Giant Cell Arteritis in a Patient Taking Etanercept and Methotrexate
To the Editor:
I describe the occurrence of giant cell arteritis (GCA) in an elderly woman with long-standing seropositive rheumatoid arthritis (RA), treated for 2 years with etanercept 25 mg twice weekly and methotrexate 7.5 mg weekly. Since etanercept and infliximab were approved for treatment of RA, there has been one report of noncaseating granulomas in pulmonary parenchyma1 and vasculitis with increased nodulosis2, and several reports of reactivation of latent tuberculosis in patients taking infliximab or etanercept3. These and other complications of tumor necrosis factor-a (TNF-a) receptor-blocking therapy involving activated endothelial cells and granuloma formation suggest a common mechanism of inflammation in susceptible patients.
In summer 2002, a 79-year-old Caucasian woman with RA presented with left jaw pain, so severe that she would not open her mouth to eat. There was associated ear pain and a sore throat. Evaluation by our dental service showed no lesion. The next week when she came into clinic for her injections, she reported worsening pain over the left temporal area. There was no antecedent trauma, and no history of temporomandibular joint involvement in her disease or previous complaints of pain at this site. Present on examination was a thin-walled, tender left temporal artery, with no erythema. There was no evidence on examination of other systemic vasculitis. A biopsy of the left temporal artery showed GCA (Figure 1). Laboratory studies were remarkable for a rising erythrocyte sedimentation rate, 46 to 86, and negative antinuclear antibodies, and she was hepatitis C antibody negative. No HLA testing was done.
Figure 1. Left temporal artery biopsy revealed GCA in a 79-year-old woman with RA (H & E, ´10).
GCA is commonly reported in association with polymyalgia rheumatica, but it is unusual to see GCA occurring in patients with RA or at least, it is reported infrequently in the literature. In 1983, 2 groups published cases describing its concurrence, and there has been speculation over the years that RA and GCA may share common pathogenetic mechanisms4-6. Studies in GCA have described TNF I (p55) receptor expression in the intima of the large arteries, with immunohistochemical staining revealing TNF-a on endothelial cells, macrophages, and giant cells7.
Etanercept is a soluble TNF receptor, a fusion protein created from human TNF p75 receptor and the Fc portion of human immunoglobulin G1. Binding to soluble TNF-a, this drug effectively blocks receptor activation in RA, resulting in clinical improvement in many patients. In 2002, Cunnane and colleagues showed images of an inflammatory cell infiltrate in a medium-size arterial wall associated with accelerated nodulosis in patients with RA treated with etanercept2. Clinical trials in Crohn's disease suggest etanercept is not effective in Th1 mediated inflammatory bowel disease, although infliximab has some record of success8. The reason for this discrepancy is unclear. Clinical studies have been undertaken to examine the efficacy of etanercept in Wegener's granulomatosis. The responses here, too, have been variable, suggesting once again that etanercept may not protect against granulomatous vasculitis9. In 2003, Tan, et al published a letter describing effective treatment of "resistant" GCA with etanercept in an elderly man with clinical polymyalgia rheumatica and biopsy negative temporal arteritis10.
The experience with our patient raises the question whether individuals undergoing TNF receptor blockade might be more vulnerable to granulomatous vasculitis, or to other infectious complications requiring effective granuloma formation and intracellular killing. Moreland, et al recently reported there is no evident impairment of delayed hypersensitivity reactions in patients taking etanercept, that immunoglobulin concentrations are normal, and there was no excessive risk of infection in the people studied11. Clearly, the complications of TNF receptor blockade reported in the literature suggest that at a cellular level, in an aging artery, arteriole, or latent tubercular granuloma, TNF receptor expression and signaling are altered in some way, in some individuals. As clinical trials using anti-TNF agents in granulomatous vasculitis are under way, it is important that clinicians realize that these agents may not be effective in these diseases, and may not protect against them.
MARGARET SETON, MD, Chief of Rheumatology, Cambridge Health Alliance, Assistant Professor of Medicine, Harvard University, Staff Physician, Arthritis Unit, Bulfinch 165, Massachusetts General Hospital, Boston, Massachusetts 02114, USA. E-mail: email@example.com
1. Peno-Green L, Lluberas G, Kingsley T, Brantley S. Lung injury linked to etanercept therapy. Chest 2002;122:1858-60.
2. Cunnane G, Warnock M, Fye KH, Daikh DI. Accelerated nodulosis and vasculitis following etanercept therapy for rheumatoid arthritis. Arthritis Rheum 2002;47:445-9.
3. Keane J, Gershon S, Wise RP, et al. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. New Engl J Med 2001;345:1098-104.
4. Horslev-Petersen K, Helin P. Coexistence of temporal arteritis and rheumatoid arthritis. J Rheumatol 1983;10:831-3.
5. Hall S, Ginsburg WW, Vollertsen RS, Hunder GG. The coexistence of rheumatoid arthritis and giant cell arteritis. J Rheumatol 1983;10:995-7.
6. Salvarani C, Cantini F, Boiardi L, Hunder G. Polymyalgia rheumatica and giant-cell arteritis. New Engl J Med 2002;347:261-71.
7. Cook A, Gallagher G, Field M. Immuno-localisation of tumour necrosis factor and its receptors in temporal arteritis. Rheumatol Int 1997;17:113-8.
8. Podolsky DK. Inflammatory bowel disease. New Engl J Med 2002;347:417-29.
9. Stone J, Uhlfelder M, Hellmann D, Crook S, Bedocs N, Hoffman G. Etanercept combined with conventional treatment in Wegener's granulomatosis: a six-month open-label trial to evaluate safety. Arthritis Rheum 2001;44:1149-54.
10. Tan AL, Holdsworth J, Pease C, Emery P, McGonagle D. Successful treatment of resistant giant cell arteritis with etanercept [letter]. Ann Rheum Dis 2003;62:373-4.
11. Moreland LW, Bucy RP, Weinblatt ME, Mohler KM, Spencer-Green GT, Chatham WW. Immune function in patients with rheumatoid arthritis treated with etanercept. Clin Immunol 2002;103:13-21.
Is Measurement of Serum Vascular Endothelial Growth Factor Reliable in Patients with Systemic Sclerosis?
To the Editor:
I read with interest the article by Choi, et al1 reporting serum concentrations of vascular endothelial growth factor (VEGF) in patients with systemic sclerosis (SSc). I bring to your attention some methodological concerns that have arisen.
First, serum VEGF levels do not reflect VEGF synthesis by peripheral tissues, and they are not representative of circulating extracellular VEGF level at the time of sampling. In serum, VEGF levels are several-fold higher than in matched plasma samples, owing to in vitro release of VEGF from platelets during blood clotting2,3. Although various blood cells such as granulocytes, monocytes, mast cells, and lymphocytes have been shown to be capable of producing VEGF, these cells are of little importance for the release of VEGF into the circulation4-6. Contradictory results on platelet aggregation in patients with SSc have been reported. In vitro studies suggest that platelets from patients with SSc are hyperactive7,8; these observations may justify the increased levels of serum VEGF measured by the authors in patients with SSc compared with healthy controls. CTAD (citrate, theophylline, adenosine, dipyridamole) plasma is recommended for the measurement of circulating extracellular VEGF9.
Second, the authors did not report the condition of processing (i.e., force of centrifugation, time and temperature between blood collection and processing). When VEGF levels are measured, standardization in the collection of serum is relevant and it should be declared. VEGF is released from the platelet in serum in a time-dependent manner. Allowing whole blood sample to clot 2 to 6 hours before serum is collected reduces time-dependent, non-uniform release of VEGF6. In addition, centrifuging samples at variable centrifugal forces or for variable times can affect platelet activation by mechanical stress, and consequently may influence VEGF levels. The authors suggest that "high VEGF levels may serve as a surrogate indicator of capillary damage in SSc." In a clinical situation, where blood samples are taken and left for variable times before processing, the contribution from the clotting process would effectively rule out the use of serum measurement. However, even if strict uniformity of clotting time is applied to all samples, the large interpersonal variation in generation of VEGF in clotted samples may invalidate the results2.
Third, platelet count is significantly correlated to VEGF serum level6. When VEGF levels are measured from serum samples, it is advisable to correct the measurement to platelet count10.
In light of these considerations, I believe the results reported by the authors should be confirmed on plasma samples. The use of plasma samples may improve the value of circulating extracellular VEGF as an indicator of capillary damage in patients with SSc.
SIMONE FERRERO, MD, Department of Obstetrics and Gynaecology, San Martino Hospital, University of Genoa, Largo R. Benzi 1, 16132 Genoa, Italy. E-mail: firstname.lastname@example.org
1. Choi JJ, Min DJ, Cho ML, et al. Elevated vascular endothelial growth factor in systemic sclerosis. J Rheumatol 2003;30:1529-33.
2. Webb NJ, Bottomley MJ, Watson CJ, Brenchley PE. Vascular endothelial growth factor (VEGF) is released from platelets during blood clotting: implications for measurement of circulating VEGF levels in clinical disease. Clin Sci Lond 1998;94:395-404.
3. Gunsilius E, Petzer A, Stockhammer G, et al. Thrombocytes are the major source for soluble vascular endothelial growth factor in peripheral blood. Oncology 2000;58:169-74.
4. Grutzkau A, Kruger-Krasagakes S, Baumeister H, et al. Synthesis, storage, and release of vascular endothelial growth factor/vascular permeability factor (VEGF/VPF) by human mast cells: implications for the biological significance of VEGF206. Mol Biol Cell 1998;9:875-84.
5. Harmey JH, Dimitriadis E, Kay E, Redmond HP, Bouchier-Hayes D. Regulation of macrophage production of vascular endothelial growth factor (VEGF) by hypoxia and transforming growth factor beta-1. Ann Surg Oncol 1998;5:271-8.
6. Werther K, Christensen IJ, Nielsen HJ. Determination of vascular endothelial growth factor (VEGF) in circulating blood: significance of VEGF in various leucocytes and platelets. Scand J Clin Lab Invest 2002;62:343-50.
7. Friedhoff LT, Seibold JR, Kim HC, Simester KS. Serotonin induced platelet aggregation in systemic sclerosis. Clin Exp Rheumatol 1984;2:119-23.
8. Goodfield MJ, Orchard MA, Rowell NR. Whole blood platelet aggregation and coagulation factors in patients with systemic sclerosis. Br J Haematol 1993;84:675-80.
9. Wynendaele W, Derua R, Hoylaerts MF, et al. Vascular endothelial growth factor measured in platelet poor plasma allows optimal separation between cancer patients and volunteers: a key to study an angiogenic marker in vivo? Ann Oncol 1999;10:965-71.
10. Hormbrey E, Gillespie P, Turner K, et al. A critical review of vascular endothelial growth factor analysis in peripheral blood: is the current literature meaningful? Clin Exp Metastasis 2002;19:651-63.
Autoantibodies to Bactericidal/Permeability-Increasing Protein and Cathepsin G in Systemic Sclerosis
To the Editor:
In a recent article, Khanna, et al1 suggest that bactericidal/permeability-increasing protein (BPI) and cathepsin G are the major antigenic targets of antineutrophil cytoplasmic antibodies (ANCA) in systemic sclerosis (SSc) or scleroderma. We investigated the prevalence of antibodies to BPI, cathepsin G, myeloperoxidase (MPO), proteinase 3 (PR3), lactoferrin, lysozyme, and elastase using a commercial Combi ELISA kit (Diamedix, Miami, FL, USA). Sera of 20 patients with SSc (7 diffuse form, 13 limited form) were examined. In agreement with the findings of Khanna, et al1, we observed that the prevalences of antibodies to BPI (9/20, 45%) and cathepsin G (11/20, 55%) were higher than those of antibodies to MPO (1/20, 5%), PR3 (4/20, 20%), lactoferrin (4/20, 20%), lysozyme (1/20, 5%), and elastase (2/20, 10%). Similarly, in our sera BPI and cathepsin G antibodies were frequently associated (8/9 and 8/11, respectively).
Recently, performing indirect immunofluorescence (IIF) studies on 115 sera from patients with diffuse (n = 55) and the limited form (n = 60) of SSc we found ANCA positivity in 23 cases (20%)2,3. In accord with Khanna, et al, the only fluorescent patterns were the atypical (15.6% of cases) and perinuclear (4.3%) stainings, while cytoplasmic fluorescence was never found. However, our IIF results were obtained employing ethanol and formalin-fixed human neutrophils (Inova Diagnostics, San Diego, CA, USA) as substrate4, while they used only ethanol-fixed human neutrophils.
When we compared2,5 the clinical and serological features of cathepsin G, MPO, and PR3 ANCA-positive patients with those of cathepsin G, MPO, and PR3 ANCA-negative patients, we found no correlation. In particular, according to the reports cited2,5, statistical analysis showed no association between cathepsin G positivity and clinical/serological features nor between MPO positivity and kidney involvement.
We confirm the high prevalence of anti-BPI and cathepsin G antibodies and their frequent association in SSc sera. The significance of their presence and that of other ANCA in scleroderma remains unknown.
PANAGIOTIS GRYPIOTIS, PhD, Research Biologist; FRANCO COZZI, MD, Assistant Professor of Rheumatology; AMELIA RUFFATTI, MD, Associate Professor of Rheumatology, Department of Medical and Surgical Sciences, Rheumatology Unit, University of Padova, Policlinico Universitario, Via Giustiniani 2, 35128 Padova, Italy. E-mail: email@example.com
1. Khanna D, Aggarwal A, Bhakuni DS, Dayal R, Misra R. Bactericidal/permeability increasing protein and cathepsin G are the major antigenic targets of antineutrophil cytoplasmic autoantibodies in systemic sclerosis. J Rheumatol 2003;30:1248-52.
2. Grypiotis P, Ruffatti A, Cozzi F, et al. Prevalence and clinical significance of cathepsin G antibodies in systemic sclerosis [Italian]. Reumatismo 2003;55:256-62.
3. Ruffatti A, Grypiotis P, Todesco S. Antineutrophil cytoplasmic antibodies in patients with systemic sclerosis [reply to letter]. J Rheumatol 2003;30:2079-80.
4. Radice A, Vecchi M, Bianchi MB, Sinico RA. Contribution of immunofluorescence to the identification and characterization of anti-neutrophil cytoplasmic autoantibodies. The role of different fixatives. Clin Exp Rheumatol 2000;18:707-12.
5. Ruffatti A, Sinico RA, Radice A, et al. Autoantibodies to proteinase 3 and myeloperoxidase in systemic sclerosis. J Rheumatol 2002;29:918-23.
Gastrointestinal Disease and Psoriatic Arthritis
To the Editor:
Observations suggest that a relationship exists between gastrointestinal (GI) disease and psoriatic arthritis (PsA). A subgroup of patients with PsA has large joint oligoarthritis that is clinically similar to reactive arthritis and the arthritis associated with inflammatory bowel disease (IBD). In addition, microscopic inflammatory changes in the bowel mucosa of patients with active psoriasis and PsA are common1,2. These findings have led to the suggestion that GI disease, as well as skin disease, may act as a portal of entry of causative antigens in PsA2. Despite these observations, the background frequency of GI disease in patients with PsA has not been well documented. This issue has growing relevance given that the treatment of PsA increasingly involves the use of disease modifying drugs that may adversely affect the GI system.
We studied the prevalence of GI disease in a group of patients with PsA.
One hundred three unselected patients with PsA were recruited from general rheumatology outpatient clinics in Oxford, UK. Ethical approval was obtained from the Central Oxford Research Ethics Committee. Patients were considered to have PsA if they had seronegative inflammatory arthritis and psoriasis3. The clinical characteristics of these patients have been reported4. A detailed history was taken for the presence of GI disease including a diagnosis of IBD and irritable bowel syndrome (IBS). Patients were also specifically questioned about symptoms of gluten sensitivity. To further screen for celiac disease, IgA antiendomysial antibodies were tested by direct immunofluorescence in 96 of the patients. The results of antiendomysial antibody testing have been reported5, and are included here for completeness. HLA-B27 was tested in 98 patients by polymerase chain reaction with sequence-specific primers. Data were analyzed using contingency tables. The frequency of IBD and IBS in the study patients was compared with that of historical UK controls6,7. P values are expressed as 2-tailed values.
Four of 103 (3.9%) patients had biopsy-proven IBD, compared with the reported prevalence of IBD in the general UK population of 0.4% (p < 0.001). There were 3 patients with ulcerative colitis and one with Crohn's disease. All patients with IBD had psoriatic nail and scalp disease. Three had polyarticular disease and one had oligoarticular disease. One patient had significant axial disease, with sacroiliitis on magnetic resonance imaging (but was HLA-B27 negative). HLA-B27 results were available for 3 patients with IBD; all were HLA-B27 negative.
No patient had GI malignancy or celiac disease. Antiendomysial antibodies were negative in all patients. The prevalence of IBS (18%) was comparable to that of the general population.
The most striking finding was the high prevalence of IBD in our patients with PsA: almost 10-fold greater than the general UK population. While this is not an age or sex matched comparison, we believe these results are clinically meaningful. Sampling bias may also lead to an overestimation of the prevalence of IBD in our PsA population. Importantly, our patients were not recruited from an institution with a specialized gastroenterology unit. Clearly, it is impossible to distinguish with complete certainty between patients with PsA and IBD, and patients with enteropathic arthritis and psoriasis. However, all patients recruited into this study met the Moll and Wright criteria for PsA, and were considered to have typical PsA by their usual rheumatologist as well as by the investigators.
Although an increased risk of IBD in patients with PsA has not been widely reported, studies of patients with IBD have shown a high prevalence of psoriasis, ranging from 5.7% to 11.2%8,9. Overall, these data support the hypothesis that common genetic or environmental determinants for IBD, psoriasis, and PsA exist. It is unlikely that HLA-B27 is a clinically significant factor in our patients. Rahman and colleagues have recently reported that the CARD15 (NOD2) Crohn's disease susceptibility gene also confers susceptibility to PsA independent of HLA-Cw*060210. It would be interesting to study a larger group of patients with IBD and PsA to determine whether shared genetic factors such as CARD15 gene variants account for this association.
We wish to raise awareness of the association between PsA and IBD. Our data suggest that clinicians should have a high index of suspicion for IBD in their patients with PsA.
LYN WILLIAMSON, MA, MRCP (Oxon), Consultant Rheumatologist; JOANNE L. DOCKERTY, MB ChB, Rheumatology Research Fellow; NICOLA DALBETH, FRACP, Rheumatology Research Fellow; B. PAUL WORDSWORTH, FRCP, Consultant Rheumatologist, Department of Rheumatology, Nuffield Orthopaedic Centre, Headington, Oxford OX3 7LD, United Kingdom. Email: firstname.lastname@example.org
Supported by an Oxford District Research Grant. Dr. Dalbeth is the Rose Hellaby Fellow 2002.
1. Scarpa R, Manguso F, D'Arienzo A, et al. Microscopic inflammatory changes in colon of patients with both active psoriasis and psoriatic arthritis without bowel symptoms. J Rheumatol 2000;27:1241-6.
2. Schatteman L, Mielants H, Veys EM, et al. Gut inflammation in psoriatic arthritis: a prospective ileocolonoscopic study. J Rheumatol 1995;22:680-3.
3. Moll JM, Wright V. Psoriatic arthritis. Semin Arthritis Rheum 1973;3:55-78.
4. Williamson L, Dockerty JL, Dalbeth N, McNally E, Ostlere S, Wordsworth BP. Clinical assessment of sacroiliitis and HLA-B27 are poor predictors of sacroiliitis diagnosed by magnetic resonance imaging in psoriatic arthritis. Rheumatology Oxford 2004;43:85-8.
5. Dockerty JL, Williamson L, Wordsworth BP. Endomysial antibodies in psoriatic arthritis patients. Rheumatology Oxford 2002;41:1195-6.
6. Rubin GP, Hungin AP, Kelly PJ, Ling J. Inflammatory bowel disease: epidemiology and management in an English general practice population. Aliment Pharmacol Ther 2000;14:1553-9.
7. Kennedy TM, Jones RH, Hungin AP, O'Flanagan H, Kelly P. Irritable bowel syndrome, gastro-oesophageal reflux, and bronchial hyper-responsiveness in the general population. Gut 1998;43:770-4.
8. Yates VM, Watkinson G, Kelman A. Further evidence for an association between psoriasis, Crohn's disease and ulcerative colitis. Br J Dermatol 1982;106:323-30.
9. Lee FI, Bellary SV, Francis C. Increased occurrence of psoriasis in patients with Crohn's disease and their relatives. Am J Gastroenterol 1990;85:962-3.
10. Rahman P, Bartlett S, Siannis F, et al. CARD15: a pleiotropic autoimmune gene that confers susceptibility to psoriatic arthritis. Am J Hum Genet 2003;73:677-81.