Heterogeneity of bone marrow-directed immune mechanisms in the pathogenesis of neutropenia of Felty's syndrome

Felty's syndrome

Felty's syndrome was first described in 1924 by the US-American physician Augustus Roi Felty as a triad of rheumatoid arthritis, splenomegaly and leucopenia. Even nearly 100 years later, this rare syndrome is still paralleled by diagnostic and therapeutic challenges and its pathogenesis is incompletely understood. Neutropenia with potentially life-threatening infections is the main problem and several pathomechanisms like Fas-mediated apoptosis, anti-neutrophil antibodies, anti-G-CSF antibodies, neutrophil consumption in the context of NETosis and suppression of granulopoiesis by T-LGLs have been suggested. Felty's syndrome has various differential diagnoses as splenomegaly and cytopenia are common features of different infectious diseases, malignancies and autoimmune disorders. Additionally, benign clonal T-/NK-LGL lymphocytosis is increasingly noticed in Felty's syndrome, which further complicates diagnosis. Today's treatment options are still sparse and are largely based on case reports and small case series. Methotrexate is the mainstay of therapy, followed by rituximab, but there is less evidence for alternatives in the case of adverse reactions or failure of these drugs. This article gives an updated review about Felty's syndrome including its pathogenesis and treatment options.

[Rheumatic and hemato-/oncological disorders]

The association of rheumatic diseases with common and some of the less common hematological features such as different forms of anemia, leukopenia and thrombopenia are described in this article. In addition, the occurrence of malignancy in rheumatoid arthritis, systemic lupus erythematosus, myositis and scleroderma and possible causes are discussed. On the other hand, this review also focuses on various rheumatological manifestations of hematological diseases such as leukemia and lymphoma. The aim of the article is to give an overview of the various associations between rheumatological and hemato-/oncological diseases that have to be considered in clinical practice.

Autoantibodies against granulocyte-macrophage colony stimulating factor and interleukin-3 are rare in patients with Felty's syndrome

OBJECTIVES

Antibodies against granulocyte colony stimulating factor are frequently found in patients with Felty's syndrome (FS). In this study, we examined the prevalence of antibodies against two other granulopoietic cytokines: granulocyte-macrophage colony stimulating factor (GM-CSF) and interleukin-3 (IL3).

METHODS

Sera of 32 patients with FS, 20 normocytic patients with rheumatoid arthritis (RA), and 72 healthy individuals were screened for the presence of antibodies against GM-CSF and IL3 by ELISA and bioassays, using the human erythroleukaemia cell line TF-1.

RESULTS

In two of the 32 patients with FS, antibodies to GM-CSF and IL3 were detectable by ELISA. Binding anti-GM-CSF antibodies were also detected in one of the 72 healthy controls, while in another healthy subject and in one of the patients with normocytic RA, anti-IL3 antibodies were present. Serum from one of the two patients with FS who tested positive for anti-IL3 and anti-GM-CSF antibodies by ELISA showed strong neutralising capacity to the biological effect of IL3, but not to GM-CSF in vitro. Patients with FS had significantly higher serum levels of GM-CSF (median; 2.82 pg/mL; interquartile range 2.64-3.19 pg/mL) compared with patients with RA (2.52 pg/mL; 2.28-2.72 pg/mL; p = 0.012) and healthy controls (2.23 pg/mL; 2.04-2.52; p<0.001). In addition, serum levels of IL3 in patients were significantly higher in FS (10.05 pg/mL; 8.94-11.98) compared with controls (4.79 pg/mL; 3.72-7.22; p<0.001), but not compared with RA patients (9.52 pg/mL; 8.32-10.42; p = 0.17).

CONCLUSIONS

Antibodies to GM-CSF and IL3 are rare in patients with FS and RA and in healthy subjects. In individual patients with FS, the presence of neutralising anti-IL3 antibodies may contribute to the development of cytopenia.

Autoantibodies against granulocyte colony-stimulating factor in Felty's syndrome and neutropenic systemic lupus erythematosus

OBJECTIVE

Cytokines and growth factors can be a target of autoantibodies in systemic inflammatory diseases. We examined whether patients with neutropenia and either Felty's syndrome (FS) or systemic lupus erythematosus (SLE) have autoantibodies against granulocyte colony-stimulating factor (G-CSF) and whether these autoantibodies are functionally relevant.

METHODS

Fifteen patients with neutropenia due to FS were matched for age, sex, and disease activity with 16 normocytic rheumatoid arthritis (RA) control patients. Sixteen patients with SLE and neutropenia were matched with 16 normocytic SLE control patients. Antibodies against G-CSF were measured by enzyme-linked immunosorbent assay and Western blotting. Antibody specificity was verified by competitive inhibition using recombinant human G-CSF. The effect of anti-G-CSF antibodies on the functional activity of their target molecule was measured in a bioassay using G-CSF-sensitive murine 32D cells.

RESULTS

IgG anti-G-CSF was found in 11 FS patients, 6 SLE patients with neutropenia, 6 SLE control patients, and none of the RA control patients. IgM anti-G-CSF was found in 6 neutropenic and 3 normocytic SLE patients. Anti-G-CSF antibodies were associated with an exaggerated serum level of G-CSF and a low neutrophil count. A neutralizing effect of anti-G-CSF antibodies on its target molecule was found in 3 of the 9 patients tested. Irrespective of the presence or absence of anti-G-CSF antibodies, neutropenic patients with FS and SLE had exaggerated serum levels of G-CSF.

CONCLUSION

Anti-G-CSF autoantibodies are common in neutropenia due to FS and SLE. In individual patients, these autoantibodies have a neutralizing capacity. In patients without neutralizing antibodies, hyposensitivity of the myeloid cells to G-CSF appears to be central to the pathogenesis of the neutropenia in FS and SLE.

Treatment of severe neutropenia due to Felty's syndrome or systemic lupus erythematosus with granulocyte colony-stimulating factor

OBJECTIVES

To examine the efficacy and safety of recombinant human granulocyte colony-stimulating factor (rhG-CSF) and recombinant human granulocyte-macrophage colony-stimulating factor (rhGM-CSF) for the treatment of severe neutropenia due to Felty's syndrome (FS) or systemic lupus erythematosus (SLE).

METHODS

Eight patients with absolute neutrophil counts (ANC) below 1,000/microL attributable to FS (n = 4) or SLE (n = 4) were treated with rhG-CSF. The hematologic and clinical response as well as side effects were recorded. In addition, reports on the use of rhG-CSF/rhGM-CSF in FS and SLE retrieved from the English language literature were analyzed.

RESULTS

RhG-CSF effectively corrected neutropenia due to FS and SLE in seven of the current eight patients. In 54 of 55 FS and SLE patients retrieved from the literature, G-CSF or GM-CSF, respectively, proved to be effective at elevating the neutrophil count, which was often associated with improvement of infectious complications. The neutrophil count often declined again when growth factor treatment was stopped but generally stabilized at a level that exceeded the pretreatment count. Side effects included rare cases of thrombocytopenia, arthralgias, and development of cutaneous leukocytoclastic vasculitis. Side effects were dose dependent and resolved when treatment was discontinued. One of our own patients and 17 previously reported patients continued to benefit from long-term administration of rhG-CSF over periods of more than 40 months.

CONCLUSIONS

RhG-CSF is an effective and generally well-tolerated treatment for neutropenia due to FS or SLE. Exacerbation of the underlying rheumatic condition due to G-CSF appears to be rare if G-CSF is administered at the lowest dose effective at elevating the ANC above 1,000/microL.

Large granular lymphocyte expansions in patients with Felty's syndrome: analysis using anti-T cell receptor V beta-specific monoclonal antibodies

Felty's syndrome (FS), the association of rheumatoid arthritis (RA) and idiopathic neutropenia, remains an unexplained phenomenon. HLA-DR4 is found in over 90% of cases. Patients with FS may have a T cell lymphocytosis of CD3+CD8+CD57+ large granular lymphocytes (LGL syndrome). In this study of 47 patients with FS, 19% had clear evidence for LGL expansions, while in total 42% had variable evidence for the LGL syndrome using currently available techniques. Of these T cell expansions, 76% were clonal, as demonstrated by Southern blotting and analysis with T cell receptor (TCR) beta chain constant region probes. This technique may fail to detect clonal populations in some patients. Cytofluorographic analysis using antibodies specific for TCR V beta chains identified patients with clonal LGL expansions with results comparable to those obtained with Southern blotting. No evidence for shared V beta usage among expansions from different patients was seen. The role of LGL in RA and FS is currently unclear, but this technique offers a practical and accessible means of identifying patients with LGL expansions, as a starting point for further investigation.

Successful treatment of chronic wound infection in neutropenia and rheumatoid arthritis with filgrastim (rhG-GSF)

A 73-year-old woman was diagnosed with seropositive destructive rheumatoid arthritis in 1981. She was treated with cortisone, chloroquine, and cyclophosphamide (Sendoxan) in 1982 and 1984 and contracted severe neutropenia. After that she only received cortisone. During 1991, again low neutrophilic counts were registered, especially granulocytopenia. At first, B-cell lymphoma was suspected, but later Felty's syndrome was established. The patient was treated with high-dose cortisone with some success and had a few minor septic episodes. In May 1992 she contracted a traumatic wound on the back of the lower leg. Conservative treatment resulted in a worsening of the condition and an increased wound area, most likely related to the neutropenic condition. In mid July the patient was hospitalized. Bacterial isolates yielded mixed gram-negative enteric bacteria from the wound. Parenteral antibiotic treatment was started, followed by oral drugs, rhG-CSF (filgrastim) was given subcutaneously once a day, starting 3 days after admission. This resulted in increased numbers of peripheral granulocytes. The ulcer started to heal and by mid August the patient received a transplant with autologous skin grafting. In mid September the wound was completely healed. It is concluded that the combination of antibiotics, skin transplantation, and G-CSF was necessary for the successful result. Actually, the bacterial growth did not call for antibiotics, but it was considered necessary to cover for staphylococci. No worsening of the underlying arthritis was observed.

Autoimmune neutropenia manifesting as recurrent oral ulceration

There are many causes of recurrent oral mucosal ulceration, particularly gastrointestinal, mucocutaneous, and hematologic disease. The present report details the features of a patient with recurrent oral ulceration caused by autoimmune neutropenia.

Successful reversal of neutropenia in Felty's syndrome with recombinant granulocyte colony stimulating factor

We report two patients with Felty's syndrome and chronic skin ulcers treated successfully with recombinant granulocyte colony stimulating factor (GCSF). In both cases granulocytes returned to the normal range within days of starting treatment, and their cutaneous ulcers improved. In one patient granulocytes were maintained at normal levels with a regimen of GCSF 3 micrograms/kg twice weekly for 14 months.

Long-term correction of neutropenia in Felty's syndrome with granulocyte colony-stimulating factor

Neutropenia in Felty's Syndrome predisposes patients to recurrent bacterial infections. We have treated a patient for more than one year with G-CSF and ascertained that this growth factor can safely correct neutropenia over a long period of time. G-CSF may constitute a new agent for the treatment and prophylaxis of infection in Felty's syndrome.

GM-CSF in the treatment of Felty syndrome

Many therapeutic agents have been tried with variable success in the treatment of Felty neutropenia, but the reports are anecdotal. We now describe the second trial of recombinant granulocyte-macrophage colony-stimulating factor (GM-CSF), in a splenectomized, infected patient with Felty syndrome.

Investigation and management of gout in the young and the elderly

Images

Cyclosporine for the treatment of granulocytopenia in Felty's syndrome

A patient with Felty's syndrome was treated with cyclosporine, initially 10 mg/kg/day and then 4 mg/kg/day. The neutrophil count increased by 6 weeks and was normal at 3 months. Over the subsequent 27 months the neutrophil count was closely related to the cyclosporine dosage and there was no evidence of cyclosporine toxicity. This case indicates that cyclosporine may have a role in the treatment of Felty's syndrome.

Effect of corticosteroid treatment on hemopoiesis in vivo and in vitro in a patient with Felty's syndrome

We have studied the CFU-GM and BFU-E in vitro growth in a neutropenic and anemic patient with Felty's syndrome, either before or one and three months after steroid therapy when neutrophils and erythrocytes returned to normal. Both CFU-GM growth and CSA production were found to be low before therapy, and prednisone was shown to raise them to normal levels. The in vitro growth of BFU-E and the production of BPA by T lymphocytes of the patient were significantly lower than normal when studied before therapy. However, the T lymphocytes incubated in vitro with hydrocortisone regained their ability to stimulate the BFU-E growth. After prednisone therapy both BFU-E growth and BPA production by T lymphocytes returned to normal. Possible pathogenetic mechanisms of impaired granulo- and erythropoiesis in Felty's syndrome are discussed. The in vitro study with hydrocortisone can help to identify steroid-sensitive patients.

Phenotypic and genotypic analysis of mononuclear cells from patients with Felty's syndrome

Phenotypic and genotypic characteristics of the peripheral blood mononuclear cells in nine patients with Felty's syndrome have been examined. One patient had an increased number and percentage of peripheral blood mononuclear cells with the phenotype CD3+ Leu-7+ CD16+ and showed a clonal rearrangement of the T cell receptor B chain gene. The remaining eight patients all showed a germline configuration of the T cell receptor B chain gene. In two patients an increased proportion of CD3+ Leu-7+ CD16- peripheral blood mononuclear cells (45 (SD 11)% of peripheral blood mononuclear cells) were found, while the remaining six patients had proportions of CD3+ Leu-7+ cells similar to those of patients with uncomplicated rheumatoid arthritis. These data confirm that patients with Felty's syndrome are heterogeneous, with at least three different peripheral blood mononuclear cell phenotypic subsets. One subset is characterised by a clonal expansion of an unusual lymphocyte subpopulation, another by polyclonal expansion, and the third subset has the same proportions of peripheral blood mononuclear cells as patients with uncomplicated rheumatoid arthritis.

CFU-GM inhibitors in neutropenia

Peripheral blood lymphocytes from 20 patients with neutropenia not consistent with aplastic anaemia were tested for their ability to inhibit the proliferation of normal granulopoietic precursor cells (CFU-GM) in agar culture. Two patients, both with features of an autoimmune disorder, had lymphocytes which were more inhibitory than normal lymphocytes to both normal and their own CFU-GM. Two other patients had lymphocytes which were more inhibitory than normal lymphocytes to either their own CFU-GM or normal CFU-GM but not both. Eight patients had lymphocytes which were significantly less inhibitory than normal lymphocytes to either normal or their own CFU-GM, but only one showed this feature against both normal and their own CFU-GM. One patient had a highly potent plasma inhibitor of CFU-GM--this patient had received multiple transfusions and had a leucocyte antibody of a broad specificity. No clinical or haematological features were common to any of these groups of patients which reflects the heterogeneity of patients studied and stresses the importance of controls.

Reversal of neutropenia with methotrexate treatment in patients with Felty's syndrome. Correlation of response with neutrophil-reactive IgG

We evaluated the clinical and hematologic response to methotrexate (MTX) in 4 women with Felty's syndrome (FS) who had had neutropenia for 1-3 years. Since immune complexes or antineutrophil antibodies are implicated in the pathogenesis of the neutropenia of FS, we also measured both direct and indirect levels of neutrophil-reactive IgG. All 4 patients showed a prompt and dramatic increase in neutrophil counts within 1-2 months of starting MTX therapy. In 3 patients, the symptoms of arthritis also improved; in the fourth patient, arthritis worsened. Recurring infections ceased in 3 patients. Neutrophil-reactive IgG levels, which were elevated in all patients prior to treatment, decreased toward normal while the patients were receiving MTX therapy. We conclude that MTX is effective in treating the neutropenia of FS, in part by lowering neutrophil-reactive IgG.

The spleen in Felty's syndrome: a histological, morphometrical, and immunohistochemical study

A histological study employing morphometrical and immunohistochemical methods was performed in 3 spleens removed because of Felty's syndrome (FS). A comparison was made with control spleens and with spleens removed from patients with idiopathic thrombocytopenic purpura (ITP) or autoimmune hemolytic anemia (AIHA). In FS, the spleen is enlarged, mainly due to expansion of the red pulp. The sinuses are enlarged, too, and contain many macrophages. 2 of our specimens showed many hyperplastic germinal centers; the 3rd contained no germinal centers. Only in the former cases did splenectomy lead to an increase in the number of circulating granulocytes. Immunohistologically, FS spleens differed from the control, the ITP, and the AIHA spleens. The results have provided further evidence indicating that the pathogenesis of granulocytopenia in FS differs from that of autoimmune hematocytopenias, and have given rise to the hypothesis that, in FS, the spleen is the site of interaction between immune complexes and granulocytes.

Immunological aspects of the anemia of rheumatoid arthritis

In order to investigate the cause of the anemia concomitant with rheumatoid arthritis (RA), we examined, using the erythroid colony assay of human bone marrow colony-forming units-erythroid (CFU-e) and burst-forming units-erythroid (BFU-e), the effects of the patients' serum and peripheral blood T lymphocytes on the CFU-e-derived colonies. The counts of erythroid colonies of RA patients were markedly lower than those of human control subjects [CFU-e: control 152.9 +/- 30.6 (n = 19), RA 51.1 +/- 13.6 (n = 7), t = 7.66567, p less than 0.01; BFU-e: control 25.2 +/- 5.9 (n = 5), RA 12.6 +/- 2.6 (n = 7), t = 4.574, p less than 0.01]. The serum from two out of seven RA patients slightly inhibited the formation of CFU-e-derived colonies of human control subjects (t = 2.31, 0.05 less than p less than 0.1); however, the serum from the other five RA patients did not significantly inhibit human control erythroid colony formation as compared with human control serum (t = 0.981, 0.3 less than p less than 0.4). On the other hand, peripheral blood T lymphocytes of the patients markedly inhibited the formation of CFU-e-derived colonies of the control subjects as compared with peripheral blood T lymphocytes from human control subjects (t = 4.24, p less than 0.01). The above-mentioned results suggest that the peripheral blood T lymphocytes of RA patients might play a role as one of the causes of the concomitant anemia of RA patients.

Regulation of hematopoiesis by T lymphocytes and natural killer cells

T lymphocytes and natural killer (NK) cells exert both stimulatory and suppressive effects that regulate growth and differentiation of hematopoietic cells. Activated T and NK cells have been demonstrated in different pathological states of bone marrow failure and are proposed to play a role in the pathogenesis of the disease. T and NK cells have also been shown to be responsible for bone marrow graft rejection in both allogeneic and syngeneic donor/recipient combinations. Lymphocytes can regulate hematopoietic cell growth by direct cellular contact or by releasing soluble factors, such as colony-stimulating factors, immune interferon, lymphotoxin, and tumor necrosis factor, active on hematopoietic precursor cells.

Depressed superoxide radical generation by neutrophils from patients with rheumatoid arthritis and neutropenia: correlation with neutrophil reactive IgG

Neutrophils of 31 patients with neutropenia and rheumatoid arthritis (RA) have been studied to assess their ability to generate superoxide radicals (O-2) on activation. Seventeen patients had classical Felty's syndrome and 14 presumed chrysotherapy related neutropenia. Results were compared with those from age and sex matched controls with uncomplicated RA and from normal subjects. Neutrophils from patients with Felty's syndrome had a significantly reduced ability to generate superoxide radicals when compared with the other three groups. In addition, serum levels of IgG polymorphonuclear leucocyte binding activity (IgG PBA) were also raised in the group with Felty's syndrome. A statistically significant inverse correlation existed between O-2 generation and IgG PBA. It is concluded that neutrophil reactive IgG may have an important role in both quantitative and qualitative defects in neutrophil function in Felty's syndrome.

Immune complexes and the pathogenesis of neutropenia in Felty's syndrome

The effect of the injection of serum from patients with rheumatoid arthritis (RA) and Felty's syndrome (FS) into mice on the number of circulating polymorphonuclear cells (PMN) was studied. The number of circulating PMN dropped to 61% (range 34-98%) of the initial counts after the injection of FS serum. This phenomenon was observed less frequently after injection of RA serum. In contrast, injection of serum from healthy controls always resulted in an immediate increase in the number of circulating PMN. No decrease in PMN counts was found after injection of FS sera pretreated with polyethylene glycol to precipitate immune complexes (IC). Gel filtration of FS sera on Sepharose 4B showed that the effect on the PMN counts in mice did not coincide with the 7S peak but occurred only in fractions containing larger material. Serum fractions from FS patients that contained IC were more active in producing neutropenia than the corresponding fractions from patients with RA. Microscopic and immunohistochemical examination of the organs from mice injected with FS serum showed sequestration of PMN and deposition of human IgG, IgA, and IgM in the vascular bed of the lungs. These results indicate that the interaction between PMN and IC of patients with FS leads to sequestration of PMN in mice and suggests that this interaction in humans may have a role in the pathogenesis of FS.

Gamma interferon and lymphotoxin, released by activated T cells, synergize to inhibit granulocyte/monocyte colony formation

We have shown that lymphocytes stimulated by PHA produce colony-forming unit of granulocyte/monocyte (CFU-GM)-stimulating and -inhibiting activities, IFN-gamma, and lymphotoxin (LT). IFN-gamma is necessary for inhibition of CFU-GM by PHA-conditioned medium (CM), as shown by experiments in which removal of IFN-gamma from PHA-CM abrogated inhibition. However, experiments in which rIFN-gamma was added to IFN-gamma-depleted PHA-CM revealed the presence, in PHA-CM, of other factors that act in synergy with IFN-gamma to inhibit CFU-GM. Fractionation of PHA-CM on a Sephadex G-100 column was used to separate IFN-gamma and LT. Colony-inhibiting activity was eluted in fractions that contained both IFN-gamma and LT activities, identifying LT as a factor present in PHA-CM that synergizes with IFN-gamma to inhibit CFU-GM. Treatment of PHA-CM with mAb against either IFN-gamma or LT completely abrogated the colony-inhibiting activity, demonstrating a requirement for both lymphokines in PHA-CM-induced inhibition of CFU-GM. Experiments using rIFN-gamma and preparations of purified LT confirmed that neither lymphokine alone, when added to bone marrow cells at the concentrations present in PHA-CM, strongly inhibited day 7 or day 14 CFU-GM, but that the two lymphokines, added together, behaved synergistically to inhibit CFU-GM by up to 70%. The inhibition observed using purified preparations of lymphokines shows that synergy between IFN-gamma and LT is sufficient to explain PHA-CM-induced inhibition of CFU-GM. Our findings suggest that activated T cells regulate hematopoiesis through the release of inhibitory as well as stimulatory factors, and that the simultaneous production of IFN-gamma and LT may represent a mechanism of suppression of hematopoiesis in the cases of bone marrow failure associated with the presence of activated T cells.

Felty syndrome: autoimmune neutropenia or immune-complex-mediated disease?

Immunofluorescence on polymorphonuclear cells (PMN) of patients with Felty syndrome (FS) revealed increased amounts of IgG, IgA, and IgM bound to the PMN surface compared with PMN of patients with rheumatoid arthritis alone. A positive correlation was found between the score for surface-bound immunoglobulins on FS-PMN and the results of the Clq binding assay in FS sera. After preincubation with sera from 20 patients with FS, immunofluorescence on PMN from healthy controls (HC) showed that these cells had bound IgG, IgA, and IgM. However F(ab')2 fragments of IgG from FS sera did not bind to PMN, although the antigen-binding reactivity of the F(ab')2 fragments was maintained as shown by control experiments. Immunoglobulins eluted from FS-PMN failed to bind to HC-PMN, whereas the corresponding IgG of patients with autoimmune neutropenia was bound. Gel filtration of FS sera on Sepharose 4B showed that the binding of IgG in FS sera to PMN did not coincide with the 7S peak but occurred mainly in fractions containing larger material. No binding of IgA and IgM to HC-PMN was found after incubation with FS sera pretreated with polyethylene glycol (PEG) to precipitate immune complexes. These results indicate that in sera of patients with FS the PMN-binding reactivity of IgG, IgA, and IgM is due to the binding of immune complexes containing these immunoglobulins and not to presence of autoantibodies directed to antigens on the neutrophil surface.