Rheumatological Manifestations of Malignant Haematological Conditions

Edward M Vital

Paul Emery

Academic Unit of Musculoskeletal Disease, Chapel Allerton Hospital, Leeds

Reports on the Rheumatic Diseases Series 5 : Topical Reviews

• Haematological malignancy may present as a rheumatic disease; rheumatologists must have a high degree of clinical suspicion when assessing new patients
  
• Certain rheumatologic presentations may be a paraneoplastic phenomenon associated with underlying haematological malignancy
  
• B-lymphoproliferative disease and autoimmune diseases such as rheumatoid arthritis, systemic lupus erythematosus and Sjögren's syndrome may have related B-cell abnormalities
• There is a move in rheumatology therapeutics to B-cell-directed therapies using a similar strategy to the treatment of B-cell lymphoma. Such strategies may be especially beneficial for patients with both malignant and autoimmune clinical features

Malignant haematological disease and rheumatic disease: a two-way association with implications for therapy

Many of the rheumatic diseases may be complicated by malignancy. The association of rheumatoid arthritis (RA), systemic lupus erythematosus (SLE) and Sjögren's syndrome with lymphoproliferative disease has received much attention. The cause of these malignancies seems to be at least partly related to the pathological process of autoimmunity itself. That this association is solely due to cytotoxic therapy seems unlikely: the increased incidence of non-Hodgkin's lymphoma (NHL) in RA was noted before cytotoxic therapy became a common treatment, and occurs frequently in patients who have never been exposed to such treatments.1 There is however continuing concern that therapies may amplify the risk of lymphoma in autoimmune disease. Another proposed hypothesis for the increased incidence of NHL in autoimmune disease is that viral infections such as Epstein–Barr virus that predispose to B-cell lymphomas are being activated by the immune suppression caused by both treatment and the disease itself. However, a recent pathological review found that the incidence of this virus in patients with autoimmune disease complicated by NHL was in fact very low.2

Overall, then, there seems to be an independent association between autoimmunity and lymphoproliferative disease. The association of these antibody-positive autoimmune diseases with lymphoproliferative disease, most frequently diffuse B-cell lymphoma, suggests there may be common features in the aetiology of these diseases. This association is made more striking by the observation that not only is lymphoproliferative disease a complication of autoimmune disease, but the reverse also seems to occur: lymphoproliferative disease may present as an autoimmune rheumatic disease, or may be complicated by rheumatic disease at a later stage. Lymphoproliferative diseases are commonly complicated by autoimmune diseases in a number of systems.

Exploration of the relationship between lymphoproliferative disease and autoimmunity may provide important clinical lessons in the diagnosis and treatment of the patient presenting with rheumatic disease, and also help in furthering our understanding of the aetiology of both rheumatological and haematological disorders. The recent realisation of the importance of the B-cell in a range of rheumatic diseases and the association of those same diseases with B-cell malignancies emphasises the argument for a focus on B-cell-directed therapies.

Arthritis in leukaemia

Musculoskeletal symptoms and presentations are common in leukaemia. A synovitis similar to inflammatory arthritis may occur due to infiltration of malignant cells in the synovium, but there are a number of other mechanisms. Leukaemic infiltration may occur in the juxta-articular bone or the metaphyseal periosteum, or there may be haemorrhage into the joint.3 Alternatively, arthritis may occur as an immune-mediated paraneoplastic phenomenon. There are reports of leukaemic synovitis in acute lymphocytic leukaemia,4,5 in myelomonocytic blast crisis in chronic myeloid leukaemia,6 and in chronic lymphocytic leukaemia (CLL).7 Commonly, patients with leukaemia complain not of arthritis but of a more non-specific bone pain, the cause of which may be unclear. In some patients osteolytic lesions, periosteal reaction or metaphyseal rarefaction are seen on plain x-ray, or the bone scan is abnormal.8

True leukaemic synovitis is more common in children than in adults and more common in acute than chronic leukaemias. In acute leukaemias arthritis is usually asymmetric oligoarthritis and forms part of the initial presentation. In chronic leukaemias a symmetric arthritis occurring later in the disease is more common as a paraneoplastic feature. Arthritis may occur in 14–65% of children with leukaemia.9 In one series, 50% of children with leukaemia had joint abnormalities, including joint effusion, soft tissue swelling or tenderness, but most were mild and only a small number of these were symptomatic.10 10% of children with leukaemia present with arthritis, and it is important to note that this may be weeks to months before the diagnosis of leukaemia and that initially full blood count and bone marrow biopsy may be normal. Non-steroidal anti-inflammatory drugs are often ineffective in arthritis due to leukaemia, and carry additional risk because of the risk of gastrointestinal haemorrhage. Most patients however respond well to chemotherapy.

In a series of 139 adults with acute leukaemia, 5.8% presented with rheumatic disease, which preceded the diagnosis of leukaemia by an average of 3.25 months.11 In these patients the most common pattern of presentation was a seronegative spondyloarthropathy or reactive arthritis pattern, with asymmetrical large joint arthritis predominantly in the lower limbs and low back pain. This syndrome, sometimes known as 'cancer polyarthritis', may occur in association with a variety of malignancies and often has an explosive onset. While such a presentation would usually be considered a reactive arthritis in a younger person, in older patients, in patients with no clear evidence of a triggering infection, or in patients who fail to respond to treatment or spontaneously resolve, underlying malignancy should be considered. Clinical outcome from leukaemia in groups with or without a rheumatic presentation was no different.

Arthritis in lymphoma

7–25% of patients with NHL develop some kind of musculoskeletal symptoms; however most of these are due to bone metastasis or a primary lymphoma of bone rather than true arthritis. When arthritis does occur it is usually a paraneoplastic, cytokine-driven phenomenon rather than infiltration of the joint by malignant cells. It is clinically important to note that in cases where synovial biopsies have been obtained that demonstrated leukaemic infiltration, synovial fluid has been only mildly inflammatory.12,7 Therefore if there is a high degree of suspicion of leukaemic infiltration on clinical and imaging features, negative synovial fluid analysis does not exclude the diagnosis.

Birlik et al13 report an interesting case of a patient presenting with seronegative symmetrical small joint polyarthritis, which was initially diagnosed and treated as seronegative RA. Following a suspicion of septic arthritis in a proximal interphalangeal joint, synovial biopsy demonstrated infiltration of the synovium by B-cell NHL. Thus it appears possible that malignant infiltration of the joint may present as a symmetric small joint polyarthritis rather than the usual large joint mono- or oligoarthritis. This case highlights the importance of vigilance for underlying malignancy in patients presenting with inflammatory arthritis. In this case, poor response to disease-modifying anti-rheumatic drugs (DMARDs) and corticosteroids, disproportionate pain for the clinical abnormality and early significant osteopenia were clues to underlying malignancy, and should prompt consideration of magnetic resonance imaging and synovial biopsy.

Another interesting feature in this patient was widespread infiltration of the synovium without visceral and nodal involvement. This suggests that malignant cells may have been subject to some kind of homing signal for the joint. The fact that cases of polyarthritis due to malignant infiltration seem to be a rare feature of NHL, however, demonstrates that such affinity for the joint is not the norm. In standard RA there is marked infiltration by non-malignant lymphocytes, which frequently form ectopic secondary lymphoid structures that correlate to severity of synovitis.14 Patients with B-cell NHL may also develop a paraneoplastic RA with a predominantly T-cell synovial infiltrate more typical of standard RA.15 In cases such as these we may therefore hypothesise a combination of autoimmune and malignant factors leading to infiltration of the joint by malignant B-cells.

In a patient presenting with inflammatory arthritis, when should I consider underlying malignancy?
Severe joint pain disproportionate to physical findings Explosive onset Poor response to steroids and DMARDs Early significant osteopenia or lytic lesions on x-ray Hepatosplenomegaly or lymphadenopathy Abnormal differential white cell count Paraprotein on serum electrophoresis

Other rheumatic paraneoplastic phenomena occurring with haematological malignancy

A number of other rheumatic disorders commonly occur as a paraneoplastic phenomenon. Some of these are particularly suggestive of underlying haematological malignancy and should prompt a high degree of clinical suspicion or further investigation.

Paraneoplastic eosinophilic fasciitis

Eosinophilic fasciitis is a disorder of fascial inflammation that results in cutaneous induration (predominantly in the extremities), tenosynovitis, inflammatory arthritis and sometimes visceral involvement such as oesophageal dysmotility. Joint contractures are common due to fascial thickening with or without arthritis. Although some of the changes are macroscopically similar to those seen in scleroderma, the aetiology and histopathology are different, with none of the vascular changes (including Raynaud's phenomenon) that are thought to be the primary abnormality in scleroderma, and fibrosis that is localised to the fascia rather than the dermis. There are a number of case reports of eosinophilic fasciitis occurring as a paraneoplastic disorder, commonly in association with haematological malignancies. A review of 12 cases revealed that many of these occur some time prior to the diagnosis of cancer, by a median of 1 year. As in many other paraneoplastic disorders, there seems to be relative resistance to steroid treatment, although the fasciitis may remit after treatment of the underlying malignancy.16

Paraneoplastic vasculitis

Vasculitis may be a presenting feature or a complication of malignancy. In a review of 200 reported cases of vasculitis associated with malignancy, 155 were associated with haematological malignancies.17 The most common vasculitides were cutaneous leukocytoclastic vasculitis (43%) and polyarteritis nodosa (PAN – 10.5%). Smaller numbers of patients had nerve or muscle microvasculitis, granulomatous angiitis of the central nervous system, temporal arteritis, Henoch–Schönlein purpura or Churg–Strauss syndrome or unclassified forms. The most common haematological malignancies causing a vasculitis were lymphoproliferative: hairy-cell leukaemia (28%), NHL (21%) and Hodgkin's disease (14%). Others had myeloid or plasma cell malignancies. Unfortunately, details of antibody status are not available for all of these patients. Many plasma cell or B-lymphoproliferative disorders may also cause a vasculitis due to the production of cryoglobulins, which are discussed separately below.

Leukocytoclastic vasculitis is a small-vessel vasculitis of the skin. Clinically it is characterised by crops of macules or urticarial papules that progress to become purpuric and necrotic. Often there will be clinical evidence of other systemic manifestations such as fever, arthralgia and elevation of inflammatory markers. Treatment of the underlying disease may result in improvement of the vasculitis, or it may respond to corticosteroids. Histologically, small vessels are affected by neutrophil infiltration of the vessel wall with leukocytoclasis (fragmented neutrophils) and extravasation of red blood cells. Immunofluorescence often demonstrates immunoglobulin M (IgM) or IgG deposition (in Henoch–Schönlein purpura IgA deposition is seen).

Skin lesions are also a common feature of PAN, but many patients present with predominantly non-specific systemic symptoms such as fever and weight loss or visceral involvement (mesenteric ischaemia, coronary vasculitis and mononeuritis multiplex are common). Histologically, small- to medium-sized arteries are affected predominantly in a segmental pattern, with a transmural infiltrate of neutrophils and lymphocytes resulting in fibrinoid necrosis and microaneurysm formation. Malignancy-associated PAN differs from classical PAN by an increased frequency of peripheral artery aneurysms, which in some cases may be due to direct invasion of the vessel wall by malignant cells, and less frequent production of antinuclear antibody and rheumatoid factor (RF). It is most strongly associated with hairy-cell leukaemia.

The pathogenesis of paraneoplastic vasculitis remains unclear, but given the heterogeneity of causative malignancies and vasculitic phenomena a single underlying mechanism seems unlikely.

Cryoglobulinaemia

A vasculitis-like syndrome may be caused by the cryoglobulins that are produced as a complication of a number of haematological diseases. A cryoglobulin is either a monoclonal antibody (Type I) or an immune complex (Type II or III, both known as mixed cryoglobulins) that precipitates below 37°C. The reasons why certain antibodies possess this property while others do not is unknown. Alterations in protein conformation may occur with lower temperatures, but in the case of hepatitis C virus (HCV)-associated cryoglobulins the presence of viral antigen within the immune complex may also be instrumental.18 Any of the haematological disorders that are usually associated with a paraprotein may produce a Type I cryoglobulin. Proliferative disorders of B-cells and plasma cells such as myeloma, Waldenström's macroglobulinaemia and CLL produce IgM, IgG, IgA or pure light chain cryoglobulins, depending on the cell of origin. Precipitation of this paraprotein results in hyperviscosity and physical occlusion of small vessels, especially in the peripheries where a purpuric rash results. Complement deposition in the vessel wall, however, is a less prominent feature than in mixed cryoglobulinaemia, resulting in fewer of the classically inflammatory features of the latter, such as arthritis.

Mixed cryoglobulinaemias result from a chronic inflammatory state, such as connective tissue disease (CTD), where polyclonal antibody is produced in response to a persistent antigen. While many cases of mixed cryoglobulinaemia were once described as 'essential', the majority of such cases are actually now known to be secondary to hepatitis C. Whatever the cause, mixed cryoglobulinaemias are characterised by production of polyclonal IgG associated with an RF that may be either mono- (Type II) or polyclonal (Type III). It is the presence of this RF that allows formation of cryoprecipitable complexes: neither the IgG nor RF component alone will precipitate. In mixed cryoglobulinaemia, immune complex deposition and complement fixation result in inflammatory vasculitis in addition to hyperviscosity and hence features such as glomerulonephritis are more common than in Type I cryoglobulinaemia. There is evidence for a clonally heterogeneous intermediate 'Type II–III' cryoglobulinaemia.19 In HCV-associated cryoglobulinaemia there may be early polyclonal expansion of IgM–RF-producing B-cells followed by later emergence of a dominant clone, so patients with oligoclonal RF may represent a transitional stage.20

Conventional treatments for cryoglobulinaemia are unsatisfactory (for a review see Ferri and Mascia 200621). The most effective treatment is usually one that is directed at correction of the underlying cause, but this is often not achievable. In cases of cryoglobulinaemia associated with CTD immunosuppression with agents such as cyclophosphamide may be effective, but effects may be transient and this therapy is limited by significant toxicity such as premature ovarian failure. In patients with HCV-associated cryoglobulinaemia there is a therapeutic dilemma: immunosuppressive agents may control the vasculitic phenomena of the cryoglobulins, but result in progression of the viraemia and worse long-term outcomes. Cryoglobulinaemia associated with haematological malignancy is often associated with high-titre cryoglobulins. Plasmapheresis may be effective in the short term. Treatment of underlying plasma cell disorders may be effective if possible.

Rituximab is an agent that promises much improved outcomes for many patients with cryoglobulinaemia that has failed to respond to conventional therapy, and may become the treatment of choice in some situations, particularly the mixed cryoglobulinaemias. B-cell depletion with rituximab is known to be an effective therapy for a number of autoimmune diseases. In addition, rituximab depletes the B-cells responsible for the RF component of a cryoglobulin. By both treating the underlying autoimmune disease and removing the RF producing B-cells, cryoglobulin production may be abrogated. Evidence is currently case reports only, but these are promising when used in essential, autoimmune-associated or HCV-associated cryoglobulinaemia.22

In Type I cryoglobulinaemia associated with haematological disease the potential efficacy of rituximab depends on the cell of origin. Cryoglobulin production by CD20– plasma cell clones will not be affected. In three published cases, one responded well.23 One CD20+ case worsened after rituximab with a rise in cryoglobulin level, and another failed to respond.24 In Waldenström's macroglobulinaemia rituximab has also shown efficacy, and it should be noted that initial rise in cryoglobulin level may occur with a good eventual outcome.25 Further evaluation of rituximab for CD20+ B-cell clonal disorders is therefore needed.

Remitting seronegative symmetric synovitis with pitting (o)edema

Remitting seronegative symmetric synovitis with pitting (o)edema (RS3PE) is an inflammatory symmetrical polysynovitis of older people first described in 198526 which predominantly affects the wrists and flexor tendon sheaths with pitting oedema of the hands and feet (for a recent review see Olivieri et al 200027). It usually has sudden onset, but responds well to corticosteroids, and often may remit permanently within 18 months without bone erosion. It may occur as a paraneoplastic phenomenon, secondary to both solid organ and haematological malignancies, that remits after treatment of the underlying tumour. In 14 reported cases, 8 had NHL, 3 had myelodysplastic syndrome, 2 had CLL and 1 had acute myeloid leukaemia.28,29 The onset of RS3PE may either precede or follow the diagnosis of malignancy. In two of these cases, the onset of RS3PE heralded the recurrence of a malignancy that had previously been in remission. Recognition of RS3PE as a paraneoplastic syndrome may therefore be important in long-term follow-up of NHL patients. One series attempted to assess for malignancies that occurred years after the diagnosis of RS3PE, but was inconclusive.30 Similarly to other paraneoplastic phenomena, relative resistance to corticosteroids may be a clue to an underlying malignancy, and worse systemic symptoms such as anorexia, fever and weight loss also seem to be more common in patients with underlying malignancy.27

Inflammatory myositis

Perhaps the most well-known association between rheumatological and malignant disease is dermatomyositis (DM) and polymyositis (PM). There has been considerable controversy over the veracity of this well-known association, but reports from a number of national registries do support an increased risk of malignancies, which is highest for DM.31,32 In DM, one of these studies found a standardised incidence ratio (SIR) for all cancers of 3.0 (CI 2.5, 3.6). The most common cancers were ovarian, lung, pancreatic, stomach and colorectal, but the SIR for NHL and Hodgkin's disease were increased at 3.6 and 5.9 respectively. For PM, although the overall rate of cancer was lower (SIR 1.3; CI 1.0, 1.6), NHL had a stronger association than other malignancies (SIR 3.7). The temporal relationship between diagnosis of myositis and malignancy was assessed by meta-analysis.33 For DM the risk of cancer was increased for the 4 years before and after diagnosis in roughly equal numbers, but for PM the risk was only increased in the 5 years after diagnosis. The increased risk of cancer before diagnosis observed for DM supports the hypothesis that DM may be a paraneoplastic phenomenon, and is against detection bias as an explanation for the association. That this is a feature of DM and not PM may be related to differences in the aetiology of these diseases, which is thought to be humoral in DM but cell-mediated in PM. One retrospective study has examined the diagnostic yield of blind investigation for malignancy in patients presenting with inflammatory myositis (Sparsa et al 2002;34 in this unselected sample most patients had DM, and were therefore a higher risk group for malignancy).

While investigation for malignancy is commonly recommended only for patients over the age of 50, one third of the patients with malignancy in this study were under 50 years. There was an association between rapid onset of disease and malignancy and also a striking negative association of Raynaud's phenomenon with malignancy. The investigators found that some blind (without suggestive symptoms or signs) computerised tomographic scans of thorax, abdomen and/or pelvis did increase diagnostic yield of malignancy for NHL and ovarian, lung and pancreatic cancer. Although this was a small-scale retrospective study, it does emphasise the need for vigilance in any patient with inflammatory myositis.

There is a lack of robust evidence for many of the treatments used in inflammatory myositis, as highlighted by a recent Cochrane review.35 Clinical experience has supported the efficacy of corticosteroids but morbidity remains high despite treatment and a proportion of patients fail to respond at all. Azathioprine, methotrexate and ciclosporin A are frequently used in the treatment of inflammatory myositis; however there is no randomised controlled trial demonstrating their efficacy. There is evidence for efficacy of intravenous immunoglobulin in DM36 and PM37 but relapse may occur after discontinuation. Response to treatment for patients with DM with and without associated malignancy was examined by one national retrospective study.38 DM improved after treatment of underlying malignancy (solid or haematological) in 17 out of 20 patients. 5-year survival was 38% as related to cancer and 16% as related to DM. Rituximab has shown promising efficacy in some case series39,40 and there appears to be a trend towards better responses in patients with myositis-specific antibodies. This therapeutic strategy may be particularly useful in patients with inflammatory myositis associated with B-cell malignancy.

Lymphoproliferative disease and other autoimmune disorders

Lymphoproliferative diseases are associated with a range of other paraneoplastic autoimmune phenomena which may be organ-specific or multisystem. These are frequent in both T- and B-cell proliferative diseases and are associated with autoantibody production. One series examined the incidence of autoimmune disease in patients with Hodgkin's disease and NHL. In Hodgkin's disease, 7.6% of patients had an autoimmune disease, of which the most common were Sjögren's syndrome, dermatological syndromes or thyroid disease. In NHL, 8.6% of patients had an autoimmune disease, the majority being thyroid disease, with smaller numbers of patients with glomerulonephritis and thrombocytopenia. The overall rate of autoimmune disease in patients with lymphoma is therefore 2–3 times the rate in the normal population. In 70% of patients the symptoms of the autoimmune disease were present at the time of lymphoma diagnosis.41

B-cell chronic lymphocytic leukaemia (B-CLL) seems to be particularly associated with autoimmunity. This association was appreciated as early as 1959.42 The B-lymphoproliferative diseases are those that most commonly arise as a complication of the autoantibody-positive diseases Sjögren's syndrome, SLE and RA. In B-CLL, 10–26% of patients develop autoimmune phenomena, the most common by far being autoimmune hemolytic anaemia (AIHA). Other autoimmune diseases that have been associated with B-CLL are autoimmune thrombocytopenia, autoimmune neutropenia, Sjögren's syndrome, nephrotic syndrome, bullous pemphigoid, Grave's disease, SLE, RA, ulcerative colitis, allergic vasculitis and pernicious anaemia (for a recent review see Hamblin 200643). B-CLL has a highly variable prognosis, with an indolent course without treatment in some patients but a more aggressive progression and poor prognosis despite chemotherapy in others, and clinical staging systems are only partially successful in predicting this.44 The presence of AIHA is a poor prognostic factor and is more frequent in patients with higher-stage disease,45 suggesting that lymphocyte function and specificity may be important in predicting prognosis as well as measures of anatomical extent of disease. Mechanisms of autoimmunity may include production of a pathogenic autoantibody by leukaemic cells, leukaemic cells functioning as antigen-presenting cells, or induction of autoimmune haemolysis by chemotherapy.43 Treatments that have been used for autoimmune complications of CLL include splenectomy, intravenous immunoglobulin and ciclosporin. More recently, rituximab has been used. Rituximab is effective in B-lymphoproliferative diseases, and can also induce remission in a wide variety of autoimmune diseases associated with CLL including AIHA, idiopathic thrombocytopenic purpura, pure red cell aplasia, cold agglutination syndrome, axonal degenerating polyneuropathy and paraneoplastic pemphigus.43

Autoantigen recognition and B-cell chronic lymphocytic leukaemia

B-CLL is characterised by malignant clones of CD5+ B-cells, a population that produces low-affinity, polyreactive, frequently autoreactive antibody and is expanded in autoimmune disease.46 Based on these observations, the CD5+ B-cells from patients with B-CLL have been examined at a clonal level for autoantibody production.47 Immunoglobulin produced by cultures of the CD5+ clones of patients with B-CLL was tested for reactivity to single- and double-stranded DNA and for RF activity. Even using this limited panel of autoantigens, clones from more than 50% of the B-CLL patients tested produced autoreactive antibody.

If the malignant transformation of B-cells that leads to B-CLL were a stochastic process, with all B-cells having equal neoplastic potential, then we would expect recognition of autoantigens to be no more frequent in B-CLL patients than in the general population. In fact, autoantigen recognition, as manifest by autoantibody production, is extremely common. A number of different types of evidence, then, point to important relations between the development and progression of lymphoproliferative disease and autoimmunity.

Some very interesting work by Chiorazzi's group and others has investigated the clinical significance and cause of this autoreactivity (for a recent review see Chiorazzi et al 200548). Analysis of Ig variable region genes from patients with B-CLL has shown that patients can be divided into 2 groups based on the VH and VL gene mutation status. Patients with mutated V genes, characteristic of post-germinal centre memory B-cells, had the best prognosis from their lymphoma. Patients with unmutated V genes and surface markers characteristic of B-cells that have not entered a germinal centre had a much worse prognosis.49 These patients also had a two-fold increased incidence of clinical autoimmune disease, already identified as a poor prognostic factor.43 These B-cells with unmutated IgV genes may represent either naive B-cells or B-cells that have been activated without entering a germinal centre and therefore without generating mutations, but the latter explanation seems more likely when the specificity of these cells is considered. B-CLL cells have a very restricted B-cell receptor repertoire, and while most IgV-unmutated CLL cells produce highly polyreactive antibody, IgV-mutated CLL cells do not. In patients with the worst prognosis and an unmutated VH genotype there is overexpression of the VH 1-69 gene, which is also expressed in disorders with clonal RF-producing B-cells such as cryoglobulinaemia, and VH 4-34, which is expressed in cold agglutinin disease and is involved in anti-DNA antibodies. B-cell receptors from B-CLL cells often resemble autoantibodies to DNA, cardiolipin and IgG (RF) or sometimes antibodies to bacteria or viruses. The malignant transformation of B-cells that causes B-CLL therefore seems to be selective for cells of particular antigen specificity, and depends on clonal stimulation by autoantigen via T-cell independent mechanisms.48

This apparent relationship between autoimmunity and B-cell malignancy may have interesting consequences now that we are entering an era dominated by B-cell therapies. The efficacy of B-cell depletion as a therapy for autoimmune disease was first noted in patients who experienced remission from NHL and associated autoimmune disease following rituximab therapy. Therapies which treat underlying B-cell abnormality, whether by depletion or manipulation of B-cell trophic factors, may also abrogate the risk of lymphoma in these diseases. Study of the dynamics of depletion and reconstitution of B-cell subsets following rituximab therapy is an area that may have consequences for both rheumatologists and haematologists.

Conclusions

An awareness of the rheumatic presentations of haematological malignancy is important to rheumatologists, haematologists and the general physician. Recognition of certain rheumatic presentations or features may allow earlier diagnosis of haematological malignancy or relapse. In understanding of B-cell related disorders and treatments there is now convergence between haematology and rheumatology which can benefit both fields of research.

Other Titles in this series

• June 2008 No 14
  Sports-Related Musculoskeletal Disorders
• October 2007 No 13
  Glucocorticoids in Rheumatic Diseases
• June 2007 No 12
  Immunisation of Immunosuppressed Patients with Rheumatic Diseases
• October 2006 No 10
  Work-Related Disorders of the Upper Limb
• June 2006 No 9
  Qualitative Research: Its Relevance and Use in Musculoskeletal Medicine
• February 2006 No 8
  Rheumatic Diseases: Serological Aids to Early Diagnosis
• October 2005 No 7
  Malignancy and the Rheumatic Diseases
• June 2005 No 6
  Systemic Amyloidosis in the Rheumatic Diseases
• February 2005 No 5
  Heritable Collagen Disorders
• October 2004 No 4
  An Update on Gout
• June 2004 No 3
  Surgery for Arthritis: Total Hip and Knee Joint Replacement
• February 2004 No 2
  Idiopathic Inflammatory Myopathies
• September 2003 No 1
  Low Back Pain

download a pdf of this publication

6631.pdf
(139Kb)