total parenteral nutrition solution.

OBJECTIVES: To compare the pharmacokinetics and effectiveness of continuously administered recombinant erythropoietin (Epo) in total parenteral nutrition (TPN) solution with daily subcutaneously administered Epo. METHODS: Forty preterm infants in the first 72 hours of life were randomly assigned to receive Epo (200 units/kg per day for 10 consecutive days), either subcutaneously (20 infants, 1051 +/- 40 gm, 28.3 +/- 0.4 weeks of gestation; mean +/- SEM), or added daily to their TPN fluids (20 infants, 1028 +/- 36 gm, 27.9 +/- 0.4 weeks of gestation). Both groups received iron supplementation (1 mg/kg per day iron dextran in the TPN solution). Absolute reticulocyte counts and complete blood cell counts with differentials were measured, and transfusions and phlebotomy losses were recorded. Pharmacokinetics were determined in the first 16 infants. RESULTS: In the infants who received Epo subcutaneously, the elimination half-life was 17.6 +/- 4.4 hours on day 3 and 11.2 +/- 1.5 hours on day 10; the volume of distribution was 802 +/- 190 ml/kg on day 3 and 1330 +/- 243 m/kg on day 10. Serum Epo concentrations were higher on day 3 than on day 10 for both groups (subcutaneous: 400 +/- 64 mU/ml vs 177 +/- 29 mU/m, p <0.05; TPN: 395 +/- 64 vs 194 +/- 41 mU/ml, p <0.05). Clearance did not differ between the two groups with regard to route of administration and increased significantly from days 3 to 10 in both groups. Reticulocyte counts were similar in both groups. There were no differences between groups in the number of transfusions given, and the overall decline in hematocrit was similar. No adverse effects of Epo were noted in either group. CONCLUSIONS: Adding Epo to the TPN solution in this population results in similar Epo concentrations, clearance, and effectiveness as subcutaneous dosing.

Maternal and umbilical serum concentrations of granulocyte colony-stimulating factor and its messenger RNA during clinical chorioamnionitis.

OBJECTIVE: To determine whether women with chorioamnionitis developed elevated serum granulocyte colony-stimulating factor (G-CSF) concentrations and, if so, whether the G-CSF crossed to the fetus. METHODS: We quantified G-CSF before, during, and after delivery and in the cord blood of 12 women with clinical chorioamnionitis and 12 matched controls. We also measured G-CSF messenger RNA (mRNA) transcripts in mononuclear cells isolated from maternal and cord blood at delivery. RESULTS: At study entry, G-CSF concentrations were much higher in women with chorioamnionitis (1397 +/- 950 pg/mL, mean +/- standard error of mean) than in controls (17 +/- 5 pg/mL, P < .05). At delivery, concentrations had increased in both groups (2752 +/- 1100 pg/mL in the chorioamnionitis group, 165 +/- 61 pg/mL in controls; P < .05). After delivery, G-CSF concentrations had diminished to 839 +/- 594 pg/mL in women with chorioamnionitis and to 83 +/- 16 pg/mL in controls (P < .05). Concentrations in the cord blood were 2729 +/- 974 pg/mL for the chorioamnionitis group and 51 +/- 17 pg/mL for controls (P < .05). Granulocyte colony-stimulating factor mRNA transcripts were more abundant in women with chorioamnionitis than in controls but were scarce in all matched cord blood samples tested. CONCLUSION: Serum G-CSF concentrations were elevated during clinical chorioamnionitis, and similar levels were found in maternal and cord serum. Because G-CSF mRNA levels were very low in cord mononuclear cells, the G-CSF in cord serum might be maternal in origin.

Neutrophil pool sizes and granulocyte colony-stimulating factor production in human mid-trimester fetuses.

We quantified neutrophils and neutrophil progenitors, and assessed granulocyte colony-stimulating factor (G-CSF) production in the liver and bone marrow of 20 human abortuses after elective pregnancy termination between 14 and 24 wk of gestation. Mature neutrophils were not observed in any of the liver specimens, but were present in the bone marrow as early as 14 wk. The concentrations of neutrophils in the fetal marrow were extremely low, by comparison with term infants and adults, with less than 5% of the nucleated cells being segmented neutrophils, band neutrophils, or metamyelocytes compared with 31-69% in term infants. Despite the low neutrophil populations, progenitors which had the capacity for clonal maturation into neutrophils in vitro were abundant in the fetal liver and fetal bone marrow. In addition, such progenitors had a dose-response relationship to recombinant G-CSF similar to that of progenitors from the bone marrow of healthy adults. At each gestational age tested, stimulation of mononuclear cells from fetal liver with IL-1 alpha generated less G-CSF protein and fewer G-CSF mRNA transcripts than did stimulation of mononuclear cells from fetal bone marrow. Mononuclear cells from the fetal bone marrow produced less G-CSF protein and mRNA than did mononuclear cells from the blood of adults. Thus, the liver of the mid-trimester human fetus is almost devoid of neutrophils, and the bone marrow contains a significantly lower proportion of neutrophils than does the marrow of term neonates or adults. These findings correlate with IL-1 alpha-induced production of G-CSF in these organs.(ABSTRACT TRUNCATED AT 250 WORDS)

Efficacy and cost analysis of treating very low birth weight infants with erythropoietin during their first two weeks of life: a randomized, placebo-controlled trial.

OBJECTIVE: We hypothesized that using a higher dose of erythropoietin (Epo) and starting treatment on the first day of life would reduce the transfusion requirements of ventilator-dependent and non-ventilator-dependent very low birth weight (VLBW) infants. Moreover, we hypothesized that this treatment would be cost-effective. METHODS: We randomly assigned 20 ill newborn VLBW infants to receive either Epo (200 units/kg per day) or placebo during their first 2 weeks of life. The caregivers were unaware of the treatment assignments, and erythrocyte transfusions were administered according to hematocrit and signs of anemia. RESULTS: On day 1, reticulocyte counts and hematocrits were similar in the two groups. During the subsequent 2 weeks, reticulocyte counts of the placebo recipients fell significantly below those of the Epo recipients, but hematocrits in the two groups did not differ. More transfusions were received by the placebo recipients (mean = 1.4 per patient) than by the Epo recipients (mean = 0.2 per patient; p < 0.01). No adverse effects of Epo were noted, and the costs in the placebo group exceeded those in the Epo group. CONCLUSIONS: We conclude that administration of Epo to VLBW infants during the first 2 weeks of life results in fewer transfusions and is cost-effective.

Possible mechanisms accounting for the growth factor independence of hematopoietic progenitors from umbilical cord blood.

Hematopoietic progenitors obtained from the bone marrow of healthy adults fail to undergo clonogenic maturation in vitro if a source of hematopoietic growth factors is not included in the culture dishes. In contrast, a fraction of similarly purified progenitors obtained from umbilical cord blood undergo clonogenic maturation even in the absence of added growth factors. We postulated that production of hematopoietic growth factors within the culture dishes containing the progenitors of umbilical cord blood origin might be responsible. We postulated further, that this production might be by non-progenitor cells co-plated along with the progenitors, or alternatively by CD34+ cells themselves, or by cells clonally derived from CD34+ cells. To test these possibilities we first assessed the effect of including in the cultures neutralizing antibody directed against various growth factors. Inclusion of anti-granulocyte macrophage colony-stimulating factor (GM-CSF) and anti-interleukin-3 (IL-3) (but not anti-IL-2) significantly reduced the growth factor independence of cord blood progenitors (P < .005 and P < .01). Inclusion of both anti-GM-CSF and anti-IL-3 almost completely ablated the spontaneous colony growth (P < .001). Inclusion of IL-10 also reduced, in a concentration-dependent fashion, the spontaneous generation of umbilical cord blood-derived colonies. Transcripts for GM-CSF and IL-3 were detected, by reverse transcriptase-polymerase chain reaction (RT-PCR), in the CD34+ cells from cord blood and from adult marrow. When plated without added growth factors, however, the CD34+ cells of adult marrow origin failed to produce colonies, whereas 6% of cord blood CD34+ cells similarly cultured did so. When these growth factor independent colonies were plucked from culture, transcripts for GM-CSF and IL-3 were identified in all. We conclude that production of GM-CSF and IL-3 occurs within culture dishes containing hematopoietic progenitors of umbilical cord origin, and that this explains some of their apparently unique features of in vitro growth.

Pilot study comparing recombinant erythropoietin alone with erythropoietin plus recombinant granulocyte-macrophage colony-stimulating factor for treatment of the anemia of prematurity.

Sixteen infants with the anemia of prematurity were randomly assigned to treatment with packed erythrocyte transfusions, recombinant erythropoietin alone, or epo plus recombinant granulocyte-macrophage colony-stimulating factor (GM-CSF). During the treatment period, blood reticulocyte concentrations remained unchanged in those randomly assigned to receive transfusions, but increased significantly in those who received erythropoietin, either alone or in combination with GM-CSF. The magnitude of increase in hematocrit was not greater in those who received erythropoietin plus GM-CSF than in those who received erythropoietin alone. Blood neutrophil concentrations fell in all four infants who received erythropoietin alone, but increased in all who received erythropoietin plus GM-CSF.

Effect of recombinant stem cell factor on clonogenic maturation and cycle status of human fetal hematopoietic progenitors.

Studies were undertaken to delineate the actions of stem cell factor (SCF) on human fetal hematopoietic progenitors in vitro. Mononuclear cells from umbilical cord blood of term fetuses were "panned" immunologically, and the resulting hematopoietic progenitors were grown in methylcellulose culture containing various concentrations of SCF alone or in combination with other recombinant hematopoietic growth factors. Neutralizing antibodies to IL-3 and granulocyte-macrophage colony-stimulating factor were added to all plates to which recombinant IL-3 or granulocyte-macrophage colony-stimulating factor were not included to decrease any confounding effect resulting from production of small quantities of these factors within the culture plates. SCF, as a single agent, supported clonogenic maturation of fetal granulocyte-macrophage progenitors (granulocyte-macrophage colony-forming unit, p < 0.05), multipotent progenitors (CFU-MIX, p < 0.05), and erythroid progenitors (erythroid burst-forming unit, p < 0.05). When combined with subplateau concentrations (0.1 microgram/L) of IL-3 or granulocyte-macrophage colony-stimulating factor, SCF had an additive or synergistic effect on clonogenic maturation of granulocyte-macrophage colony-forming unit and CFU-MIX. When combined with higher concentrations (5.0 micrograms/L) of IL-3 or granulocyte-macrophage colony-stimulating factor, SCF generally did not enhance colony formation but did increase the number of cells per colony. Like other pleiotropic cytokines such as IL-6, IL-9, and IL-11, SCF had a broad spectrum of action of fetal hematopoietic progenitors.