Non-immune chronic idiopathic neutropenia of adult: an overview

Copyright # Munksgaard 2001 Eur J Haematol 2001: 67: 35–44 Printed in UK. All rights reserved EUROPEAN JOURNAL OF HAEMATOLOGY ISSN 0902-44441 Non-immune chronic idiopathic neutropenia of adult: an overview Papadaki HA, Palmblad J, Eliopoulos GD. Non-immune chronic idiopathic neutropenia of adult: an overview. Eur J Haematol 2001: 67: 35–44. # Munksgaard 2001. Helen A. Papadaki1, Jan Palmblad2, George D. Eliopoulos1 1 Abstract: There is strong evidence that non-immune chronic idiopathic neutropenia of adult is a cytokine-mediated syndrome characterized by (a) neutropenia of varying degree associated with a low number of lineage-specific CD34+ cells and increased production of inhibitors of hematopoiesis, including transforming growth factor-b1 and tumor necrosis factor-a; (b) lymphopenia due to selective loss of primed/ memory T-cells and NK cells; (c) increased splenic volume on ultrasonography in 48.1% of patients; (d) osteopenia and/or osteoporosis in 60.0% of patients; (e) anemia, mostly of the type of anemia of chronic disease, in 15.6% of patients; (f) features of chronic antigenic stimulation, including increased proportion of bone marrow plasma cells, increased serum levels of IgG1 and/or IgA, increased frequency of monoclonal gammopathy of undetermined significance, increased frequency of antinuclear antibodies with specific reactivity, and increased serum levels of circulating immune complexes; and (g) increased concentrations of a variety of macrophage-derived proinflammatory cytokines and chemokines capable of affecting bone metabolism, bone marrow function, and leukocyte trafficking. All these findings are suggestive of the existence of an unrecognized lowgrade chronic inflammatory process which may be involved in the pathogenesis of the disorder. Neutropenia in these patients is probably the result of a combination of at least three factors, reduced neutrophil production in bone marrow, enhanced neutrophil extravasation, and increased sequestration and/or extravasation of neutrophils into the spleen. Chronic idiopathic neutropenia (CIN) is a widely used descriptive term denoting the ‘‘unexplained’’ reduction in the number of circulating neutrophils below the lower limit of the normal distribution in a given ethnic population for prolonged period of time, namely for more than three months (1–3). At least two types of the condition can be recognized, immune-mediated CIN due to antineutrophil antibodies (1–10), neutrophil-associated immunoglobulin (11), circulating immune complexes (12), antibodies to myeloid precursors (13), or cell-mediated suppression of myelopoiesis (14), and non-immune CIN due to unknown cause and pathogenetic mechanism. However, during the last Department of Haematology of the University of Crete School of Medicine, University Hospital of Heraklion, 2 Crete, Greece and Department of Medicine, The Karolinska Institutet at Huddinge University Hospital, Stockholm, Sweden Key words: non-immune chronic idiopathic neutropenia of adult; pro-inflammatory cytokines and chemokines; leukocyte trafficking; low-grade chronic inflammation; chronic antigenic stimulation; osteopenia; osteoporosis Correspondence: Helen A. Papadaki, Assistant Professor of Haematology, University Hospital of Heraklion, PO Box 1352, Crete, Greece Tel: +30 81 392.805 Fax: +30 81 318.028 e-mail: [email protected] Accepted for publication 18 February 2001 few years, evidence is accumulating that nonimmune CIN of adult (NI-CINA) is probably mediated by pro-inflammatory cytokines and chemokines (15–17), and may constitute a separate syndrome presented with a variety of clinical manifestations and laboratory findings suggestive of the existence of an underlying low-grade chronic inflammatory process possibly involved in the pathogenesis of the disorder (15–22). In this paper we summarize recently acquired knowledge on NI-CINA in an attempt to outline its possible pathogenetic mechanism. The basis is an extensive evaluation of a Cretan population. Other NI-CINA-like patient cohorts are also presented; 35 Papadaki et al. however, due to differences in the investigational approach, it is not possible to make full comparisons at this time. Diagnostic criteria and grading Diagnostic criteria of NI-CINA are listed in Table 1 (15, 20). The condition can be identified among other types of chronic neutropenia by its acquired character, usually benign outcome, lack of chromosomal abnormalities, negative tests for antineutrophil antibodies, lack of history of exposure to ionizing irradiation, contact with chemical compounds or intake of drugs capable of causing neutropenia, and lack of clinical and/or laboratory evidence for any underlying disease to which neutropenia could be ascribed (15, 18, 20). Despite the fact that many of these criteria are exclusion criteria, this definition seems to describe a rather homogeneous subset of patients with ‘‘unexplained’’ chronic neutropenia. Other NI-CINA-like entities with more pronounced susceptibility to infections have also been described (23). On the basis of the number of circulating neutrophils in the Cretan patient population, neutropenia has been characterized as ‘‘mild’’ if the number of neutrophils ranged from 1800 to 2499 per ml of blood (18, 24), ‘‘moderate’’ if it ranged from 500 to 1799 per ml, and ‘‘severe’’ if it was below 500 per ml of blood. The term ‘‘agranulocytosis’’ was restricted to cases of severe neutropenia with neutrophil counts below 100 per ml of blood (24), while that of ‘‘pronounced neutropenia’’ was often applied to denote all cases of moderate and severe neutropenia taken together (15, 18, 20). Admittedly, these limits for ‘‘mild’’ and ‘‘moderate’’ neutropenia are arbitrarily chosen and they do not discriminate for proneness to infection or biological characteristics. Others have arbitrarily advocated the limits of 500–1000 neutrophils per ml for ‘‘moderate’’, and 1000–1500 neutrophils per ml of blood for ‘‘mild’’ neutropenia (23, 25). anti-inflammatory drugs did not account for the development of NI-CINA (26). An increased frequency of HLA-DRB1*1302 haplotype, defining a relative risk of 8.36 for the development of the disorder, has recently been reported from our laboratory (28). Clinical manifestations NI-CINA patients are usually asymptomatic, and diagnosis is made on the occasion of a blood cell count performed for unrelated reasons. However, some of these patients may complain of clinical symptoms and laboratory abnormalities requiring further investigation and treatment. Infections Recurrent respiratory, urinary or cutaneous infections, defined as fever above 38 uC for at least 3 d requiring treatment with antibiotics, have been observed in a limited number of patients with chronic idiopathic neutropenia (2, 29). Persistent gingivitis has also been reported (30–32). In a group of 183 Cretan NI-CINA patients with neutrophil counts ranging from 393 to 2499 per ml of blood, the incidence of infection has been estimated as 2.09% per year (18). However, in a Swedish group of neutropenic patients with neutrophil counts ranging from 500 to 1499 per ml of blood, severe and recurrent infections were noted in 40% of patients (23). This difference is probably due to the different criteria applied for NI-CINA in the Swedish study, the lower cut-off value in the definition of neutropenia, the relatively small number of patients studied, and, perhaps, the different definition of infection. The risk of contracting bacterial infection in NI-CINA patients depends on the severity of neutropenia, and also on the low plasma levels of soluble Fcc-RIII (soluble CD16) (33). Generally, NI-CINA patients respond to granulocyte colonystimulating factor (G-CSF) administration by increasing rapidly the number of circulating neutrophils, even at low doses of the cytokine (34, 35). Epidemiology The incidence of NI-CINA in the general population is unknown, but the prevalence of pronounced neutropenia in the Cretan population was estimated as 1.67% (26). The condition affects women preferentially (27). In a series of 183 patients registered in our Neutropenia Unit, the female to male ratio was 5:1 (18). In the same series of patients, about 65% were 30–59 yr, 20% 15–30 yr and 15% 60–79 yr old at diagnosis (18). Occupation, use of agricultural drugs, contact with organic solvents or intake of non-steroidal 36 Splenomegaly Clinically detectable splenomegaly is very rare in NI-CINA patients (2). Only 1 of our 183 Cretan patients had a mild palpable splenomegaly (18). However, increased splenic volume, as estimated by determining the ‘‘corrected splenic index’’ (CSI) on ultrasonography, was found in 48.1% of patients (36). In the entire group of patients, individual CSI values inversely correlated with the number of circulating neutrophils and lymphocytes but not with the levels of hemoglobin or the number of Non-immune chronic idiopathic neutropenia of adult Table 1. Criteria applied for NI-CINA diagnosis (15, 18, 20) 1. Number of circulating neutrophils below 2500 per ml of blood found at least in four consecutive blood cell counts performed monthly within the last three months of observation. 2. Exclusion of cyclic and familial neutropenia by appropriate studies. 3. No evidence of any underlying disease known to be associated with neutropenia following a detailed hematological, biochemical, virologic, cytogenetic, immunologic, parasitologic, and ultrasonic investigation. 4. No previous exposure to ionizing irradiation, contact with organic solvents or intake of cytotoxic or other drugs capable of causing neutropenia. 5. Negative leukoagglutination and immunofluorescent tests for anti-neutrophil antibodies. circulating monocytes or platelets (36). It has been suggested that the increased splenic volume in these patients may reflect hyperplasia of activated splenic macrophages and activated endothelial cells, given that the levels of macrophage-derived pro-inflammatory cytokines and endothelial cell-derived soluble cell adhesion molecules were both significantly increased in patients’ sera (36). On the basis of these observations, it has been postulated that the increased splenic volume in NI-CINA patients may affect leukocyte trafficking by increasing the sequestration and/or extravasation of neutrophils and lymphocytes into the spleen (36). Osteopenia and osteoporosis About 38% of Cretan NI-CINA patients presented myoskeletal pains due to osteoarthritis or osteoporosis, compared to only 6% of age- and sexmatched normal controls (18). In a study of 45 unselected Cretan NI-CINA patients never treated with G-CSF, bone mineral density (BMD) was significantly reduced, and individual BMD values strongly correlated with the degree of neutropenia (37). On the basis of previously reported criteria (38), osteopenia was diagnosed in 44.4% and osteoporosis in 15.6% of patients (37). These patients also had increased urine concentrations of the amino-telopeptide of collagen I (NTx), suggestive of increased bone resorption, and increased serum osteocalcin (OC), suggestive of increased bone formation. The values of both NTx and OC were correlated inversely with BMD and positively with the levels of serum tumor necrosis factor-a (TNF-a) (37). These alterations in bone metabolism biochemical markers are suggestive of an increased bone turnover leading in the time to bone loss. It has been suggested that the increased bone resorption is probably related with the increased production of pro-inflammatory cytokines, mainly TNF-a, found in these patients (37). Natural history As mentioned above, neutropenia can persist for years or throughout the life in NI-CINA patients. Some fluctuations in neutrophil counts are not uncommon, but spontaneous correction of neutro- penia is rarely seen (2, 23, 27). No treatment of neutropenia per se is recommended. Patients with severe neutropenia and infection must be treated with antibiotics and G-CSF. It has been reported that no case of lupus erythematosus, rheumatoid arthritis, leukemia or other systemic disease developed during long periods of patient observation (27). However, Auner and co-workers reported recently on two cases of chronic idiopathic neutropenia preceding acute myeloid leukemia, and they postulated that the disorder may be a preleukemic state (39). One of the Cretan patients developed acute myeloid leukemia and a second acute multilineage leukemia 34 and 64 months, respectively, after NI-CINA diagnosis, but a causal relationship between leukemia and the underlying NI-CINA could not be proved (40). Laboratory findings Peripheral blood cells NI-CINA may present either as isolated neutropenia or as neutropenia combined with anemia and or thrombocytopenia (2), but lymphopenia and monocytopenia may also occur (18, 19, 41–43). Neutropenia. In previous studies, probably including cases of both immune-mediated CINA and NI-CINA, neutropenia has been attributed to increased loss of neutrophils in the periphery because of an excessive margination (44), increased neutrophil destruction (4–7, 9) or enhanced neutrophil extravasation (15, 45), or to impaired neutrophil production in bone marrow because of a low number of myeloid progenitors (16, 46) or a dysfunction of hematopoietic microenvironment in supporting efficient myelopoiesis (47, 48) due to inadequate release of hematopoietic growth factors, namely G-CSF, by the stromal cells (49, 50), or to increased production of inhibitors of hematopoiesis, including transforming growth factor b1 (TGF-b1) and TNF-a (16). However, the mechanism of neutropenia in NICINA appears to be more complex considering recent reports (see below). 37 Papadaki et al. Cretan NI-CINA patients responded to intravenous hydrocortisone by increasing the number of circulating neutrophils by 300% or more (unpublished data). Similar results have been reported by others on various forms of chronic neutropenia using etiocholanolone, endotoxin, prednisone or epinephrin to mobilize neutrophils from the marginal pool or marrow reserve to pass into the circulating granulocyte pool (51, 52). Functional studies on NI-CINA neutrophils are very limited. Myeloperoxidase and alkaline phosphatase expression by patient neutrophils were found within normal range in a number of patients studied (unpublished data). Abnormalities in b2integrins (45) and an increased density of neutrophil surface L-selectin expression have been reported (53). No mutations in elastase encoding ELA2 genes could be demonstrated in a number of Cretan NICINA patients studied (Horwitz P, personal communication). Lymphopenia. Peripheral blood lymphocytes were significantly decreased in the Cretan patients with NI-CINA, and individual lymphocyte counts correlated with the degree of neutropenia (19, 43). About 37% of patients had lymphocyte counts below 1500 per ml of blood. Lymphocyte depletion was mainly due to the reduction of T-cells and, to a lesser degree, to a decline in the number of natural killer cells (NK cells) (43). Both helper/ inducer CD4+ and suppressor/cytolytic CD8+ Tcells decreased in parallel, so that the CD4+/ CD8+ T-cell ratio remained within normal limits. T-Cell depletion reflected mainly the reduction in the number of primed/memory CD45RO+/CD4+ and CD45RO+/CD8+ but not of virgin/naive CD45RA+/ CD4+ or CD45RA+/CD8+ T-cells (43). The absolute numbers of T-cells carrying the CD25, CD38, CD71 or HLA-DR cell activationrelated surface markers were within normal limits, suggesting that activated T-lymphocytes are not increased in the circulation of NI-CINA patients (43). Natural killer (NK) cells, defined by the expression of CD16 and CD56 cell surface markers, were reduced in the Cretan NI-CINA patients (43). NK cell activity (NKa) of peripheral blood mononuclear cells (PBMCs) was also significantly reduced, and individual NKa values were inversely correlated with the degree of neutropenia (41). Patient NKa increased in the presence of recombinant human interleukin-2 or recombinant human interferon-alpha, but it did not reach the respective values of normal subjects (41). The capacity of CD56+ cells to bind to K562 cell targets, and the expression of perforin and granzyme B by patient 38 CD16+ cells, were normal (41). On the basis of these data, it has been suggested that the low NKa in NI-CINA is probably due to low numbers of NK cells present in patient PBMC suspensions rather than to an intrinsic defect of NK cells (41, 42). Peripheral blood B-cells, defined by the expression of CD19 and CD20 cell surface markers, as well as IgD-positive cells, did not show any significant change in the NI-CINA patients studied (43). Monocytopenia and thrombocytopenia. About 36% of Cretan NI-CINA patients had peripheral blood monocyte counts below 250/ml of blood, and individual absolute monocyte numbers correlated with the degree of neutropenia (18). Thrombocytopenia, defined as the reduction of the number of platelets below 150000/ml of blood, was observed in about 10% of patients (18). None of our 66 normal controls had thrombocytopenia, while only four had monocytopenia (6%). The cause and the underlying mechanisms leading to these disturbances still remain unclear. Anemia. Anemia has been reported in patients with chronic idiopathic neutropenia of adult by several investigators, and it has been attributed to iron deficiency (27), disturbances in iron turnover (2), follate and/or cobalamin deficiency (54), or to shortened erythrocyte survival (2, 54). In a recent study anemia, defined as a reduction of hemoglobin levels below 12 g/dl in women and 13.3 g/dl in men (55, 56), was found in 15.6% of Cretan NICINA patients (57). Interestingly, individual hemoglobin values correlated positively with the number of circulating neutrophils and inversely with the levels of serum erythropoietin (EPO) and TNF-a concentrations (57). Based on previously reported criteria (58, 59), it has been suggested that anemia of chronic disease is the more frequent type of anemia seen NI-CINA patients (57). Bone marrow cells Morphologically defined cell compartments. Bone marrow granulocytic series have previously been characterized as hyperplastic, normal or hypoplastic in patients with chronic idiopathic neutropenia (60). These contradictory conclusions were probably due to the great heterogeneity of neutropenic patients studied, including cases with immunemediated CIN and, perhaps, cases with myelodysplasia. In the group of Cretan NI-CINA patients (22), the bone marrow granulocytic series was significantly reduced in 79% of patients with pronounced neutropenia. The myeloid to erythroid cell ratio was decreased and correlated with the Non-immune chronic idiopathic neutropenia of adult degree of neutropenia. The proportion of cells reflecting the proliferating granulocyte pool (PGP) was increased, while that of cells reflecting the maturating granulocyte pool (MGP) was significantly reduced (22). Thus, a striking shift to the left of the granulocytic series was apparent. The low MGP has been attribulted to enhanced transfer of cells through the compartment of maturating pool as a response to neutropenia, while the elevated PGP may represent an effort of the bone marrow to compensate the low MGP (22). The inability of marrow to fully compensate MGP is probably due to an insufficient supply of new myeloblasts from the reduced compartment of myeloid progenitor cells (16). Leukokinetic studies previously performed with DF32P in patients with chronic idiopathic neutropenia (61) are consistent with this suggestion. Megaloblastoid features of erythroid series were noted in about 25% of patients (22). Megacaryocytes were within normal limits, but some micromegacaryocytes were occasionally seen (22). The proportions of lymphocytes and eosinophils were within normal limits or slightly increased (22). In contrast, the proportion of plasma cells was significantly elevated and individual plasma cell percentages were inversely correlated with the number of circulating neutrophils (22). Lymphocyte subpopulations. The proportion of bone marrow T cells and NK cells did not show any significant change in the Cretan NI-CINA patients, while B cells and CD57+ cells were slightly increased (22). Bone marrow primed/ memory CD45RO+ T cells were reduced in patients with pronounced neutropenia. No significant alterations were found in the proportion of virgin/naive CD45RA+ T cells or the proportion of T cells bearing cell surface activation markers. A significant rise in CD57+ cells in bone marrow sections has been noted in unselected neutropenic patients by Picker et al. (62). Histological findings. In histology, bone marrow cellularity has been characterized as reduced in patients with pronounced neutropenia (22). A shift to the left of the granulocytic series was noted in about 67% of patients (22). An increased proportion of disperse lymphocytes was found in 10.5% of patients, but aggregates of lymphocytes (nodules) in paratrabecular areas without germinal centers were found in 28.9% of patients (22). An increased proportion of plasma cells (2–8%) was noted in 50% of patients. Eosinophils were increased in about 10%, and histiocytes in 12% of patients. Immunohistochemical studies did not reveal any important abnormality. In no case could p53 protein be detected in any cell lineage, while bcl-2 protein was detected in all lymphocytic aggregates (22). Granulocyte progenitor cells and hematopoietic microenvironment. It has been reported that patients with chronic idiopathic neutropenia have normal (47, 48), reduced (46), or increased (47) numbers of marrow granulocyte-macrophage colony-forming units (CFU-GM). It has also been suggested that a dysfunction of hematopoietic microenvironment (47, 48), leading to an inadequate release of G-CSF because of an abnormal expression of G-CSF mRNA (49) or a defect at the post-transcriptional level (50), or to increased release of inhibitors of hematopoiesis such as TGF-b1 and TNF-a (16), may be involved in the pathogenesis of neutropenia in affected subjects. Recent studies showed that NI-CINA patients had a low frequency of lineage-specific CD34+cells (CD34+/lin+) expressing CD33 or CD71 or HLA-DR or CD45RA cell surface markers (16). Interestingly, the number of marrow CFU-GM correlated with the degree of neutropenia in these patients (16). Moreover, in bone marrow longterm cultures, patient stromal cells failed to support CFU-GM colony formation by normal myeloid progenitors, compared to normal stromal cells (16). Supernatants from patient bone marrow long-term cultures had increased concentrations of G-CSF, IL-6, TNF-a and TGF-b1, and normal or slightly increased levels of GM-CSF (16). These findings suggest that stromal cells, e.g. macrophages or mesenchymal stem cells, are probably activated in NI-CINA patients and they produce increased amounts of both hematopoietic growth factors and inhibitors of hematopoiesis. Preliminary data from our laboratory are suggestive of increased apoptosis within the CD33+ cell fraction. It is thus possible that activated marrow stromal cells may suppress myelopoiesis not only by releasing increased amounts of inhibitors of hematopoiesis affecting progenitor cell proliferation and overcoming the inductive action of growth factors, but also by producing increased amounts of TNF-a which activate the cell death pathway in the granulocyte progenitor cell compartment. More studies are certainly needed to clarify the true role of cytokines in the regulation of granulopoiesis in these patients. Serum proteins Serum immunoglobulins. About 18.5% of Cretan NI-CINA patients studied had serum gammaglobulins above the upper 95% confidence limit 39 Papadaki et al. of the control range, and about 20% of these patients had increased serum levels of IgG or IgA but not IgM (20, 63). Moreover, 7.14% of patients with neutrophil counts below 1500/ml of blood, and 2.61% of patients with neutrophil counts ranging from 1500 to 2499 per ml of blood, had monoclonal gammopathy of undetermined significance (MGUS) (63). None of our 66 normal controls or 157 age- and sex-matched patients hospitalized for various non-malignant diseases had any evidence of MGUS in the serum (63). The increased levels of IgG probably reflected the rise in IgG1, since IgG2 and IgG4 did not show any significant change and IgG3 levels were severely reduced (20). Low serum IgG3 in nonsevere neutropenia patients have also been noted in a Swedish study (23). Individual IgG3 levels correlated inversely, and individual IgA positively, with the levels of serum TGF-b1 (20). It has been suggested that changes in these immunoglobulins in NI-CINA patients may be related to increased amounts of TGF-b1, which is a cytokine normally affecting transcriptional factors involved in immunoglobulin class switching (64). Acute phase proteins. With the exception of a1antitrypsin (AAT), all other serum acute phase reactants did not present any significant change in a group of NI-CINA patients studied (63). The cause and the biological significance of the increased levels of AAT in these patients needs further investigation, but one can hypothesize that they may be related to the expected low serum levels of neutrophil elastase for which AAT is the natural inhibitor. Circulating immune complexes, rheumatoid factor, complement factors C3 and C4 and soluble FccRIII. It has been reported that circulating immune complexes (CICs) may be increased in some patients with chronic idiopathic neutropenia (12). Increased levels of CICs were found in the sera of 32.5% of the Cretan NI-CINA patients (21). No significant changes were observed in the levels of serum rheumatoid factor or complement factors C3 and C4 (21). In a group of 66 Dutch patients with NI-CINA and neutrophil counts below 1500 per ml of blood, plasma levels of soluble Fcc-RIII were found within the normal range, but patients with low Fcc-RIII values had an increased risk of contracting infection (33). Serum autoantibodies Antinuclear antibodies (ANA) at a titer equal to or higher than 1:80 were detected in 33.1% of Cretan NI-CINA patients compared to 9.5% found in a 40 group of age- and gender-matched healthy controls (21). The pattern of ANA reactivity was speckled in 84.6% and diffuse in 15.4%. Both types of granulation, coarse and fine, were observed with about equal frequency (21). Antibodies to a variety of extracted nuclear antigens and antibodies to organspecific and organ-non-specific antigens, including anti-neutrophil cytoplasmic antigens, were either negative or positive at low titer in a small proportion of patients similar to that seen in the normal controls (21). Low-titer serum anti-dsDNA antibodies were transiently observed only in one patient (21). Viruses and other pathogens Antibodies to hepatitis B and C virus, cytomegalovirus, Epstein–Barr virus, herpes viruses, human immunodeficiency viruses, Coxiella burnetii, Brucella melitensis, Salmonella typhi, Toxoplasma gondii and Leismania donovani were tested in a large group of Cretan NI-CINA patients and were found negative for active infection (IgM antibodies) or positive for old infection (IgG antibodies) at a low proportion of patients similar to that noted in the normal controls (18). Ehrlichia genomic material could not be detected in bone marrow stroma tested in a number of NI-CINA patients (unpublished data). Cytokines and chemokines Serum levels of a variety of pro-inflammatory cytokines and chemokines, including IL-1b, TNFa, soluble TNF receptor p55 (sTNF-RI), TGF-b1, IL-6, IL-8 and RANTES, were found significantly increased in Cretan NI-CINA patients, and individual cytokine and chemokine values strongly correlated with the degree of neutropenia (15, 17, 20, 42). Interestingly, all these molecules are mainly produced by activated macrophages. In contrast, serum levels of soluble CD23, soluble IL-2 receptor, IL-4 and interferon-gamma, which are mainly produced by activated lymphocytes, did not show any significant alteration in the patients studied (17). Increased concentrations of TGF-b1, IL-6, TNF-a and G-CSF have also been found in patient bone marrow long-term culture supernatants, indicating that marrow stromal cells are also activated in these patients (16). It is important to note that the increased serum TNF-a levels correlated with the values of patient CSI expressing an estimation of splenic volume (36). Taken together these findings are suggestive of the existence of an unrecognized low-grade chronic inflammatory process preferentially affecting macrophages in patients with NI-CINA (15, 17, Non-immune chronic idiopathic neutropenia of adult 20, 42). The alternative possibility, that the lowgrade chronic inflammation may be the result rather than the cause of neutropenia, seems unlikely here for many reasons and also from the fact that the levels of serum pro-inflammatory cytokines and chemokines were not increased in other patients with neutropenia of the same degree lasting for more than 3 months after the cessation of chemotherapy for breast cancer (18). Serum soluble cell adhesion molecules Serum levels of endothelial cell-derived soluble cell adhesion molecules, such as soluble E-selectin (sEselectin), soluble intercellular cell adhesion molecule (sICAM) and soluble vascular cell adhesion molecule (sVCAM), were all significantly increased in Cretan NI-CINA patients, and the observed individual values correlated inversely with the number of circulating neutrophils and positively with the levels of most of the aforementioned macrophage-derived pro-inflammatory cytokines and chemokines (20, 42). These observations indicate that NI-CINA patients have activated endothelium, and that endothelial cell activation is probably mediated by the increased levels of IL1b and TNF-a, two potent endothelial cell activators (62). On the basis of previously reported studies on the role of activated endothelium on leukocyte trafficking (65), it has been suggested that an enhanced neutrophil extravasation, probably mediated by interactions of b2-integrins expressed on neutrophils with ICAM-1 and VCAM-1 molecules expressed on activated endothelium, and under the influence of the chemokine IL-8, may occur in NI-CINA patients, and may be a contributing factor in the determination of the degree of neutropenia (15). An enhanced extravasation has also been suggested for primed/ memory CD45RO+ T cells and NK cells constitutively expressing a4b1 integrins (very late antigen-4, VLA-4) (65). Using this molecule, lymphocytes interact with endothelial ICAM-1 and VCAM-1, adhere firmly to the vessel wall and pass into the tissues under the influence of the chemokine RANTES (19). Possible pathogenetic mechanism Based on the above discussed data we have suggested that NI-CINA is probably related to an underlying low-grade chronic inflammation which may be involved in the pathogenesis of the disorder. The genetically defined predisposition of HLADRB1*1302 haplotype-bearing individuals to develop NI-CINA (28) coroborates this suggestion. Fig. 1. Schematic representation of a mechanism possibly operating in the pathogenesis of NI-CINA. An unrecognized low-grade chronic inflammation leads to both chronic antigenic stimulation and macrophage activation. Chronic antigenic stimulation is indicated by the increased proportion of marrow plasma cells, the increased serum IgG1 and/or IgA, the increased frequency of MGUS, the increased incidence of antinuclear antibodies (ANA) and the increased levels of circulating immune complexes (CICs). Activated macrophages produce pro-inflammatory cytokines and chemokines affecting bone marrow function, vascular endothelium and bone metabolism. The effect on bone marrow is indicated by the increased production of inhibitors of myelopoiesis overcoming the production of hematopoietic growth factors and leading to reduced neutrophil production. The effect on vascular endothelium is indicated by endothelial cell activation resulting to enhanced neutrophil extravasation. The effect on bone metabolism is indicated by the increased frequency of osteopenia and osteoporosis. Thus, neutropenia in NI-CINA patients may be the result of a combination of at least three factors, reduced neutrophil production, enhanced neutrophil extravasation, and increased sequestration and/or extravasation of neutrophils into the spleen. The existence of such an inflammation may lead to both chronic antigenic stimulation and macrophage activation. Antigenic stimulation is indicated by the increased serum IgG1 and/or IgA (20), the increased frequency of MGUS (63), the increased proportion of marrow plasma cells (22), the increased frequency of serum ANA (21), and the increased levels of CICs (21). Macrophage activation is indicated by the increased levels of macrophage-derived proinflammatory cytokines and chemokines (17, 20) and the increased concentrations of stromal cellderived cytokines in long-term bone marrow culture supernatants (16). Both chronic antigenic stimulation and macrophage activation may explain the increased splenic volume found on ultrasonography in the patients studied (36). The lack of increased proportion of activated lymphocytes in peripheral 41 Papadaki et al. blood (46) and bone marrow (22), and the lack of a rise in lymphocyte-derived cytokines in patient sera (17), do not exclude the possibility of the existence of such an underlying chronic inflammation. Macrophage-derived pro-inflammatory cytokines may affect vascular endothelium, bone metabolism and marrow function (Fig. 1). The effect on vascular endothelium, mediated mainly by the increased levels of IL-1b and TNF-a, is indicated by endothelial cell activation leading to enhanced cell extravasation (15). Such an effect may account, to some degree, for the development of neutropenia and lymphopenia in NI-CINA patients, and can explain the rarity of bacterial infections in these patients, given that the numbers of leukocytes present in the tissues are probably higher than those expected from the low numbers of leukocytes in the circulation. Pro-inflammatory cytokines from the periphery and locally produced by bone marrow stromal cells may affect myelopoiesis by inhibiting proliferation and enhancing apoptosis of myeloid progenitor cells. Inhibition of proliferation may be mediated by the increased levels of TGF-b1 and, perhaps, other inhibitory substances (16), while enhanced apoptosis may be mediated by the increased concentrations of TNF-a. Probably, the increased production of inhibitors of myelopoiesis may overcome the action of the increased amounts of locally produced hematopoietic growth factors and may lead to impaired neutrophil production (16). Anemia of chronic disease, seen in a number of NI-CINA patients, can also be explained from the increased concentrations of TNF-a in bone marrow microenvironment (57). Pro-inflammatory cytokines may also affect bone metabolism by activating osteoclasts (66). IL-1b and TNF-a are two potent osteoclast activators (67). It is possible that the osteopenia and osteoporosis seen in NI-CINA patients may be a result of the increased production of these cytokines (37). The increased splenic volume may also account for the low number of circulating neutrophils and lymphocytes in NI-CINA patients, given that the increased CSI values in these patients correlated with the degree of neutropenia and lymphopenia (36). It is also possible that the spleen may be a putative site of chronic inflammation, and therefore may be an organ in which neutrophil and lymphocyte sequestration and extravasation may occur (36). Conclusion There is strong evidence that NI-CINA is a cytokine-mediated syndrome characterized by a variety of clinical and laboratory findings, including 42 neutropenia of varying degrees. NI-CINA patients have increased levels of macrophage-derived proinflammatory cytokines and chemokines in both blood serum and long-term bone marrow culture supernatants, suggestive of the existence of an unrecognized low-grade chronic inflammatory process. These cytokines may affect leukocyte trafficking by activating endothelium, bone metabolism by activating osteoclast, and bone marrow function leading to defective neutrophil production. Thus, neutropenia in NI-CINA patients may be the result of a combination of at least three factors, decreased production of neutrophils in the bone marrow, enhance neutrophil extravasation into the tissues, and increased sequestration and/or extravasation of neutrophils into the spleen. References 1. KYLE RA, LINMAN JW. Chronic idiopathic neutropenia: a newly recognized syndrome? N Engl J Med 1968;279:1015– 1019. 2. DALE DC, GUERRY D, WEWERKA J, BULL J, CHUSID M. Chronic neutropenia. Medicine (Baltimore) 1979;58:128–144. 3. HADDY TB, RANA SR, CASTRO O. Benign ethnic neutropenia: What is a normal absolute neutrophil count? J Lab Clin Med 1999;133:15–22. 4. BOXER LA, GREENBERG MS, BOXER GJ, STOSSEL TP. Autoimmune neutropenia. N Engl J Med 1975;293:748–753. 5. LALEZARI P, JIANG A-F, YEGEN L, SANTORINEOU M. Chronic autoimmune neutropenia due to anti-NA2 antibody. N Engl J Med 1975;293:744–747. 6. SHASTRI KA, LOGUE GL. Autoimmune neutropenia. Blood 1993;81:1984–1995. 7. LOGUE GL, SHASTRI KA, LAUGHLIN M, SHIMM DS, ZIOLKOWSKI LM, IGLEHART JL. Idiopathic neutropenia: antineutrophil antibodies and clinical correlations. Am J Med 1991;90: 211–216. 8. HARTMAN KR, WRIGHT DG. Identification of autoantibodies specific for the neutrophil adhesion glycoproteins CD11b/ CD18 in patients with autoimmune neutropenia. Blood 1991;78:1096–1104. 9. BUX J, BEHRENS G, JAEGER G, WELTE K. Diagnosis and clinical course of autoimmune neutropenia in infancy: analysis of 240 cases. Blood 1998;91:181–186. 10. PAPADAKI HA, GIBSON FM, RIZZO S, GORDON-SMITH EC, MARSH JCW. Assessment of bone marrow stem cell reserve and function and stromal cell function in patients with autoimmune cytopenias. Blood 2000;96:3272–3275. 11. LIGHTSEY AL, CHAPMAN R, MCMILLAN R, MUSHOVIC J, YELENOSKY R, LONGMIRE RL. Immune neutropenia. Ann Intern Med 1977;86:60–62. 12. CALIGARIS-CAPPIO F, GAMUSSI G, GAVOSTO A. Idiopathic neutropenia with normocellular bone marrow: An immune complex disease. Br J Haematol 1979;43:595–605. 13. HARTMAN KR, LARUSSA VF, ROTHWELL SW, ATOLAGBE TO, WARD FT, KLIPPLE G. Antibodies to myeloid precursor cells in autoimmune neutropenia. Blood 1994;94:625–631. 14. BAGBY GC, LAWRENCE HJ, NEERHOUT RC. T-Lymphocytemediated granulopoietic failure: in vitro identification of prednisone-responsive patients. N Engl J Med 1983;309: 1073–1078. 15. PAPADAKI HA, ELIOPOULOS GD. Enhanced neutrophil extravasation may be a contributing factor in the determination of Non-immune chronic idiopathic neutropenia of adult 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. neutropenia in patients with chronic idiopathic neutropenia of adults. Eur J Haematol 1998;61:272–277. PAPADAKI HA, GIOUREMOU K, ELIOPOULOS GD. Low frequency of myeloid progenitor cells in chronic idiopathic neutropenia of adults may be related to increased release of TGF-b1 by bone marrow stromal cells. Eur J Haematol 1999;63:154–162. PAPADAKI HA, COULOCHERI SA, ELIOPOULOS GD. Patients with chronic idiopathic neutropenia of adults have increased serum concentrations of inflammatory cytokines and chemokines. Am J Hematol 2000;65:271–277. PAPADAKI HA, PSYLLAKI M, ELIOPOULOS GD. Clinical and laboratory findings on 183 patients with non-immune chronic idiopathic neutropenia of adult. Br J Haematol 2001;(submitted). PAPADAKI HA, ELIOPOULOS GD. Selective loss of peripheral blood CD45RO+ T-lymphocytes correlates with increased levels of serum cytokines and endothelial cell-derived soluble cell adhesion molecules in patients with chronic idiopathic neutropenia of adults. Ann Hematol 1998;77:153–159. PAPADAKI HA, PALMBLAD J, KAPSIMALI V, ANAGNOU NP, ELIOPOULOS GD. Increased serum IgA and decreased IgG3 strongly correlate with increased serum levels of TGF-b1 in patients with non-immune chronic idiopathic neutropenia of adults. Eur J Haematol 2000;65:237–244. PAPADAKI HA, PSYLLAKI M, ELIOPOULOS DG, TSIROYANNI A, ELIOPOULOS GD. Increased frequency and specific reactivity of serum antinuclear antibodies in patients with non-immune chronic idiopathic neutropenia of adults. Acta Haematol 2001;105:13–20. PAPADAKI HA, CHATZIVASSILI A, STEFANAKI K, KOUMAKI V, KANAVAROS P, ELIOPOULOS GD. Morphologically defined myeloid cell compartments, lymphocyte subpopulations and histological findings of bone marrow in patients with nonimmune chronic idiopathic neutropenia of adults. Ann Hematol 2000;79:563–570. PALMBLAD J, KARLSTRÖM R, GUSTAFSON R. Chronic mild neutropenia in adults – relation to IgG3 deficiency and infection susceptibility. Blood 2000;96:31b (Abstract). HOFFBRAND AV, PETTIT JE. Essential Haematology. 3rd edn. Oxford: Blackwell Scientific Publications, 1993;156. DALE DC. Immune and idiopathic neutropenia. Curr Opin Hematol 1998;5:33–36. PAPADAKI HA, XYLOURI I, COULOCHERI SA, KALMANTI M, KAFATOS A, ELIOPOULOS GD. Prevalence of chronic idiopathic neutropenia of adults among an apparently healthy population living on the island of Crete. Ann Hematol 1999;78:293–297. KYLE RA. Natural history of chronic idiopathic neutropenia. N Engl J Med 1980;302:908–909. PAPADAKI HA, ELIOPOULOS GD, COULOCHERI SA, SPYROPOULOU M, STAVROPOULOS-GIOKAS C. Increased frequency of DRB1*1302 HLA-haplotype in patients with nonimmune chronic idiopathic neutropenia of adults. Blood 2001;97: 850–851. SIEVERS EL, DALE DC. Non-malignant neutropenia. Blood Rev 1996;10:95–100. KYLE RA, LINMAN JW. Gingivitis and chronic idiopathic neutropenia: report of two cases. Mayo Clin Proc 1970; 45:494–504. KALKWARF KL, GUTZ DP. Periodontal changes associated with chronic idiopathic neutropenia. Pediatr Dent 1981;3: 189–195. KAMMA JJ, LYGIDAKIS NA, NAKOU M. Subgingival microflora and treatment in prepubertal periodontitis associated with chronic idiopathic neutropenia. J Clin Periodontol 1998;25: 759–765. KOENE HR, DE HAAS M, KLEIJER M, HUIZINGA TW, ROOS D, vON DEM BORNE AE. Clinical value of soluble IgG Fc receptor 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. type III in plasma from patients with chronic idiopathic neutropenia. Blood 1998;91:3962–3966. FURUKAWA T, TAKAHASHI M, MORIYAMA Y, KOIKE T, KUROKAWA I, SHIBATA A. Successful treatment of chronic idiopathic neutropenia using recombinant granulocyte colony-stimulating factor. Ann Hematol 1991;62:22–24. HEUSSNER P, HAASE D, KANZ L, FONATSCH C, WELTE K, FREUND M. G-CSF in the long-term treatment of cyclic neutropenia and chronic idiopathic neutropenia in adult patients. Int J Hematol 1995;62:225–234. PAPADAKI HA, CHAROULAKIS N, ELIOPOULOS DG, PSYLLAKI M, ELIOPOULOS GD. Patients with nonimmune chronic idiopathic neutropenia of adults have increased splenic volume on ultrasonography. Clin Lab Haematol (in press). PAPADAKI HA, MARGIORIS AN, MILIAKI M, STERIOPOULOS C, VALATAS V, ELIOPOULOS GD. Chronic idiopathic neutropenia of adults is associated with decreased bone mineral density and alterations in bone turnover biochemical markers. Eur J Haematol 1999;62:311–316. WHO Study Group. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. World Health Organiz Techn Rep Ser 1994;843:1–129. AUNER HW, KLINTSCHAR M, CREVENNA R, et al. Two case studies of chronic idiopathic neutropenia preceding acute myeloid leukaemia. Br J Haematol 1999;105:431–433. PAPADAKI HA, KOSTEAS M, GEMETZI C, ALEXANDRAKIS M, ELIOPOULOS GD. Acute myeloid leukemia with aberrant phenotype and complex karyotye but no mutations in G-CSF receptor in two patients with non-immune chronic idiopathic neutropenia of adult. Ann Hematol 2001 (submitted). PAPADAKI HA, COULOCHERI SA, XYLOURI I, CHATZIVASSILI A, KARKAVITSAS N, ELIOPOULOS GD. Defective natural killer cell activity of peripheral blood lymphocytes correlates with the degree of neutropenia in patients with chronic idiopathic neutropenia of adults. Ann Hematol 1998;76:127–134. PAPADAKI HA, KATRINAKIS G, ELIOPOULOS AG, COULOCHERI S, ALEXANDRAKIS M, ELIOPOULOS GD. Low numbers of peripheral blood natular killer cells in patients with chronic idiopathic neutropenia of adults are probably due to enhanced extravasation into the tissues. Haema 1998;1:33–40. KYRIAKOU D, PAPADAKI HA, SAKELLARIOU D, ELIOPOULOS AG, KAPSIMALI V, ELIOPOULOS GD. Flow-cytometric analysis of peripheral blood lymphocytes in patients with chronic idiopathic neutropenia of adults. Ann Hematol 1997;75:103–110. BISHOP CR, ROTHSTEIN G, ASHENBRUCKER HE, ATHENS JW. Leukokinetic studies. XIV. Blood neutrophil kinetics in chronic steady-state neutropenia. J Clin Invest 1971;50:1678– 1689. BIASI D, BAMBARA L-M, CARLETTO A, et al. Increased in vitro neutrophil adherence in a case of chronic idiopathic neutropenia. Br J Haematol 1994;86:421–424. RICKARD KA, BROWN RD, YUEN E, KRONENBERG H. Chronic idiopathic neutropenia. Granulopoietic assessment by bone marrow culture and granulocyte kinetics. Am J Hematol 1980;9:193–198. GREENBERG P, MARA B, STEED S, BOXER L. The chronic idiopathic neutropenia syndrome: correlation of clinical features with in vitro parameters of granulopoiesis. Blood 1980;55:915–921. ELIOPOULOS GD, MELETIS J, FESSAS P, ANAGNOU N. Defective CSA-dependent granulopoiesis in patients with chronic neutropenia. Haematologia 1990;23:101–109. KOHGO Y, HIRAYAMA Y, MATSUNAGA T, KATO J, SAKAMAKI S, NIITSU Y. Chronic idiopathic neutropenia associated with abnormal expression of granulocyte colony-stimulating factor mRNA of bone marrow stromal cells. Int J Hematol 1994;59:177–180. MORIKAWA K, YAMAMORI S, OIWA K, et al. Impaired G-CSF 43 Papadaki et al. 51. 52. 53. 54. 55. 56. 57. 44 production at post-transcriptional level in a patient with chronic idiopathic neutropenia. Br J Haematol 1993;85:200– 202. VOGEL JM, KIMBALL HR, WOLFF SM, PERRY S. Etiocholanolone in the evaluation of marrow reserves in patients receiving cytotoxic agents. Ann Int Med 1967;67:1226–1238. DALE DC, FAUCI AS, GUERRY D, WOLFF SM. Comparison of agents producing a neutrophilic leukocytosis in man. Hydrocortisone, prednisone endotoxin, and etiocholanolone. J Clin Invest 1976;56:808–813. PAPADAKI HA, GIOUREMOU K, ELIOPOULOS GD. Overexpression of L-selectin may account for enhanced neutrophil extravasation in patients with chronic idiopathic neutropenia of adults. Haema 2000;3:29–35. FINCH SC. Neutropenia. In: WILLIAMS WJ et al. (eds), Hematology, 3rd edn, London: McGraw-Hill, 773–793. MILLER CB, JONES RJ, PIANTADOSI S, ABELOFF MD, SPIVAK JL. Decreased erythropoietin response in patients with anemia of cancer. N Engl J Med 1990;322:1689–1692. BLAKE DR, WATERWORTH RF, BACON PA. Assessment of iron stores in inflammation by assays of serum ferritin concentrations. Br Med J 1981;283:1147–1148. PAPADAKI HA, ELIOPOULOS DG, VALATAS V, ELIOPOULOS GD. Anemia of chronic disease is the more frequent type of anemia seen in patients with chronic idiopathic neutropenia of adults. Ann Hematol 2001;80:195–200. 58. TEPPO A-M, MAURY CPJ. Radioimmunoassay of tumor necrosis factor in serum. Clin Chem 1987;33:2024–2027. 59. SPIVAK J. Cancer-related anemia: its causes and characteristics. Semin Oncol 1994;21:3–8. 60. DANCEY JT, BRUBAKER LH. Neutrophil marrow in chronic benign idiopathic neutropenia. Am J Med 1980;68:251–254. 61. CLINE JM. The White Cell. Cambridge, MA: Harvard University Press, 1975;155. 62. PICKER LJ, FURST A, ROBINSON SH, KADIN ME. Immunoarchitecture of bone marrow in neutropenia: increased HNK1+ cells define a subset of neutropenic patients. Am J Hematol 1987;25:29–41. 63. PAPADAKI HA, ELIOPOULOS DG, PONTICOGLOU C, ELIOPOULOS GD. Increased frequency of monoclonal gammopathy of undetermined significance in patients with non-immune chronic idiopathic neutropenia syndrome. Int J Hematol 2001;73:339–345. 64. ISLAM KB, BASKIN B, NILSSON L, HAMMARSTROM L, SIDERAS P, SMITH CI. Molecular analysis of IgA deficiency. Evidence for impaired switching to IgA. J Immunol 1994;152:1442–1452. 65. CARLOS TM, HARLAN JM. Leukocyte-endothelial adhesion molecules. Blood 1994;84:2068–2101. 66. AZUMA Y, KAJI K, KATOGI R, TAKESHITA S, KUDO A. Tumor necrosis factor-alpha induces differentiation of and bone resorption by osteoclasts. J Biol Chem 2000;275:4858–4854. 67. ROODMAN GD. Cell biology of the osteoclast. Exp Hematol 1999;27:1229–1241.