Beneficial Effects of Soluble Guanylyl Cyclase Stimulation and Activation in Sickle Cell Disease Are Amplified by Hydroxyurea: In Vitro and In Vivo Studies.

The complex pathophysiology of sickle cell anemia (SCA) involves intravascular hemolytic processes and recurrent vaso-occlusion, driven by chronic vascular inflammation, which result in the disease's severe clinical complications, including recurrent painful vaso-occlusive episodes. Hydroxyurea, the only drug frequently used for SCA therapy, is a cytostatic agent, although it appears to exert nitric oxide/soluble guanylyl cyclase (sGC) modulating activity. As new drugs that can complement or replace the use of hydroxyurea are sought to further reduce vaso-occlusive episode frequency in SCA, we investigated the effects of the sGC agonists BAY 60-2770 (sGC activator) and BAY 41-2272 (sGC stimulator) in the presence or absence of hydroxyurea on SCA vaso-occlusive mechanisms and cell recruitment both ex vivo and in vivo. These agents significantly reduced stimulated human SCA neutrophil adhesive properties ex vivo in association with the inhibition of surface ß2-integrin activation. A single administration of BAY 60-2770 or BAY 41-2272 decreased tumor necrosis factor cytokine-induced leukocyte recruitment in a mouse model of SCA vaso-occlusion. Importantly, the in vivo actions of both agonists were significantly potentiated by the coadministration of hydroxyurea. Erythroid cell fetal hemoglobin (HbF) elevation is also a major goal for SCA therapy. BAY 41-2272 but not BAY 60-2770 at the concentrations employed significantly induced γ-globin gene transcription in association with HbF production in cultured erythroleukemic cells. In conclusion, sGC agonist drugs could represent a promising approach as therapy for SCA, for use either as stand-alone treatments or in combination with hydroxyurea. SIGNIFICANCE STATEMENT: This preclinical study demonstrates that stimulators and activators of sGC are potent inhibitors of the adhesion and recruitment of leukocytes from humans and in mice with sickle cell anemia (SCA) and may represent a promising approach for diminishing vaso-occlusive episode frequency in SCA. Hydroxyurea, a drug already frequently used for treating SCA, was found to potentiate the beneficial effects of sGC agonists in in vivo studies, implying that these classes of compounds could be used alone or in combination therapy.

Targeted disruption of cd39/ATP diphosphohydrolase results in disordered hemostasis and thromboregulation.

CD39, or vascular adenosine triphosphate diphosphohydrolase, has been considered an important inhibitor of platelet activation. Unexpectedly, cd39-deficient mice had prolonged bleeding times with minimally perturbed coagulation parameters. Platelet interactions with injured mesenteric vasculature were considerably reduced in vivo and purified mutant platelets failed to aggregate to standard agonists in vitro. This platelet hypofunction was reversible and associated with purinergic type P2Y1 receptor desensitization. In keeping with deficient vascular protective mechanisms, fibrin deposition was found at multiple organ sites in cd39-deficient mice and in transplanted cardiac grafts. Our data indicate a dual role for adenosine triphosphate diphosphohydrolase in modulating hemostasis and thrombotic reactions.

Hematopoietic progenitor cell rolling in bone marrow microvessels: parallel contributions by endothelial selectins and vascular cell adhesion molecule 1.

We have used intravital microscopy to study physiologically perfused microvessels in murine bone marrow (BM). BM sinusoids and venules, but not adjacent bone vessels, supported rolling interactions of hematopoietic progenitor cells. Rolling did not involve L-selectin, but was partially reduced in wild-type mice treated with antibodies to P- or E-selectin and in mice that were deficient in these two selectins. Selectin-independent rolling was mediated by alpha4 integrins, which interacted with endothelial vascular cell adhesion molecule (VCAM)-1. Parallel contribution of the endothelial selectins and VCAM-1 is not known to direct blood cell trafficking to other noninflamed tissues. This combination of constitutively expressed adhesion molecules may thus constitute a BM-specific recruitment pathway for progenitor cells analogous to the vascular addressins that direct selective lymphocyte homing to lymphoid organs.

P-Selectin glycoprotein ligand 1 (PSGL-1) is expressed on platelets and can mediate platelet-endothelial interactions in vivo.

The platelet plays a pivotal role in maintaining vascular integrity. In a manner similar to leukocytes, platelets interact with selectins expressed on activated endothelium. P-selectin glycoprotein ligand 1 (PSGL-1) is the main P-selectin ligand expressed on leukocytes. Searching for platelet ligand(s), we used a P-selectin-immunoglobulin G (IgG) chimera to affinity purify surface-biotinylated proteins from platelet lysates. P-selectin-bound ligands were eluted with ethylenediaminetetraacetic acid. An approximately 210-kD biotinylated protein was isolated from both human neutrophil and platelet preparations. A band of the same size was also immunopurified from human platelets using a monoclonal anti-human PSGL-1 antibody and could be blotted with P-selectin-IgG. Under reducing conditions, both the predicted PSGL-1 approximately 210-kD dimer and the approximately 120-kD monomer were isolated from platelets. Comparative immunoelectron microscopy and Western blotting experiments suggested that platelet PSGL-1 expression is 25-100-fold lower than that of leukocytes. However, patients with chronic idiopathic thrombocytopenic purpura who harbor predominantly young platelets displayed greater expression, indicating that PSGL-1 expression may be decreased during platelet aging. By flow cytometry, thrombin-activated platelets from normal individuals exhibited greater expression than those unstimulated. An inhibitory anti-PSGL-1 antibody significantly reduced platelet rolling in mesenteric venules, as observed by intravital microscopy. Our results indicate that functional PSGL-1 is expressed on platelets, and suggest an additional mechanism by which selectins and their ligands participate in inflammatory and/or hemostatic responses.

Role of P-selectin cytoplasmic domain in granular targeting in vivo and in early inflammatory responses.

P-selectin is an adhesion receptor for leukocytes expressed on activated platelets and endothelial cells. The cytoplasmic domain of P-selectin was shown in vitro to contain signals required for both the sorting of this protein into storage granules and its internalization from the plasma membrane. To evaluate in vivo the role of the regulated secretion of P-selectin, we have generated a mouse that expresses P-selectin lacking the cytoplasmic domain (DeltaCT mice). The deletion did not affect the sorting of P-selectin into alpha-granules of platelets but severely compromised the storage of P-selectin in endothelial cells. Unstored P-selectin was proteolytically shed from the plasma membrane, resulting in increased levels of soluble P-selectin in the plasma. The DeltaCT-P-selectin appeared capable of mediating cell adhesion as it supported leukocyte rolling in the mutant mice. However, a secretagogue failed to upregulate leukocyte rolling in the DeltaCT mice, indicating an absence of a releasable storage pool of P-selectin in the endothelium. Furthermore, the neutrophil influx into the inflamed peritoneum was only 30% of the wild-type level 2 h after stimulation. Our results suggest that different sorting mechanisms for P-selectin are used in platelets and endothelial cells and that the storage pool of P-selectin in endothelial cells is functionally important during early stages of inflammation.

Cytokine-induced meningitis is dramatically attenuated in mice deficient in endothelial selectins.

Leukocyte accumulation in cerebrospinal fluid and disruption of the blood-brain barrier are central components of meningitis and are associated with a poor prognosis. Genetically engineered deficiencies or functional inhibition of endothelial leukocyte adhesion receptors P-, or P- plus E-selectins, lead to deficits in leukocyte rolling and extravasation. However, their impact on meningeal inflammation has not been tested previously. An acute cytokine-induced meningitis model associated with significant cerebrospinal fluid leukocyte accumulation (averaging 14,000 leukocytes/microl as early as 4 h) and blood-brain barrier permeability was developed in adult mice. This model was applied to mice deficient in P-selectin and mice doubly deficient in P- and E-selectins. Partial inhibition of cerebrospinal fluid leukocyte influx and permeability was noted in P-selectin-deficient mice. Mice doubly deficient in P- and E-selectins displayed a near complete inhibition of these parameters. Our results suggest that P- and E-selectins cooperatively contribute to meningitis and that functional blocking of both endothelial selectins in conjunction with antibiotics may provide a therapeutic approach for treatment of bacterial meningitis.

The combined role of P- and E-selectins in atherosclerosis.

P- and E-selectins are adhesion molecules mediating the first step in leukocyte extravasation. Because their function in leukocyte adhesion is overlapping, we hypothesized that there might be a combined effect of these selectins on the development of atherosclerotic lesions. We bred P- and E-selectin-double-deficient mice onto the low-density lipoprotein receptor (LDLR)-deficient background (LDLR-/- P/E-/-) and compared lesion development in these mice to that in mice wild type for both selectins (LDLR-/- P/E+/+). After 8 wk on atherogenic diet, the LDLR-/- P/E-/- mice developed fatty streaks in the aortic sinus that were five times smaller than those in LDLR-/- P/E+/+ mice. The density of macrophages in the fatty streaks was comparable between LDLR-/- P/E+/+ and LDLR-/- P/E-/- mice. After 22 wk on the diet, the lesions spread throughout the aorta but this process was delayed in LDLR-/- P/E-/- mice. At 37 wk on diet, the lesions progressed to the fibrous plaque stage in both genotypes. However, the lesions in the aortic sinus in LDLR-/- P/E-/- mice were 40% smaller and less calcified than those of LDLR-/- P/E +/+ mice. Our results suggest that P- and E-selectins together play an important role in both early and advanced stages of atherosclerotic lesion development.

Insights into selectin function from knockout mice.

The development of animal models through gene targeting was very useful to the selectin field. Selectins are found on endothelium, platelets and leukocytes and, they mediate adhesion among these cell types. The removal of a single selectin gene taught us that P-selectin on the vessel wall mediates leukocyte rolling in the absence of inflammation and that all three selectins contribute to leukocyte rolling during inflammation. Similarly, P-selectin is responsible for early neutrophil recruitment while the other selectins contribute in later stages. The knockout animals also confirmed the important role of L-selectin in lymphocyte homing. Removal of both endothelial selectins uncovered the hidden importance of E-selectin in leukocyte homeostasis and showed that the endothelial selectins were as important for leukocyte extravasation as the leukocyte beta 2 integrins. The submission of selectin-deficient mice to models of various human diseases can provide invaluable information on conditions in which an anti-selectin therapy may prove beneficial.

Experimental murine acid aspiration injury is mediated by neutrophils and the alternative complement pathway.

Acid aspiration may result in the development of the acute respiratory distress syndrome, an event associated with significant morbidity and mortality. Although once attributed to direct distal airway injury, the pulmonary failure after acid aspiration is more complex and involves an inflammatory injury mediated by complement (C) and polymorphonuclear leukocytes. This study examines the injurious inflammatory cascades that are activated after acid aspiration. The role of neutrophils was defined by immunodepletion before aspiration, which reduced injury by 59%. The injury was not modified in either P- or E-selectin-knockout mice, indicating that these adhesion molecules were not operative. C activation after aspiration was documented with immunochemistry by C3 deposition on injured alveolar pneumocytes. Animals in which C activation was inhibited with soluble C receptor type 1 (sCR1) had a 54% reduction in injury, similar to the level of protection seen in C3-knockout mice (58%). However C4-knockout mice were not protected from injury, indicating that C activation is mediated by the alternative pathway. Finally, an additive effect of neutrophils and C was demonstrated whereby neutropenic animals that were treated with sCR1 showed an 85% reduction in injury. Thus acid aspiration injury is mediated by neutrophils and the alternative C pathway.

P- and E-selectin-deficient mice are susceptible to cerebral ischemia-reperfusion injury.

We examined brain sections from P- and E-selectin-deficient mice (-/-) and their nontransgenic littermates (+/+) after focal cerebral ischemia and reperfusion (I/R) tissue injury. There was no difference in the subsequent infarct volume after 3 h of ischemia and 21 h of reperfusion. These data indicate that selectin-independent mechanisms mediate tissue injury after a prolonged period of transient focal ischemia.

Generation and characterization of a novel adhesion function blocking monoclonal antibody recognizing both rat and mouse E-selectin.

The prerequisite for the recruitment of circulating leukocytes to sites of inflammation is adhesion to vascular endothelial cells. Selectins play a significant role in the initiation of this multistep process by mediating "rolling" of the leukocytes on the endothelium. Investigation of selectin-dependent cell interactions using function blocking monoclonal antibodies (MAb) provides insights into the mechanisms involved in leukocyte migration into inflammation. Until now most studies in inflammation models in rats have relied on cross-reactive or polyclonal antibodies against rat E-selectin. In an E-selectin knockout mouse, we aimed to generate an adhesion function blocking MAb to rat E-selectin by immunization with rat E-selectin transfected Chinese hamster ovary cells (RESEC). An IgG1 kappa MAb was identified that reacts with RESEC but not with untransfected Chinese hamster ovary cells, as well as with recombinant mouse E-selectin protein as assessed by ELISA. This MAb is designated RME-1. It does not cross-react with rat L-selectin or rat P-selectin or E-selectin expressed on human umbilical vein endothelium. Adhesion of the HL-60 myeloid cells to immobilized mouse E-selectin was completely inhibited by MAb RME-1 under static conditions and adhesion of rat polymorphonuclear leukocytes to recombinant mouse E-selectin was blocked under rotation condition. This novel antibody thus recognizes a function-related epitope on rodent E-selectin.

Alternative CD44 splicing in intestinal stem cells and tumorigenesis.

As an important player in stem cells and cancer, CD44 is expressed in multiple isoforms via alternative mRNA splicing. Whether, and if so, how various isoforms play distinct roles in normal stem cells and tumorigenesis remains unclear. In this issue of Oncogene, Zeilstra et al. report studies showing that intestinal stem cells express a specific CD44 variant that promotes intestinal tumorigenesis induced by the activation of Wnt signaling, whereas the more commonly expressed standard CD44 isoform is not expressed by stem cells and does not promote tumor formation. This finding demonstrates an isoform-specific function of CD44 in intestinal stem cells and tumorigenesis.

Sulfated glycans induce rapid hematopoietic progenitor cell mobilization: evidence for selectin-dependent and independent mechanisms.

The adhesive mechanisms leading to the mobilization of hematopoietic progenitor cells (HPCs) from the bone marrow into the blood are poorly understood. We report on a role for selectins and fucoidan in progenitor mobilization. Baseline levels of circulating HPCs are increased in endothelial selectin-deficient (P/E-/-) mice. Similar levels are observed when E-selectin null (E-/-) mice are treated with anti-P-selectin antibody or with fucoidan (which inhibits P- and L-selectin function). In particular, administration of 2 doses of fucoidan (25 mg/kg) over 6 hours produces profound mobilization of progenitors in wild-type mice and the response is greatly enhanced in E-/- and P/E-/- mice. Competitive reconstitution experiments reveal that fucoidan also elicits long-term (more than 6 months) repopulating stem cells. Mobilization assays using chimeric mice harboring L-selectin-deficient progenitors and wild-type progenitors expressing the green fluorescence protein suggest that L-selectin expression is not required but confers an advantage for fucoidan-induced mobilization. Sulfation is critical as desulfated fucoidan is ineffective. In addition, sulphogalactosylceramide (sulfatide) but not heparin can induce HPC mobilization. Our results indicate that administration of sulfated glycans, especially with concurrent inhibition of E-selectin function, represents a powerful novel method for rapid mobilization of long-term-repopulating stem cells. These findings may help elucidate the mechanisms of HPC trafficking during development and adult life.

Platelets roll on stimulated endothelium in vivo: an interaction mediated by endothelial P-selectin.

P-selectin, found in storage granules of platelets and endothelial cells, can be rapidly expressed upon stimulation. Mice lacking this membrane receptor exhibit a severe impairment of leukocyte rolling. We observed that, in addition to leukocytes, platelets were rolling in mesenteric venules of wild-type mice. To investigate the role of P-selectin in this process, resting or activated platelets from wild-type or P-selectin-deficient mice were fluorescently labeled and transfused into recipients of either genotype. Platelet-endothelial interactions were monitored by intravital microscopy. We observed rolling of either wild-type or P-selectin-deficient resting platelets on wild-type endothelium. Endothelial stimulation with the calcium ionophore A23187 increased the number of platelets rolling 4-fold. Activated P-selectin-deficient platelets behaved similarly, whereas activated wild-type platelets bound to leukocytes and were seen rolling together. Platelets of either genotype, resting or activated, interacted minimally with mutant endothelium even after A23187 treatment. The velocity of platelet rolling was 6- to 9-fold greater than that of leukocytes. Our results demonstrate that (i) platelets roll on endothelium in vivo, (ii) this interaction requires endothelial but not platelet P-selectin, and (iii) platelet rolling appears to be independent of platelet activation, indicating constitutive expression of a P-selectin ligand(s) on platelets. We have therefore observed an interesting parallel between platelets and leukocytes in that both of these blood cell types roll on stimulated vessel wall and that this process is dependent on the expression of endothelial P-selectin.

Susceptibility to infection and altered hematopoiesis in mice deficient in both P- and E-selectins.

We describe the phenotype of mice lacking both endothelial selectins after sequential ablation of the genes encoding P- and E-selectins. In contrast with the rather mild phenotypes observed in mice deficient in a single selectin gene, the doubly deficient mice present extreme leukocytosis, elevated cytokine levels, and alterations in hematopoiesis. Granulocytopoiesis is increased both in bone marrow and spleen, while erythropoiesis is partially translocated to the spleen. Virtual lack of leukocyte rolling and low extravasation at sites of inflammation make these animals susceptible to opportunistic bacterial infections, to which they succumb. Our results show that the absence of endothelial selectins severely affects leukocyte homeostasis and indicate that these two selectins are as important for normal leukocyte function as are the leukocyte beta2 integrins.

DNA and protein components of nuclear acceptor sites for androgen receptors in the rat prostate.

Androgen receptor-acceptor complexes in nuclei from rat ventral prostates were cross-linked in situ with formaldehyde and partially purified using affinity chromatography. To isolate acceptor DNA, the cross-linked receptor-acceptor complexes in formaldehyde-treated chromatin samples were adsorbed to dihydrotestosterone-17 beta-succinyl agarose, eluted with 75 microM dihydrotestosterone-1% SDS, digested with proteinase K and extracted with phenol-chloroform. After 32P end-labelling and PAGE, this DNA contained two distinct bands of DNA (about 300 and 400 base pairs respectively) which were unique relative to the total prostatic DNA. As an alternative approach for characterizing acceptor DNA, the DNA in prostatic nuclei and cross-linked chromatin was labelled with 32P by nick translation and analysed in glycerol density gradients for associations with cross-linked androgen receptors. A symmetrical 7s peak of 32P-DNA with a small amount of coincident receptor was observed in the gradients after mild trypsin treatment. In the absence of trypsin treatment, both the cross-linked receptors and the labelled DNA sedimented to the bottom of the gradients. Isolation of acceptor proteins involved iodination of cross-linked chromatin with 125I and androgen affinity chromatography. A comparison of the relative efficiency of retention and elution of 125I-proteins from different affinity columns revealed that testosterone-17 beta-succinyl agarose was potentially most suitable for purification of acceptor proteins. After electrophoresis on SDS-polyacrylamide gels, the eluates from this type of affinity matrix were found to contain two major peaks of 125I-labelled proteins--one corresponding to a protein with a similar molecular weight as the nuclear androgen receptor (33,000 Da); the other having a molecular weight of 20,000 Da. While the precise identity of this latter entity is unknown, its enrichment and retention by the affinity gel implies that it is closely associated with the androgen receptor and may be a component of the acceptor sites.

The diagnostic value of CA 27-29, CA 15-3, mucin-like carcinoma antigen, carcinoembryonic antigen and CA 19-9 in breast and gastrointestinal malignancies.

In the past decade, considerable interest has arisen for defining the role of various tumor markers in the diagnosis of cancer. This cross-sectional study evaluates four breast cancer markers (CA 27-29, CA 15-3, MCA and CEA) and two gastrointestinal (GI) markers (CA 19-9 and CEA) in 213 patients. Receiver operating curves (ROC) revealed a sensitivity for the 90% specificity cutoff for breast cancers compared to breast benign diseases of 70% for CA 27-29, 67.5% for CA 15-3, 52.5% for MCA and 40% for CEA. When GI tumors were compared to benign GI disease, the sensitivity for 90% specificity was 40.3% for CEA and 32.3% for CA 19-9. Comparison of breast cancer and GI malignancies with other malignancies leads to a marked shift of the ROC curve to the right and loss of specificity. Late stage for all breast and GI tumor markers was found to be a predictor of high serum antigen level (p < 0.001). The presence of liver metastases in breast cancer was associated with abnormal levels of CA 27-29 (p = 0.028). Pancreas adenocarcinomas had a higher CA 19-9 antigen level (p < 0.001) than other GI malignancies. CA 27-29 appears to be at least as sensitive and specific as CA 15-3 in patients with breast cancer. None of the above markers retain their specificity when compared with a control group consisting of other malignancies.

Platelet-endothelial interactions in inflamed mesenteric venules.

The selectins are membrane glycoproteins promoting adhesive events between leukocytes, platelets, and endothelial cells. We have previously demonstrated that platelets roll on P-selectin expressed on stimulated endothelium. In this study, we wished to examine the function of both the platelet and endothelial selectins, P- and E-selectins, in mediating platelet-endothelial interactions during inflammation. We demonstrate, using intravital microscopic examination of venules inflamed with tumor necrosis factor-alpha (TNF-alpha), that resting platelets interact with both P- and E-selectins and that the leukocyte alpha(1,3)fucosyltransferases FucT IV and FucT VII do not provide platelets with selectin ligand activity. We also show that after thrombin activation of wild-type (+/+) platelets, platelet P-selectin can mediate interactions on a TNF-alpha-inducible endothelial ligand. To evaluate the potential role of platelet P-selectin in the recruitment of leukocytes to inflammatory sites, we reconstituted the bone marrow of mice deficient in both P- and E-selectins (P/E-/-) with wild-type (+/+) or P-selectin-deficient (P-/-) bone marrow containing megakaryocytic precursors. Providing +/+ platelets to P/E-/- mice by bone marrow transplantation did not rescue the immunodeficient phenotype, suggesting that platelet P-selectin does not have an active function in the recruitment of leukocytes into inflammatory sites. To participate in inflammatory or hemostatic responses, platelets may use the endothelial selectins.

P- and E-selectins are required for the leukocyte recruitment, but not the tissue swelling, associated with IgE- and mast cell-dependent inflammation in mouse skin.

Many studies, in both experimental animal and human systems, have indicated that P- and/or E-selectins may contribute importantly to the leukocyte recruitment that occurs in association with mast cell-dependent inflammatory responses. We used mice that genetically lack P-selectin (P -/-), E-selectin (E -/-), or both selectins (P/E -/-) to investigate the possible roles of these selectins in the IgE- and mast cell-dependent recruitment of neutrophils to the skin of mice. We found that a lack of either or both selectins had little or no effect on the extent of mast cell degranulation or the tissue swelling associated with these reactions. Moreover, a lack of either P- or E-selectin alone did not reduce the neutrophil infiltration at the reaction sites. However, mice lacking both P- and E-selectins exhibited an almost complete ablation of IgE- and mast cell-dependent neutrophil recruitment. These findings show that P- and E-selectins can express overlapping functions in leukocyte recruitment associated with IgE- and mast cell-dependent cutaneous late-phase reactions in mouse skin, and that a lack of both selectins results in a virtual elimination of IgE-dependent leukocyte recruitment.

Granulocyte-macrophage colony stimulating factor (GM-CSF) priming in the treatment of elderly patients with acute myelogenous leukemia.

Standard intensive induction therapy is tolerated poorly by elderly patients with acute myeloblastic leukemia (AML). We treated 19 elderly patients with AML, including seven with a prior myelodysplastic syndrome (MDS) with a combination of low dose cytarabine, hydroxyurea, and GM-CSF. The percentage of blasts in S-phase was evaluated prior to and 24 hr after starting the GM-CSF infusion. Cell cycle analysis was performed by flow cytometry using propidium iodine staining with fluorescein isothiocyanate-conjugated monoclonal antibody to the myeloid antigen CD 33. Seven out of nineteen (37%) achieved a complete remission (CR) and six (31%) a partial remission (PR) for an overall response rate of 68% (13/19). There were three early deaths from infectious complications or organ failure. One patient died from disseminated fungal infection after attaining a PR. The medial overall survival was 9.5 months with a range of 1 to 23+ months. The projected median survival for the patients with de novo AML is greater than 23 months. The percentage of CD 33+ cells in S-phase increased from a mean of 11.6+/-2.7 (SEM) pre GM-CSF to 19.0+/-3.7 (SEM) post GM-CSF (P < 0.001). Patients with prior MDS demonstrated a greater increment (post-pre) in S-phase activity after GM-CSF administration (P = 0.02). There was a correlation between the increase in percent of CD 33+ cells in S-phase and the degree of cytoreduction as determined by the day 14 bone marrow biopsy (r = .78). The toxicity of the regimen was limited to the hematopoietic system. Sixteen out of nineteen patients (84%) and 12/13 (92%) of the responding patients had bone marrow aplasia on day 14. No patients experienced > grade 2 gastrointestinal toxicity. There was no neurologic or cardiac toxicity. These data suggest that the combination of hydroxyurea, GM-CSF, and cytarabine is an effective remission-induction regimen in elderly patients with AML.