Corticosteroids alter hematopoiesis in vitro by enhancing human monocyte secretion of granulocyte colony-stimulating factor.

The mechanism of corticosteroid alteration of hematopoiesis is not completely elucidated. Employing an endotoxin free system, we examined the mechanisms by which hydrocortisone succinate (HCS) enhanced human bone marrow (BM) colony forming unit granulocyte-macrophage (CFU-GM) proliferation. Interleukin-1beta (IL-1) (1 ng/mL), granulocyte-macrophage colony-stimulating factor (GM-CSF) (1 ng/mL), or the combination, induced minimal CFU-GM proliferation unless HCS was added to the cultures (10-25 vs. 80-125 colonies/4x10(5) BM mononuclear cells). Supernatants produced by incubating mononuclear cells with IL-1 + GM-CSF +/- HCS were examined for their capacity to induce CFU-GM proliferation: IL-1 and/or GM-CSF failed to induce supernatants capable of supporting CFU-GM proliferation unless HCS was present. Analysis of the cytokines produced by mononuclear cell subpopulations demonstrated that HCS markedly enhanced IL-1-induced monocyte secretion of granulocyte (G)-CSF. Furthermore, the minimal effective concentration of IL-1 required to induce G-CSF release was reduced 10-fold (from 1 to 0.1 ng/mL) and the G-CSF released was increased 5-fold at an IL-1 concentration of 1 ng/mL. In contrast, IL-1-induced monocyte secretion of tumor necrosis factor (TNF) was inhibited by HCS. HCS enhanced G-CSF secretion at physiologic concentrations (10 microg/dL), whereas progesterone had no effect. HCS alone had no effect on G-CSF secretion or mRNA expression while IL-1+HCS resulted in a 3-fold increase in G-CSF mRNA levels. These data suggest for the first time that corticosteroids increase secretion of an essential component of the lymphohematopoietic cytokine-growth factor system.

Reduction in carboplatin hematopoietic toxicity in tumor bearing mice: comparative mechanisms and effects of interleukin-1 beta and corticosteroids.

Increasing clinical evidence suggests that treatment of certain cancers is more effective with high dose chemotherapy compared to standard dose chemotherapy. Efforts at reduction of high dose chemotherapy hematotoxicity have focused on post-chemotherapy administration of hematopoietic growth factors or stem cells. A pretreatment strategy to induce hematopoietic resistance has not been extensively examined experimentally or clinically. Pretreatment with agents can provide an alternative or additional method to reduce high-dose chemotherapy hematotoxicity. We have previously described a murine model in which normal mice were injected with high dose carboplatin (CB, 600 mg/m2, intravenously). Within 7-12 days post-therapy, severe granulocytopenia and thrombocytopenia were induced and resulted in a granulocytopenic mortality of 70-80%. Utilizing this model, we observed that pretreatment of mice with interleukin (IL)-1 beta and/or a corticosteroid effectively reduced CB hematotoxicity. In the current studies, we demonstrated that C3H/HeJ mice bearing 0.25-0.5 cm RIF-1 tumors, exhibited a tumor associated decrease in hematopoietic tolerance to CB compared to normal mice. However, IL-1 beta (1 ug/mouse/day for 7 days), cortisone acetate (CA 0.5 mg/mouse/day for 7 days) or both CA and IL-1 beta, decreased CB induced mortality rates (control = 73%, IL-1 beta = 46%, CA = 30%, IL-1 beta + CA = 5%). Pretreatment with IL-1 beta, CA, or both ameliorated CB-induced absolute granulocyte, lymphocyte and platelet count nadirs. Bone marrow granulocyte-macrophage colony forming units (CFU-GM) from mice treated with IL-1 beta demonstrated increased ex-vivo resistance to cisplatin, without altering its sensitivity to high specific activity 3H-thymidine (a measure of cell fraction in S-phase). Bone marrow CFU-GM from mice treated with CA exhibited increased resistance to both cisplatin and to high specific activity 3H-thymidine. Pretreatment with IL-1 beta, CA or both did not alter tumor sensitivity to CB in vivo. These data suggest that IL-1 beta, CA or the combination may be clinically useful in reducing the hematotoxicity of CB.

Effects of endotoxin on proliferation of human hematopoietic cell precursors.

In examining the effects of corticosteroids on hematopoiesis in vitro, we observed that results were highly dependent on the lot of commercial fetal calf serum (FCS) utilized. We hypothesized that this variability correlated with the picogram (pg) level of endotoxin contaminating the FCS. Randomly obtained commercial lots of FCS contained 0.39 to 187 pg/ml of lipopolysaccharide (LPS). Standard FCS concentrations in hematopoietic precursor proliferation assays (granulocyte-marcrophage colony forming units [CFU-GM]) resulted in final LPS levels as high as 40 pg/ml. LPS (2-5 pg/ml) added to essentially endotoxin-free cultures, induced human mononuclear cell release of interleukin (IL)-1, IL-6 and granulocyte colony stimulating factor (G-CSF). Lots of FCS induced the release of IL-1, IL-6, and G-CSF from human mononuclear cells and the release of these factors correlated with the level of contaminating LPS. Human bone marrow CFU-GM proliferation, in response to granulocyte-macrophage colony stimulating factor (GM-CSF), positively correlated with the level of LPS contaminating the FCS and the FCS-induced release of IL-6 from mononuclear cells. CFU-GM proliferation of human bone marrow cluster of differentiation (CD) 34+CD14-cells were not affected by the presence of endotoxin. These data suggest that LPS at 2-5 pg/ml may induce bone marrow accessory cell release of hematopoietic growth factors, thus altering proliferative response of hematopoietic precursors and confounding the study of exogenously added cytokines to culture systems.

Corticosteroid alteration of carboplatin-induced hematopoietic toxicity in a murine model.

Corticosteroids exhibit extensive hematopoietic effects both in vitro and in vivo. Some of the previously studied effects suggested that corticosteroids may alter hematopoietic toxicity of chemotherapeutic agents. In this study, we examined (1) the optimum dose and schedule of cortisone acetate (CA) to reduce hematopoietic toxicity of carboplatin (CB) and (2) possible mechanisms involved in this protective effect. CA given subcutaneously at 0.5 mg/d per mouse for 7 days before CB reduced CB-induced mortality due to neutropenia from 88% in controls to 14% in CA-treated mice (P < .05). Lower CA doses were not effective. Three days of pretreatment (but not 1 day) was as effective as 7 days. CA given after CB had no effect on mortality. Pharmacokinetic studies of CA at 0.5 mg per mouse demonstrated blood levels of cortisol achievable in patients (peak level, 82 micrograms/dL). CA treatment markedly reduced spleen cell number and colony-forming units-granulocyte/macrophage (CFU-GM) as well as bone marrow CFU-GM. Bone marrow CFU-GM removed from CA-treated mice demonstrated increased resistance to platinum and increased resistance to high specific activity 3H-thymidine. These findings suggest that treatment of mice with CA induces cellular resistance of hematopoietic precursors to platinum and, thus, reduces CB hematotoxicity. CA or other corticosteroids may be useful in reducing clinical toxicity of CB.

Corticosteroid modulation of interleukin-1 hematopoietic effects and toxicity in a murine system.

Interleukin-1 (IL-1) has been shown to ameliorate the hematopoietic toxicities of antitumor chemotherapeutic agents in both mice and humans. However, IL-1 toxicity in humans is considerable and is similar to the systemic inflammatory toxicities induced by IL-3, IL-6, and other cytokines with pleiotropic biologic activities, eg, fever, nausea, malaise, and hypotension. We hypothesized that corticosteroids may reduce IL-1 toxicity without reducing IL-1 hematopoietic effects in vivo. C3H/HeJ mice (female, 6 weeks) were treated for 7 days subcutaneously with cortisone acetate (CA), (0.1, 0.25, or 0.5 mg/d/mouse), intraperitoneally with IL-1 (1 or 2 micrograms/d/mouse), or both. As expected, IL-1 increased white blood cell counts, splenic granulocyte-macrophage colony-forming units, and spleen cell number, and protected mice from lethal doses of carboplatin (200 mg/kg; Paraplatin, Bristol Laboratories, Evansville, IN) administered the day after completion of the 7 days of IL-1 administration. CA did not significantly block the hematopoietic effects of IL-1 or the ability of IL-1 to protect mice from the hematopoietic toxicity of carboplatin. IL-1 administered to mice at 8 micrograms/d/mouse for 5 days induced decreased activity, roughening of hair, diarrhea, pancytopenia, multiple metabolic abnormalities, and death in 60% of mice. IL-1 at the therapeutic doses (0.5 to 2 micrograms/d) was not toxic. CA in a dose-dependent manner blocked all of the above mentioned toxicities when administered 24 hours and 30 minutes before each IL-1 injection. CA also decreased IL-1-induced increase in plasma tumor necrosis factor levels at the time point examined.