rt404 2 rheum 6.2 05 - uk healthcare cecentral · raimon sanmartí, maria victoria hernández,...

VOLUME 6 NUMBER 2 2008 WWW.ADVANCESINRHEUMATOLOGY.COM

INTERNATIONAL JOURNAL OF ADVANCES IN

RHEUMATOLOGYEditors-in-ChiefMichael E Weinblatt, Boston, MA, USAFerdinand C Breedveld, Leiden, The Netherlands

Pharmacogenetics of Tumor Necrosis FactorAntagonists in Rheumatoid ArthritisHamid Bashir and Prabha Ranganathan

Cardiovascular Aspects of Rheumatoid ArthritisMichael T Nurmohamed

Serum Autoantibodies in Rheumatoid ArthritisRaimon Sanmartí, Maria Victoria Hernández, José A Gómez-Puerta, Eduard Graell, and Juan D Cañete

Case Study: Just Another Case of NeuropsychiatricSystemic Lupus Erythematosus?Leroy R Lard, Darius Soonawala, Abraham Schoe, and Irene Speyer

Jointly sponsored by the University of Kentucky Colleges of Pharmacy and Medicine and Remedica.

The University of Kentucky is an equal opportunity university.This journal is supported by aneducational grant from Abbott.

RT404_2_Rheum_6_2_COV_US_02.qxd 8/29/08 5:11 PM

International Journal of Advances in Rheumatology is supported by an unrestrictededucational grant from Abbott Immunology

The International Journal of Advances in Rheumatology is currently distributed to approximately 14 000 rheumatologists in 18 countries. You can visit the journal online at:

www.advancesinrheumatology.com

We would like to thank all those readers who have already taken time to provide feedback on theInternational Journal of Advances in Rheumatology. We are delighted that the comments have been

overwhelmingly positive and that the journal continues to be regarded as a useful resource byrheumatologists working in this fast-developing field

Please help us to keep improving our journal by providing feedback via the websitewww.advancesinrheumatology.com/feedback

Faculty DisclosuresThe following are relevant financial relationships declared by the journal’s Editors-in-Chief, Editors, and Editorial Board members. Ferdinand C Breedveld: Abbott, Schering-Plough, Wyeth. Michael E Weinblatt: Abbott, Alza, Amgen, AstraZeneca, BioAssets, Biogen Idec, Bristol-Myers Squibb, CanFite, Celgene, Centocor, CombinatoRx, CriticalTherapeutics, Eisai, EntreMed, Genentech, Gilead, Lilly, Merrimack, Merck, Millennium, Novartis, Pfizer, Praecis, Proprius, Rigel, Roche, Sanofi-Aventis, Scios, Serono, Synta,TAP, Trubion, Wyeth, UCB, Vascular Biogenics. Tom Huizinga: None declared. Peter Nigrovic: None declared. Eric Ruderman: Abbott, Amgen, Biogen Idec, Bristol-MyersSquibb, Genentech, Roche. Hendrik Schulze-Koops: Abbott, Aspreva, Essex, Roche, Wyeth. Cornelia F Allaart: None declared. Steven Abramson: Amgen, GlaxoSmithKline,Novartis, Pfizer. Carol Black: None declared. Gerd Burmester: Abbott, Essex, Merck, Novartis, Pfizer, Roche, Sanofi-Aventis, UCB, Wyeth. Maxime Dougados: Abbott, Bristol-Myers Squibb, Roche, Wyeth. Paul Emery: Abbott, Amgen, Bristol-Myers Squibb, Centocor, Roche, Schering-Plough, Wyeth. Daniel Furst: Abbott, Amgen, Bristol-MyersSquibb, Centocor, Genentech, Merck, Novartis, Roche, UCB. Mark Genovese: Biogen Idec, Bristol-Myers Squibb, Genentech, Roche, Serono, Wyeth. Gabriel Herrero-Beaumont: None declared. Joachim Kalden: Abbott, Roche, Schering-Plough, Wyeth. Arthur Kavanaugh: Abbott, Amgen, Biogen Idec, Bristol-Myers Squibb, Centocor,Genentech. Edward Keystone: Abbott, Amgen, AstraZeneca, Bristol-Myers Squibb, Novartis, Pfizer, Roche, Schering-Plough, Wyeth. Lars Klareskog: Abbott, Astra-Zaneca,Bristol-Myers Squibb, Roche, Schering-Plough, Wyeth. Vicente Rodriguez-Valverde: Abbott, Bristol-Myers Squibb, Schering-Plough. Josef Smolen: Abbott, Bristol-MyersSquibb, Centocor, Roche, Sanofi-Aventis, Schering-Plough, Wyeth. Désirée van der Heijde: Abbott, Amgen, Bristol-Myers Squibb, Centocor, Chugai, Dainippon Sumitomo,GlaxoSmithKline, Roche, Schering-Plough, UCB, Wyeth.

Editorial Policy International Journal of Advances in Rheumatology is an independent journal published by Remedica Medical Education and Publishing. Editorial control is the soleresponsibility of the Editors-in-Chief, Editorial Advisory Board, and the Editors. Before publication, all material submitted to the journal is subjected to rigorous review by theEditors-in-Chief, Editorial Advisory Board, Editors, and/or independent reviewers for suitability of scientific content, scientific accuracy, scientific quality, and conflict of interest.

Aims and Scope International Journal of Advances in Rheumatology is designed to bring a critical analysis of the world rheumatology literature, written by clinicians, for clinicians, to aninternational, multidisciplinary audience. Our mission is to promote better understanding of rheumatological medicine across the global healthcare system by providing anactive forum for the discussion of clinical and healthcare policy issues.

Leading Articles - These major review articles are chosen to reflect topical clinical and healthcare issues in rheumatology. All contributions undergo a strict editorial review process.

Clinical Reviews - The most important papers from the best of the international literature on rheumatology are systematically selected by an internationally recognized panel ofexperts. The Editors then prepare concise and critical analyses of each paper, and, most importantly, place the findings into clinical context.

Meeting Reports - International Journal of Advances in Rheumatology also provides incisive reportage from the most important international congresses.

Publisher’s Statement ©2008 Remedica Medical Education and Publishing. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form orby any means, electronic, mechanical, photocopying, recording, or otherwise without the prior permission of the copyright owners. While every effort is made by the publishersand editorial board to see that no inaccurate or misleading data, opinions, or statements appear in this journal, they wish to make it clear that the material contained in thepublication represents a summary of the independent evaluations and opinions of the authors and contributors. As a consequence, the board, publishers, and any supportingcompany accept no responsibility for the consequences of any such inaccurate or misleading data or statements. Neither do they endorse the content of the publication or theuse of any drug or device in a way that lies outside its current licensed application in any territory. International Journal of Advances in Rheumatology (ISSN 1478-856X) ispublished four times a year by Remedica Publishing Ltd and distributed by USA Mail Agent CMM LLC, 147–29 182nd Street, Jamaica, NY 11413. Subscription price $170 peryear. Periodicals Postage Pending at Jamaica, NY. POSTMASTER: Please send address changes to Remedica Publishing Ltd, c/o CMM 147–29 182nd Street, Jamaica, NY 11413.

Remedica Medical Education and Publishing Ltd., Commonwealth House, 1 New Oxford Street, London, WC1A INU, UK.Tel: +44 (0)20 7759 2999 Fax: +44 (0)20 7759 2951 Email: [email protected]

Remedica Medical Education and Publishing Inc., 20 North Wacker Drive, Suite 1642, Chicago, IL 60606. USATel: +1 (312) 372 4020 Fax: +1 (312) 372 0217 Email: [email protected]

Editorial Team: Rhian Phillips, Nicola Seymour Editorial Manager: Scott Millar Publishers: Ian Ackland-Snow, Simon Kirsch Design and Artwork: AS&K Skylight Creative Services

ISSN 1478-856X

RT404_2_Rheum_6_2_COV_US_02.qxd 8/29/08 5:11 PM

Contents

Leading ArticlesPharmacogenetics of Tumor Necrosis FactorAntagonists in Rheumatoid Arthritis: An Update Hamid Bashir and Prabha Ranganathan 34

Cardiovascular Aspects of Rheumatoid ArthritisMichael T Nurmohamed 40

Serum Autoantibodies in Rheumatoid ArthritisRaimon Sanmartí, Maria Victoria Hernández,José A Gómez-Puerta, Eduard Graell,and Juan D Cañete 47

Case StudyJust Another Case of Neuropsychiatric Systemic Lupus Erythematosus?Leroy R Lard, Darius Soonawala, Abraham Schoe, and Irene Speyer 53

Clinical ReviewsTreatment Strategies 57

Genetics 60

Pathogenesis 61

Infectious and Malignant Complications 63

Cardiovascular Risk 65

Prognosis and Assessment 66

Miscellaneous 68

Meeting ReportEULAR 2008 70

Editors-in-ChiefFerdinand C Breedveld, MDProfessor of Rheumatology, Department of Rheumatology, Leiden UniversityMedical Center, Leiden, The Netherlands

Michael E Weinblatt, MDProfessor of Medicine, Harvard Medical School, Division of Rheumatology,Immunology and Allergy, Brigham and Women’s Hospital, Boston, MA, USA

EditorsTom WJ Huizinga, MDProfessor, Department of Rheumatology, Leiden University Medical Center,Leiden, The NetherlandsPeter Nigrovic, MDDirector, Center for Adults with Pediatric Rheumatic Illness, Division ofRheumatology, Immunology and Allergy, Brigham and Women's HospitalBoston, MA, USAEric M Ruderman, MDAssociate Professor, Division of Rheumatology, Northwestern UniversitySchool of Medicine, Chicago, IL, USAHendrik Schulze-Koops, MD, PhDProfessor, Division of Rheumatology, University of Munich, Munich, Germany

Case Study EditorCornelia F Allaart, MD, PhDAssociate Professor, Department of Rheumatology, Leiden University MedicalCenter, Leiden, The Netherlands

Editorial Advisory BoardSteven B Abramson, MDProfessor of Medicine and Pathology, New York University School of Medicine,New York, NY, USACarol M Black, MDProfessor of Rheumatology, Department of Rheumatology, Royal Free Hospital,London, UKGerd R Burmester, MDProfessor of Medicine, Department of Rheumatology and Clinical Immunology,Charite University Hospital, Humboldt-University of Berlin, Berlin, GermanyMaxime Dougados, MDProfessor of Rheumatology, Department of Rheumatology, Hospital Cochin,Paris, FrancePaul Emery, MDACR Professor of Rheumatology, Academic Department of MusculoskeletalDisease, Department of Rheumatology, Leeds General Infirmary, Leeds, UKDaniel E Furst, MDCarl M Pearson Professor of Rheumatology, UCLA Medical School, Los Angeles, CA, USAMark C Genovese, MDAssistant Professor of Medicine, Division of Rheumatology, StanfordUniversity Medical Center, Palo Alto, CA, USAGabriel Herrero-Beaumont, MDProfessor of Rheumatology, Inflammation Research Unit, Fundacion JimenezDiaz, Universidad Autonoma, Madrid, SpainJoachim R Kalden, MDProfessor of Internal Medicine, University of Erlangen-Nuremberg, Erlangen, GermanyArthur F Kavanaugh, MDProfessor of Medicine, Division of Rheumatology, Allergy and Immunology,Center for Innovative Therapy, UCSD, La Jolla, CA, USA Edward Keystone, MDProfessor of Medicine, University of Toronto, Director, Center for AdvancedTherapeutics in Arthritis, Mount Sinai Hospital, Toronto, ON, CanadaLars Klareskog, MDProfessor of Rheumatology, Department of Medicine, Karolinska Hospital,Karolinska Institute, Stockholm, SwedenVicente Rodriguez-Valverde, MDProfessor of Medicine and Chief, Rheumatology Service, HospitalUniversitario Marqués de Valdecilla, Facultad de Medicina, Universidad de Cantabria, Santander, SpainJosef S Smolen, MDProfessor, 2nd Department of Medicine, Krankenhaus Lainz, Vienna, AustriaDésirée van der Heijde, MDProfessor of Rheumatology, Department of Rheumatology, Leiden University Medical Centre, Leiden, The Netherlands

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 1

Rheumatoid arthritis (RA) is a chronic, systemic, inflammatorydisease of unknown etiology that is associated withprogressive disability and an increased mortality rate. Theprimary site of pathology is the synovium of the joints, whichbecomes inflamed and proliferates to form a pannus [1].Many cell populations, including monocytes, macrophages, B cells, T cells, endothelial cells, and fibroblasts, participate inthe ongoing inflammatory process [2]. B cell–T cellinteractions are integral to the inflammatory process in RA; Bcells may function as antigen presenting cells and provideimportant co-stimulatory signals required for CD4+ T cellclonal expansion and effector functions [3,4]. RA is thoughtto be initiated by T lymphocytes recognizing antigens in thesynovial tissue. Activated T cells, macrophages, andfibroblasts produce pro-inflammatory cytokines that play akey role in the synovitis and tissue destruction in RA. Tumornecrosis factor-α (TNF-α) and interleukin-1 (IL-1) are the twomajor pro-inflammatory cytokines that enhance synovialproliferation and stimulate secretion of other inflammatorycytokines, matrix-degrading metalloproteinases, adhesionmolecules, and prostaglandin E2 – all of which areinstrumental in tissue destruction in RA [5]. Recognition of

the crucial roles of these cytokines in inflammation in RA hasled to the development of targeted biological therapies, suchas agents that block the action of TNF-α. These biologicaltreatments have revolutionized the treatment of RA.Randomized clinical trials have shown that these agents yieldsignificant clinical benefit in patients with inadequateresponses to other disease-modifying drugs [6,7].

Despite the widespread use of these novel therapeuticregimens, only 50–70% of patients receiving anti-TNFtherapy achieve at least an ACR20 response during clinicaltrials, suggesting that there remains a significant proportionof patients that does not respond [8]. As these drugs areexpensive and have the potential to cause serious toxicity[9–12], identifying patients most likely to respond favorablyis desirable. There has been a growing interest in recentyears in pharmacogenetic approaches to identifying suchpatients. These approaches are particularly relevant in RAwhere long-term therapy is required and several therapeuticoptions exist without a universal drug of choice.

Inherited differences in drug effects and drug metabolismwere first documented in the 1950s [13,14]. These conceptsevolved into a field of research called “pharmacogenetics”,which focuses on the interactions between genetic variationsand drug effects in individuals. Pharmacogenetics has beenrediscovered by a broader spectrum of academia and industry,creating the term “pharmacogenomics”. This term applies togenome-wide approaches (rather than single or multiple genes

LEA

DIN

G A

RTIC

LE

Pharmacogenetics of Tumor NecrosisFactor Antagonists in Rheumatoid Arthritis:An Update

Hamid Bashir, MD, and Prabha Ranganathan, MDDepartment of Medicine, Washington University School of Medicine, St Louis, MO, USA

Rheumatoid arthritis (RA) is a chronic inflammatory disease with a worldwide prevalence of approximately 1%. There isgood agreement on early intervention and tight control of the disease, and prevention of irreversible joint damage.However, despite the introduction of targeted biological therapy, 25–30% of RA patients do not respond to treatment.Individualizing treatment decisions can be problematic because of a lack of predictability of patients’ response to therapy.The results of recent investigations indicate that differential drug responses may be explained by genomic variationsbetween individual patients; this represents a rapidly evolving field called “pharmacogenomics”. Incorporation of suchgenomic data in clinical practice may lead to individualized treatment of RA with more efficient use of the tumor necrosisfactor (TNF) antagonists. This review highlights recent literature on the pharmacogenetics of TNF blockers in RA and theissues that surround pharmacogenomic applications in the future. Int J Adv Rheumatol 2008;6(2):34–9.

INTERNATIONAL JOURNAL OF ADVANCES IN RHEUMATOLOGY Vol 6 No 2 200834

Address for correspondence: Prabha Ranganathan, Division of

Rheumatology, Washington University School of Medicine, 660 South

Euclid Avenue, Campus Box 8045, St Louis, MO 63110, USA.

Email: [email protected]

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 34

PHARMACOGENETICS OF TNF ANTAGONISTS IN RA

INTERNATIONAL JOURNAL OF ADVANCES IN RHEUMATOLOGY Vol 6 No 2 2008 35

of interest) that are used to identify genetic variations thatgovern response to medications [15]. Unlike other factors thatinfluence drug response, inherited determinants offer theadvantage of remaining stable throughout a person’s lifetime[16]. Inter-individual differences in drug response may becaused by common genetic variations (called polymorphisms)in genes encoding drug-metabolizing enzymes, drugtransporters, or drug targets. These genetic variants of interestcome in many forms; single nucleotide polymorphisms (SNPs)are the most common, with possibly 15 million in the humangenome. SNPs are caused by a difference in one base-pair inthe DNA sequence, which may result in a change in thefunction or quantity of the transcribed protein depending onthe nucleotide change and the location [15]. There are >30 families of drug-metabolizing enzymes in humans, andessentially all are encoded by genes with genetic variants,many of which translate into functional changes in the proteinsencoded [16].

TNF blockers in RA TNF-α belongs to a family of proteins important in immuneregulation and programmed cell death. It has a key role inthe inflammatory process in RA and is expressed by manycells within the immune system. TNF-α is involved instimulating cytokine (including its own) production, inenhancing the expression of adhesion molecules, and inneutrophil activation; it is also a costimulator of T cellactivation and antibody production by B cells [17–19]. Itinduces an array of biological responses in innate as well asadaptive immunity by binding to one of two receptors: TNF-RI (subsequently referred to as the p55 TNF receptor) andTNF-RII (the p75 TNF receptor). These receptors areexpressed on nearly all human cells [20].

Three anti-TNF agents (TNF blockers) are currentlyapproved for use in the treatment of RA. Etanercept is afusion protein consisting of two p75 TNF receptors linked tothe Fc portion of human immunoglobulin G1 (IgG1). Thisagent binds to soluble TNF-α and lymphotoxin-α (LT-α),thereby preventing binding of these molecules to their cell-surface receptors. However, etanercept does not bind totransmembrane TNF-α or induce lysis of target cells. Besidesetanercept, two monoclonal antibodies against TNF-α areavailable: the chimeric antibody infliximab, and the fullyhumanized antibody adalimumab. These two monoclonalantibodies directly bind TNF-α on the cell surface, fixcomplement, and induce lysis of the target cells [21–23].

Pharmacogenetic studies of TNF blockers in RA Pharmacogenetic studies of anti-TNF therapies in RA haveinvolved the obvious candidate gene, TNF, coding for TNF-α,and the genes encoding TNF-α receptors. In recent years,

polymorphisms in genes encoding proteins related to TNF-αhave been identified that may be associated with treatmentoutcomes in RA [24–26]. Several SNPs have been describedin the TNF locus (Fig. 1), both within the TNF gene itself andin proximity to it [27,28]. Among these, SNP –308 A/G inthe promoter region of TNF has been the most widelystudied for its effects on treatment responses to TNF blockersin RA patients. This polymorphism may influence the bindingof transcription factors and control TNF production afterlipopolysaccharide (LPS) stimulation [25,29,30]. Anotherpromoter SNP, TNF –857 C/T, may directly affect thetranscription efficiency of the TNF gene [24] although itseffects on cellular levels of TNF remain controversial [31,32].

Another category of polymorphisms described in the TNFlocus are the DNA microsatellites, which are highlypolymorphic and serve as genetic markers. Microsatellites arerepeat sequences of the bases A and T found in the intronicportions of DNA. Although they are not transcribed intomature mRNA, they can alter DNA folding and conformationand hence the transcription of various enzymes and proteins.The TNF locus has five such microsatellites, TNFa–e, furtherdesignated based on the number of repeat sequences [33].Genetic polymorphisms have also been described for othercytokines and their receptors, such as IL-1, IL-10, and the Fcgamma (Fcγ) receptor. For example, the gene encoding theIL-1 receptor antagonist (IL-1ra), IL1RN, has a variable allelicpolymorphism. The IL1RN*2 allele has been described as afactor determining severity in several autoimmune diseases andhas paradoxically been associated with increased production ofIL-1ra by monocytes in vitro [34]. The Fcγ receptor found onthe surface of cells in the immune system binds the Fc portionof an Ig molecule and affects several cell-specific functions,such as phagocytosis, degranulation, antibody dependent cell-mediated cytotoxicity, cytokine release, and regulation ofantibody production [35]. Fcγ receptor polymorphisms mayalter Fcγ receptor function by enhancing or diminishing itsaffinity for Igs [36]. The most frequent polymorphism of Fcγreceptor IIIA, a subclass of Fc receptor, is a point mutationaffecting amino acids in codon 158 in the extracellular domain [37].

TNF promoter polymorphismsFonseca et al. demonstrated that RA patients with the –308G/G genotype had a better response to infliximab treatmentthan those who had the A/A or A/G genotype [38]. Theystudied the influence of the TNF –308 A/G polymorphism onlong-term responses to infliximab in a prospective fashion in22 consecutive patients. Patients who had the A/A or A/G genotype were compared with those who had the G/G genotype. After 24 months of treatment with infliximab, patients with the –308 G/G genotype had a

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 35

HAMID BASHIR AND PRABHA RANGANATHAN


significantly better response (mean decrease in DAS28 –2.4;p<0.01) than patients in the other groups. The study also demonstrated that the TNF –308 A/G genotype wasassociated with sustained (>1 year) high disease activity and functional decline despite treatment with infliximab,suggesting that carrying the A allele may portend a moresevere disease course in RA with a poorer response to treatment.

Guis et al. reported that RA patients with the –308 G/Ggenotype responded better to etanercept than patients withthe –308 A/G or A/A genotypes [39]. The study involved 86RA patients who were treated with etanercept andgenotyped for the –308 A/G TNF polymorphism. Patientswere subdivided into group A (A/A and A/G genotypes) andgroup G (G/G genotype). The clinical response to etanerceptin groups A and G was compared after 6 and 12 months,using the DAS28. After a 6-month treatment period, 55.6%of patients in group A and 82.4% of patients in group Ghad a DAS28 improvement of >1.2 (p=0.027). After 1 yearof treatment, among 48 patients followed up, 47% ofpatients in group A and 87% in group G maintained animprovement in DAS28 >1.2 (p=0.005).

Kang et al. studied several genetic polymorphisms withinthe TNF and LT-α gene region and showed that the –857C/T SNP in the promoter region of the TNF gene wasassociated with response to etanercept [40]. In this study, 70Korean patients were examined to determine whether thesepolymorphisms were associated with treatment outcomewith etanercept. Patients were divided into responders andnon-responders according to the ACR20 and ACR70response criteria. Only the –857 C/T SNP was significantlyassociated with response; the frequency of the T allele was5% in the ACR20 non-responders and 39% in the ACR70responders (odds ratio [OR] 12, 95% confidence interval[CI] 1.4–105; p=0.0077). Among ACR70 responders, therewere more carriers of the T allele (39%) than the C allele(13%), although this did not reach statistical significance.Moreover, the ratio of ACR70 responders to ACR20 non-responders among the T-allele carriers was >10-fold higherthan in the C allele homozygotes (OR 12, 95% CI 1.2–120;p=0.033), indicating that this SNP may be a useful geneticmarker for predicting response to etanercept.

A recent pharmacogenetic study by Miceli-Richardet al. suggested that a particular TNF haplotype

Figure 1. Chromosomal location of the TNF-α gene and polymorphisms in the region.

HLA: human leukocyte antigen; LT-α: lymphotoxin-α; TNF-α: tumor necrosis factor-α. Redrawn with permission from [28].

Telomere

HLA Class ITNFlocus

320 kb

LT-αa

+249

b c LT-β

850 kb

Microsatellite markers found in the TNF gene locusinclude TNFa, TNFb, TNFc, TNFd and TNFe

Single nucleotide polymorphisms in theregion of the LT-α gene

Single nucleotide polymorphisms in theregion of the TNF-α gene

Centromere

HLA Class II

TNF-α d e

+365 +720 –375

–308 –238

–244 –163 –70 +488

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 36



(–238G/–308G/–857C) may be associated with a lower rate ofresponse to adalimumab in RA patients [41]. The investigatorsstudied 388 patients receiving adalimumab alone or incombination with a disease-modifying antirheumatic drug(DMARD). All patients were genotyped for the HLA-DRB1shared epitope (SE) alleles and three TNF gene polymorphisms(–238A/G, –308A/G, and –857C/T). Extended haplotypesspanning the HLA-DRB1 and TNF loci were constructed.Neither the number of HLA-DRB1 SE copies nor the presenceof the three individual TNF polymorphisms was significantlyassociated with an ACR50 response to adalimumab at 12weeks. However, the –238G/–308G/–857C haplotype wasassociated with a lower rate of ACR50 response toadalimumab at this time point (34% vs. 50% in patientswithout the haplotype; p=0.003).

IL-10 microsatellite polymorphismsA recent study by Schotte et al. looked at microsatellitepolymorphisms in the IL10 gene as predictors of response toetanercept therapy [42]. The IL10 microsatellites IL10.R andIL10.G were genotyped in 50 RA patients. Patients weretreated for up to 4 years with etanercept and the treatmentresponse was assessed by the EULAR criteria. IL10.R3 and theIL.R3-G9 haplotype were associated with a good response toetanercept (OR 5.5, 95% CI 1.6–18, and OR 5.1, 95% CI1.5–18, respectively). In addition, IL10.G13 and the IL10.R2-G13 haplotype were more common among patients with amoderate or no response to etanercept (OR 0.18, 95% CI

0.05–0.61, and OR 0.14, 95% CI 0.04–0.50, respectively).These results suggest that microsatellite polymorphisms in theIL10 gene may play a role in determining response toetanercept in RA patients.

Fcγγ receptor polymorphisms Tutuncu et al. investigated the Fcγ receptor type IIIA (FcγRIIIA)polymorphism and demonstrated that the FcγRIIIA-158 F allelewas a marker of favorable response to anti-TNF therapy [37].The study was designed to assess the clinical outcome in 35patients with RA and psoriatic arthritis treated with the threeanti-TNF agents (adalimumab, etanercept, and infliximab).Each patient’s rheumatologist was asked to categorize thepatient as a good responder or a poor responder based onphysician’s global assessment of disease activity. All patientswere genotyped for the FcγRIIIA-158 polymorphism. Thedistribution of the FcγRIIIA-158 genotype was as follows: Fhomozygous (F/F) 31.5%, V homozygous (V/V) 11.5%, andVF heterozygous (V/F) 57%. Among very good responders(n=23), the distribution of alleles was as follows: F/F 48%,V/V 13%, and V/F 39%. Among non-responders (n=12), thedistribution of alleles was as follows: F/F 0%, V/V 8%, andV/F 92% (p<0.01). Thus, the FcγRIIIA-158 F allele appeared tobe a marker of response to anti-TNF therapy.

IL-1RN/SE polymorphismsMarotte et al. investigated IL1RN and TNF polymorphisms in198 RA patients and found that these polymorphisms did

Table 1. Summary of pharmacogenetic studies of TNF antagonists in RA.

Gene/polymorphism Postulated function of polymorphism Possible clinical effect Reference

TNF –857 Transcription/production of TNF-α TNF –857 T-allele was associated with [40] positive response to etanercept in RA

TNF –308 Transcription/production of TNF-α TNF –308 GG was associated with positive [38,39] response to infliximab and etanercept in RA

TNF –238 Transcription/production of TNF-α TNF –238 G-allele was associated with [41] negative response to adalimumab in RA

IL10 microsatellites Regulation of IL-10 production IL10 microsatellites IL10.G13 and [42] IL10.R2-G13 haplotype were associated with a decreased response to etanercept in RA

FcγRIIIA-158 Binding to IgG leading to phagocytosis FcγRIIIA-158 F-allele was associated [37] with positive response to various TNF blockers in RA and psoriatic arthritis

HLA-DRB1 SE Influence on severity and susceptibility to RA Number of HLA-DRB1 SE copies associated [43] with increased likelihood of selection for infliximab therapy in RA

IgG: immunoglobulin G; IL-10: interleukin-10; FcγRIIIA: Fcγ receptor type IIIA; HLA: human leukocyte antigen; RA: rheumatoid arthritis; SE: shared epitope; TNF-α: tumor necrosis factor-α.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 37

HAMID BASHIR AND PRABHA RANGANATHAN


not predict the clinical response to infliximab [43]. The studysought to determine whether the response to infliximab wasassociated with SE carrier status or selected cytokine genepolymorphisms in IL1RN and TNF (including the –308G/Apromoter polymorphism). The response to infliximab wasdefined using ACR response rates. Neither SE carrier statusnor the polymorphisms predicted response to infliximab.However, the authors demonstrated an association betweenselection for infliximab treatment and SE status, in thatpatients with the SE were almost twice as likely to beselected for infliximab treatment (OR 1.89, 95% CI1.35–2.65; p<0.001). In addition, a dose effect wasobserved between the SE copy number and selection forinfliximab treatment; patients with one SE copy were 1.5times more likely (OR 1.58, 95% CI 1.09–2.27; p=0.01) andpatients with two copies three times more likely (OR 2.96,95% CI 1.92–4.56; p<0.001) to be selected for infliximabtreatment than those with no copies of the SE allele.

Discussion Although biological therapies blocking TNF-α constitute amajor advance in the management of RA, selection of RApatients for these therapies is still empirical and there are nostandard guidelines. Pharmacogenetic testing has significantpotential to optimize therapy with these agents in RApatients, but there are certain facts about the studiesreviewed in this article that need to be considered (Table 1).Most of these studies had small sample sizes and hence wereunderpowered, raising questions about the validity andgeneralizability of the results. In addition, considerablecontradiction exists amongst the available studies. Forexample, some studies showed a positive associationbetween the TNF –308 G/G genotype and response toinfliximab and etanercept [38,39], while others demonstratedno such association [40,43]. More importantly, given thecentral role of TNF in the pathophysiology of RA,polymorphisms in the TNF locus may potentially influencedisease severity and this fact should be taken into accountwhen interpreting such pharmacogenetic data. For instance,Fonseca et al. observed that the TNF-α –308 A/G genotypewas associated with high disease activity and functionaldecline, despite treatment with infliximab [38]. Hence, it maybe difficult to differentiate between the effects of the TNFpolymorphisms on RA disease activity and severity, andresponse to anti-TNF therapy.

SummaryIn summary, although the results of recent pharmacogeneticstudies on the TNF antagonists in RA are intriguing, theyremain hypothesis-generating and need to be replicated andvalidated in larger patient populations. The greatest

challenge to the incorporation of pharmacogenetic testinginto clinical care is not the availability of technologies todetermine large numbers of genotypes, but rather theprecise definition of drug-response phenotypes. Large, well-powered clinical trials are needed before genotype-guidedtherapy can be incorporated into routine clinical care [15].Furthermore, rather than single SNP association studies,large multicenter studies examining composite SNPs andhaplotypes may be more useful in defining the genotypesthat determine drug response. With rapid advances in theInternational HapMap Project (www.hapmap.org) and thededication of major funding agencies to pharmacogeneticsresearch (www.nigms.nih.gov/pharmacogenetics), suchstudies should be feasible and available in the near future.

Disclosures The authors have no relevant financial interests to disclose.

References1. Felson DT, Zhang Y, Hannan MT et al. The incidence and natural history of knee

osteoarthritis in the elderly. The Framingham Osteoarthritis Study. Arthritis Rheum1995;38:1500–5.

2. Panayi GS. The pathogenesis of rheumatoid arthritis: from molecules to the whole patient.Br J Rheumatol 1993;32:533–6.

3. Takemura S, Braun A, Crowson C et al. Lymphoid neogenesis in rheumatoid synovitis. J Immunol 2001;167:1072–80.

4. Dörner T, Burmester GR. The role of B cells in rheumatoid arthritis: mechanisms andtherapeutic targets. Curr Opin Rheumatol 2003;15:246–52.

5. Firestein GS. Etiology and Pathogenesis of Rheumatoid Arthritis. 6th edition. Kelley’sTextbook of Rheumatology. S Ruddy, ED Harris, CB Sledge et al., editors. WB SaundersCompany, Philadelphia, PA. 921–966.

6. Weinblatt ME, Kremer JM, Bankhurst AD et al. A trial of etanercept, a recombinant tumornecrosis factor receptor:Fc fusion protein, in patients with rheumatoid arthritis receivingmethotrexate. N Engl J Med 1999;340:253–9.

7. Maini RN, Breedveld FC, Kalden JR et al. Therapeutic efficacy of multiple intravenousinfusions of anti-tumor necrosis factor alpha monoclonal antibody combined with low-dose weekly methotrexate in rheumatoid arthritis. Arthritis Rheum 1998;41:1552–63.

8. Hyrich KL, Watson KD, Silman AJ et al. Predictors of response to anti-TNF-alpha therapyamong patients with rheumatoid arthritis: results from the British Society forRheumatology Biologics Register. Rheumatology (Oxford) 2006;45:1558–65.

9. Askling J, Fored CM, Brandt L et al. Time-dependent increase in risk of hospitalisation withinfection among Swedish RA patients treated with TNF antagonists. Ann Rheum Dis2007;66:1339–44.

10. Breedveld FC, Weisman MH, Kavanaugh AF et al. The PREMIER study: a multicenter,randomized, double-blind clinical trial of combination therapy with adalimumab plusmethotrexate versus methotrexate alone or adalimumab alone in patients with early,aggressive rheumatoid arthritis who had not had previous methotrexate treatment.Arthritis Rheum 2006;54:26–37.

11. Klareskog L, van der Heijde D, de Jager JP et al.; TEMPO (Trial of Etanercept andMethotrexate with Radiographic Patient Outcomes) study investigators. Therapeutic effectof the combination of etanercept and methotrexate compared with each treatment alonein patients with rheumatoid arthritis: double-blind randomised controlled trial. Lancet2004;363:675–81.

12. Maini RN, Breedveld FC, Kalden JR et al.; Anti-Tumor Necrosis Factor Trial in RheumatoidArthritis with Concomitant Therapy Study Group. Sustained improvement over two yearsin physical function, structural damage, and signs and symptoms among patients withrheumatoid arthritis treated with infliximab and methotrexate. Arthritis Rheum2004;50:1051–65.

13. Kalow W, Maykut MO. The interaction between cholinesterases and a series of localanesthetics. J Pharmacol Exp Ther 1956:116:418–32.

14. Weinshilboum R. Inheritance and drug response. N Engl J Med 2003;348:529–37.

15. Lanfear DE, McLeod HL. Pharmacogenetics: using DNA to optimize drug therapy. Am FamPhysician 2007;76:1179–82.

16. Evans WE, McLeod HL. Pharmacogenomics – drug disposition, drug targets, and sideeffects. N Engl J Med 2003;348:538–49.

17. Podolsky DK. Inflammatory bowel disease. N Engl J Med 2002;347:417–29.

18. Feldmann M, Brennan FM, Maini RN. Role of cytokines in rheumatoid arthritis. Ann RevImmunol 1996;14:397–440.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 38



19. Choy EH, Panayi GS. Cytokine pathways and joint inflammation in rheumatoid arthritis. N Engl J Med 2001;344:907–16.

20. Rigby WF. Drug insight: different mechanisms of action of tumor necrosis factorantagonists – passive-aggressive behavior? Nat Clin Pract Rheumatol 2007;3:227–33.

21. Elliott MJ, Maini RN, Feldmann M et al. Treatment of rheumatoid arthritis with chimericmonoclonal antibodies to tumor necrosis factor alpha. Arthritis Rheum 1993;36:1681–90.

22. Maini RN, Elliott MJ, Brennan FM et al. Monoclonal anti-TNF alpha antibody as a probe ofpathogenesis and therapy of rheumatoid disease. Immunol Rev 1995;144:195–223.

23. Coenen MJ, Toonen EJ, Scheffer H et al. Pharmacogenetics of anti-TNF treatment inpatients with rheumatoid arthritis. Pharmacogenomics 2007;8:761–73.

24. van Heel DA, Udalova IA, De Silva AP et al. Inflammatory bowel disease is associated witha TNF polymorphism that affects an interaction between the OCT1 and NF(-kappa)Btranscription factors. Hum Mol Genet 2002;11:1281–9.

25. Louis E, Franchimont D, Piron A et al. Tumour necrosis factor (TNF) gene polymorphisminfluences TNF-alpha production in lipopolysaccharide (LPS)-stimulated whole blood cellculture in healthy humans. Clin Exp Immunol 1998;113:401–6.

26. Waldron-Lynch F, Adams C, Amos C et al. Tumour necrosis factor 5’ promoter singlenucleotide polymorphisms influence susceptibility to rheumatoid arthritis (RA) inimmunogenetically defined multiplex RA families. Genes Immun 2001;2:82–7.

27. Ranganathan P. Pharmacogenomics of tumor necrosis factor antagonists in rheumatoidarthritis. Pharmacogenomics 2005;6:481–90.

28. Knight JC, Kwiatkowski D. Inherited variability of tumor necrosis factor production andsusceptibility to infectious disease. Proc Assoc Am Physicians 1999;111:290–8.

29. Baseggio L, Bartholin L, Chantome A et al. Allele-specific binding to the –308 singlenucleotide polymorphism site in the tumour necrosis factor-alpha promoter. Eur JImmunogenet 2004;31:15–9.

30. Bouma G, Crusius JB, Oudkerk Pool M et al. Secretion of tumour necrosis factor alpha andlymphotoxin alpha in relation to polymorphisms in the TNF genes and HLA-DR alleles.Relevance for inflammatory bowel disease. Scand J Immunol 1996;43:456–63.

31. Uglialoro AM, Turbay D, Pesavento PA et al. Identification of three new single nucleotidepolymorphisms in the human tumor necrosis factor-alpha gene promoter. Tissue Antigens1998;52:359–67.

32. Higuchi T, Seki N, Kamizono S et al. Polymorphism of the 5’-flanking region of the humantumor necrosis factor (TNF)-alpha gene in Japanese. Tissue Antigens 1998;51:605–12.

33. Udalova IA, Nedospasov SA, Webb GC et al. Highly informative typing of the human TNFlocus using six adjacent polymorphic markers. Genomics 1993;16:180–6.

34. Perrier S, Coussediere C, Dubost JJ et al. IL-1 receptor antagonist (IL-1RA) genepolymorphism in Sjogren’s syndrome and rheumatoid arthritis. Clin ImmunolImmunopathol 1998;87:309–13.

35. Binstadt BA, Geha RS, Bonilla FA. IgG Fc receptor polymorphisms in human disease:implications for intravenous immunoglobulin therapy. J Allergy Clin Immunol2003;111:697–703.

36. de Haas M, Koene HR, Kleijer M et al. A triallelic Fc gamma receptor type IIIApolymorphism influences the binding of human IgG by NK cell Fc gamma RIIIa. J Immunol1996;156:2948–55.

37. Tutuncu Z, Kavanaugh A, Zvaifler N et al. Fcgamma receptor type IIIA polymorphismsinfluence treatment outcomes in patients with inflammatory arthritis treated with tumornecrosis factor alpha-blocking agents. Arthritis Rheum 2005;52:2693–6.

38. Fonseca JE, Carvalho T, Cruz M et al. Polymorphism at position –308 of the tumournecrosis factor alpha gene and rheumatoid arthritis pharmacogenetics. Ann Rheum Dis2005;64:793–4.

39. Guis S, Balandraud N, Bouvenot J et al. Influence of –308 A/G polymorphism in the tumornecrosis factor alpha gene on etanercept treatment in rheumatoid arthritis. ArthritisRheum 2007;57:1426–30.

40. Kang CP, Lee KW, Yoo DH et al. The influence of a polymorphism at position –857 of thetumour necrosis factor alpha gene on clinical response to etanercept therapy inrheumatoid arthritis. Rheumatology (Oxford) 2005;44:547–52.

41. Miceli-Richard C, Comets E, Verstuyft C et al. A single tumour necrosis factor haplotypeinfluences the response to adalimumab in rheumatoid arthritis. Ann Rheum Dis2008;67:478–84.

42. Schotte H, Schluter B, Drynda S et al. Interleukin 10 promoter microsatellitepolymorphisms are associated with response to long term treatment with etanercept inpatients with rheumatoid arthritis. Ann Rheum Dis 2005;64:575–81.

43. Marotte H, Pallot-Prades B, Grange L et al. The shared epitope is a marker of severityassociated with selection for, but not with response to, infliximab in a large rheumatoidarthritis population. Ann Rheum Dis 2006;65:342–7.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 39

The mortality rate in rheumatoid arthritis (RA) is higher thanthat in the general population, and standardized mortalityratios (SMRs) range from 1.3–3.0 [1]. This increasedmortality rate is largely attributable to cardiovascular disease(CVD), principally atherosclerosis. CV morbidity also appearsto be at least two-fold greater in RA in comparison with thegeneral population [2]. The raised CV risk in RA may haveseveral causes. Firstly, the prevalence of CV risk factors, suchas dyslipidemia, diabetes mellitus, hypertension, a higherbody mass index (BMI), a higher waist-to-hip ratio, or lackof physical activity, may be increased. Secondly, in chronicdiseases, unrelated disorders such as hypertension arefrequently under-treated [3,4]. Finally, the chronicinflammatory process in RA might itself mediate a higher CV risk.

Cardiovascular mortalityMost studies indicate that the risk of dying from CV diseasein RA, particularly coronary heart disease, is increased bymore than two-fold in comparison with the generalpopulation. However, some recent reports described noenhanced mortality rate in RA, which might be due to theinclusion of inception cohorts (as the enhanced mortality

rate becomes more apparent after 10 years of diseaseduration) [1,5]. Another possible explanation for thisdiscrepancy might be the more intensive treatment of RAemployed nowadays, which leads to a longer overall lifeexpectancy. Nevertheless, this survival improvement remainsbelow that of the general population; thus, SMRs remainenhanced. This is illustrated by a recent study in which noreduction in mortality rate was observed; in fact, a relativeincrease in comparison with the general population was seen[6]. The mortality rates between 1965 and 2005 wererelatively constant at 2.4 and 2.5 per 100 person-years forfemale and male RA patients, respectively. In contrast, theexpected mortality rate in female population control subjectsdeclined from 1.0 in 1965 to 0.2 per 100 person-years in2000. For male subjects, these figures were 1.3 and 0.2 per100 person-years in 1965 and 2000, respectively. Thesefindings indicate a widening mortality gap between RApatients and the general population.

Cardiovascular morbidityAs stated above, several investigations have indicated anelevated rate of CVD in RA. In a prospective Dutch study,the magnitude of this enhanced CV risk relative to thatconferred by type 2 diabetes – a well-established CV riskfactor – was investigated [2]. The prevalence of CVD wasdetermined in 353 RA patients in the CARRÉ (CardiovascularResearch and Rheumatoid Arthritis) study, and inparticipants of a population-based cohort study on CVD and

LEA

DIN

G A

RTIC

LE

Cardiovascular Aspects ofRheumatoid Arthritis

Michael T Nurmohamed, MD, PhDDepartments of Internal Medicine and Rheumatology, VU University Medical Centre; Department ofRheumatology, Jan van Breemen Institute, Amsterdam, The Netherlands

The increased mortality rate in rheumatoid arthritis (RA) is predominantly due to atherosclerotic cardiovascular (CV) disease.CV morbidity is also elevated in comparison with the general population. This increased CV risk may be caused by a greaterprevalence of CV risk factors, under-treatment of these risk factors, or RA itself – particularly its chronic inflammatorycomponent. However, CV risk factors only partly explain the enhanced CV risk and it is becoming increasingly acknowledgedthat the inflammatory process in RA plays a pivotal role. This is probably related to the fact that atherosclerosis also has aninflammatory component and RA accelerates this process. Equally, it is possible that effective suppression of inflammation bydisease-modifying antirheumatic drugs and/or biological agents could lower the CV risk. There is now accumulating evidencethat the enhanced CV risk in RA is similar to that observed in type 2 diabetes, and RA should therefore be seen as a new CVrisk factor for which appropriate risk management is mandatory. Int J Adv Rheumatol 2008;6(2):40–6.


Address for correspondence: Michael T Nurmohamed, Departments of

Internal Medicine and Rheumatology, VU University Medical Centre,

De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands.

Email: [email protected]

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 40

CARDIOVASCULAR ASPECTS OF RA


its risk factors (the Hoorn study). Non-diabetic RA patients(n=294) were compared with non-diabetic, non-RAindividuals (n=258), and individuals with type 2 diabetes(n=194) from the Hoorn study. The adjusted odds ratios(OR) for CVD were 3.1 in RA patients and 2.3 in individualswith type 2 diabetes in comparison with non-diabeticindividuals. This investigation indicates that the prevalenceof CVD in RA is at least comparable with its prevalence indiabetes, and is increased at least two-fold relative to thegeneral population.

Myocardial infarctionRA patients who experience a myocardial infarction (MI) havean approximately doubled rate of multivessel coronary diseasein comparison with MI patients without RA, indicatingaccelerated atherosclerosis in RA [7]. In addition, the casefatality rate in RA patients following an MI is almost two-foldgreater than that in individuals without RA [8,9]. It is alsoimportant to realize that RA patients are less likely to reportangina and are twice as likely to have a silent MI and suddendeath in comparison with the general population [10].

To date, two follow-up studies with CV endpointassessment have been published. In the first, 234 patientswith RA were followed up for 1 year, and 15 CV events(predominantly MIs) occurred during 252 patient-years [11].As a comparator group, >4600 participants of a community-based cohort were followed for up to 8 years; in this group,200 CV events occurred. The age- and sex-adjustedincidence risk ratio of CV events associated with RA was 4.0(95% confidence interval [CI] 1.86–8.43), which decreasedto 3.2 when adjusting for CV risk factors. The second studyconsidered the 3-year follow-up of the CARRÉ investigation[12]. In that analysis, CV events were reported in 9% of theRA patients and in 4% of the general population. The age-and gender-adjusted relative risk (RR) of a CV event was 2.0(95% CI 1.3–3.1).

Congestive heart failureOne of the first investigations revealing an increasedlikelihood of congestive heart failure in RA indicated an RR of 1.6 during follow-up of 450 RA patients and 450control subjects [13]. A much larger sample of patients camefrom the National Data Bank for Rheumatic Diseases [14]. Atotal of 13 171 RA patients and 2568 osteoarthritis patientswere followed over a 2-year period; heart failure wasobserved in 461 (3.9%) of those with RA and in 87 (2.3%)osteoarthritis patients. Other database investigations haveindicated similar results. However, an important pitfall of theheart failure studies is that diagnosis has always been basedon clinical criteria, rendering it less reliable. Obviously, theuse of echocardiography for assessment of heart failure

might solve this problem. In this regard, congestive heartfailure is generally the result of either systolic or diastolicventricular dysfunction, and an increased frequency of leftventricular diastolic dysfunction has been found in patientswith long-standing RA who do not have CV risk factors orclinically evident CV manifestations [15]. Moreover, anincreased frequency of subclinical pulmonary hypertensionhas also been identified in these patients compared with matched controls.

StrokeWolfe et al. reported the results of a US survey of CVD in9093 RA patients and 2479 osteoarthritis patients [16].Compared with osteoarthritis patients, RA patients had anOR of 1.7 (95% CI 1.3–2.2) for current stroke and an OR of1.1 (95% CI 0.9–1.2) for lifetime stroke. In contrast,Solomon et al. followed 25 385 British Columbian adultswith RA over 5 years and found an incidence rate per 1000 years of 5.1 in RA and 2.7 in non-RA subjects,resulting in a rate ratio of 1.9 (95% CI 1.7–2.1) [17].Another database investigation in 11 633 RA patients in theUK, with a follow-up of 5 years, revealed an incidence rateof stroke of 7.0 per 1000 patient-years and a relative risk of1.4 (95% CI 1.3–1.5) versus non-RA subjects [18]. Thesedata strongly suggest that the risk for stroke is enhanced inRA patients, although the increase may be less incomparison with the risk for an MI.

Peripheral arterial diseaseA database investigation from an integrated US health planincluding 28 208 RA patients revealed a prevalence ofperipheral arterial disease of 4.7 in RA patients and 1.7 inmatched controls (OR 2.3; 95% CI 2.3–2.6) [19]. Liang etal. found a 30-year cumulative incidence rate of peripheralarterial disease of 20% in 609 incident RA patients whowere diagnosed during 1955–1994 and followed up to theyear 2000 [20]. This is approximately 30% higher thanwould have been expected from prevalence data in non-RAsubjects [20]. In an elegant study, Del Rincón et al.investigated the stiffness of limb arteries as a marker forperipheral arterial disease in 234 RA patients and 102control subjects [21]. Among the RA patients, 66 of 931arteries (7%) were incompressible and 30 (3%) wereobstructed; for the control subjects, the percentages were0.7% and 1%, respectively.

Preclinical atherosclerosisIt is increasingly acknowledged that carotid artery intima-media thickness (IMT) is an important marker for early,preclinical atherosclerosis and a strong predictor of futureCV events. IMT is assessed non-invasively with

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 41

MICHAEL T NURMOHAMED


echocardiographic techniques (carotid ultrasound). In alandmark study of 4476 general population personsfollowed for 6 years, every 0.20 mm increase of themaximum IMT of the common carotid artery was associatedwith a 30% increase in the incidence of new CV events [22].Consequently, carotid artery IMT in RA has been assessed byseveral investigators. One of the first studies was conductedby Japanese researchers who found significantly increasedcommon carotid IMT in 138 RA patients (mean age55 years) in comparison with 94 matched controls (meanage 52 years), with values of 0.64 mm and 0.58 mm,respectively (p<0.05) [23].

Accurate longitudinal data on the progression of subclinicalatherosclerosis in RA are still sparse, but data were recentlyindirectly derived from a cross-sectional investigation. delRincón et al. found that the rate at which the IMT increasedwas related to disease duration and ranged from 0.15 mm/10 years among RA patients with a disease duration of<7 years, to 0.30 mm/10 years in RA patients with a diseaseduration of >20 years [24]. Therefore, patients with prolongedRA have more advanced atherosclerosis than patients of thesame age but who have a shorter disease duration, whichsuggests that RA accelerates atherosclerosis. This is in line withthe observation that atherosclerosis represents a chronicinflammatory process of the artery and, therefore, that longerRA disease duration will lead to a greater atheroscleroticburden (discussed below in “Atherosclerosis andinflammation”). Gonzalez-Juanatey and colleagues recentlyfound that carotid artery IMT had a high predictive power forthe development of CV events over a 5-year follow-up periodin 47 patients with RA who did not have clinically evident CVdisease at the time of the carotid ultrasonography evaluation [25].

An emerging technique for assessing coronary athero-sclerosis, including preclinical disease, is the determination ofcoronary artery calcification by computed tomography [26];the presence of such coronary calcifications is a strongpredictor of subsequent coronary heart disease. Chung et al.found coronary calcification in 61% of patients withestablished RA, 43% of early RA patients, and in 38% ofcontrol subjects [27]. The OR for more severe coronarycalcification was 3.4 in patients with established disease. Therelationship between coronary calcification and RA diseaseduration was recently confirmed in another study [28].

Cardiovascular risk factors in RADyslipidemia Increased levels of total cholesterol (TC) and low-densitylipoprotein cholesterol (LDL-C), and a decreased level ofhigh-density lipoprotein cholesterol (HDL-C) are associatedwith a higher CV risk. The available literature on lipid profiles

in patients with RA is contradictory but it appears that thereis an inverse relationship between disease activity and lipidlevels [29].

When does the dyslipidemia start?Dyslipidemia is already present in early RA and so the questionof whether or not this phenomenon starts in the preclinicalphase of the disease has emerged. Hence, the present authorand colleagues have investigated the lipid profile over time,and its relationship with inflammation, in subjects who laterdeveloped RA [30]. Future RA patients displayed 4% higherTC, 9% lower HDL-C, and 17% higher triglyceride levels thanmatched controls (p≤0.05), at least 10 years before the onsetof symptoms. Although the differences between the variouslipid values were small, they are clinically relevant, particularlyin the light of results from other studies [31].

Functional properties of lipid particlesProtecting LDL from oxidation is one the anti-atherogenicroles of normal HDL. This anti-inflammatory HDL is distinctfrom the so-called pro-inflammatory HDL, which is devoid ofthese properties and may actually promote atherosclerosis.Indeed, in a small study that included 48 women with RA and72 healthy subjects, McMahon and colleagues showed thatpro-inflammatory HDL was detected more frequently in RApatients (20%) than in control subjects (4%) [32].

Antirheumatic treatment and lipid profileTreatment with disease-modifying antirheumatic drugs(DMARDs) has beneficial effects on the lipid profile inpatients with early active RA [33]. These agents do mediatean increase in TC, but they induce a more pronouncedincrease in HDL-C, resulting in a lower (more favorable)atherogenic index (TC/HDL-C ratio), which is an importantprognostic CV risk factor.

Investigations of tumor necrosis factor (TNF)-blockingagents reveal a transient increase in TC and HDL-C levels,mostly accompanied by improvement of the atherogenicindex, during the first few months of the treatment.Thereafter, the results become divergent between thestudies (reviewed in [33]). This may be due to differences indisease activity, changes in co-medication (particularlyprednisone), dietary intake, and physical activity. Hence,future studies should appropriately address these potentialconfounders in order to reach valid conclusions.

Lifestyle factors and medical comorbiditiesSmoking is an important risk factor for CV disease in thegeneral population. It is conceivable that smoking mightplay a role in CV disease in RA as it increases thesusceptibility for the development of RA as well its severity

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 42



[34]. A recent investigation revealed that smoking was moreprevalent in RA patients and that there was an increased CVrisk in these subjects compared with that in non-smoking RApatients [35]. However, the effect of smoking wassignificantly less than that observed in control (non-RA)subjects (hazard ratio [HR] for CV disease was 1.3 and 2.2for smoking vs. non-smoking RA patients and smoking vs.non-smoking controls, respectively).

Published data on the prevalence of type 2 diabetes in RAare not concordant; however, as there is increasing evidencefor elevated rates of insulin resistance in RA [36], it isconceivable that the prevalence of diabetes might beenhanced. Similarly, data regarding hypertension in RA arecontradictory and no conclusions can be reached with respectthe contribution of body mass index to the CV risk in RA.

In summary, the abovementioned evidence is notuniform, adequate comparative investigations are lacking,and studies are limited by a lack of control for importantconfounders. Nevertheless, the prevalences of hypertensionand smoking appear to be increased in RA patients.

Very recently, a systematic review investigated theeffectiveness of exercise interventions in patients with RA[37]. Unfortunately, no studies investigating the effect ofexercise interventions in relation to CV disease in RA wereidentified. Therefore, further studies are required to addressthis topic.

Under-treatment of hypertension in RAIt is well known that comorbidity is under-treated in patientswith chronic diseases [38]. Moreover, it is known thatcomorbidities in patients with RA do not differ fromcomorbidities observed in other chronic diseases [39]. Thusfar, just one study has addressed this topic [4], with a totalof 400 consecutive RA patients investigated forhypertension and antihypertensive drug use. Hypertensionwas present in 71% of the patients; only 61% of thesereceived antihypertensives. These results clearly indicate thatfuture research in this area is necessary.

Atherosclerosis and inflammationFormerly, atherosclerosis was seen as an accumulation of lipidswithin the arterial wall. However, during the last few decades,it has become acknowledged that atherosclerosis represents achronic inflammatory process in the artery (Fig.1). Endothelialdysfunction is the first step of atherosclerosis and is induced bydifferent cardiovascular risk factors, e.g. oxidized LDL-C,smoking, hypertension, or diabetes. The endothelium becomesmore permeable to lipoproteins and acquires procoagulantrather than anticoagulant properties. Increased permeability toinflammatory and muscle cells also occurs. Inflammatorymediators such as TNF-α and interleukin-1 cause increased

binding of (modified) LDL-C to the endothelium and musclecells. Modified LDL-C accumulates within macrophagesresulting in the formation of foam cells and subsequent fattystreaks. This lesion progresses and a fibrous cap is formed,which consists of smooth muscle cells and a collagen matrixthat separates the atherosclerotic plaque lesion from thearterial lumen [40,41].

During this process, the arterial wall becomes thicker;however, initially, the lumen remains unaltered owing todilatation. Accumulation and activation of macrophages and Tlymphocytes leads to the release of several mediators causingfurther damage, and, ultimately, narrowing of the artery doesoccur. Platelet activation by the dysfunctional endotheliumresults in the formation of thromboxane – a potentvasoconstrictor and platelet aggregator. Finally, plaque ruptureand thrombosis leads to unstable angina or MI. This processaccounts for up to 70% of the acute coronary syndromes [41].The cellular interactions that are seen in the development ofatherosclerosis are similar to those observed in chronicinflammatory diseases such as RA, which might explain theincreased CV risk in patients with RA. In keeping with thisview, high-resolution B-mode ultrasound of the commoncarotid artery identified a strong correlation between thecarotid IMT and markers of systemic inflammation in patientswith RA and in healthy subjects [24,42].

Antirheumatic treatment and CV risk in RAAcetaminophenChan et al. examined the relationship between acetaminophenuse and major CV events in a prospective cohort of 70 971women [43]. Frequent usage, defined as on ≥22 days/month,was linked with an elevated risk of CV events. The highest riskwas associated with the intake of ≥15 tablets per week (RR1.7, 95% CI 1.1–2.6), with 69% of patients taking tablets of≥500 mg acetaminophen. The observed association might bemediated through the induction of hypertension due tocyclooxygenase-2 (COX-2) inhibition [44].

GlucocorticoidsThe place of glucocorticoids in RA therapy is a continuingmatter of debate in view of their associated CV side effects,which include hypertension, dyslipidemia, insulin resistance,and diabetes. These side effects are particularly linked toprolonged exposure to high-dose glucocorticoids, withhypertension demonstrated to be especially prominent inpatients receiving ≥7.5 mg/day prednisone for a period of≥6 months [45]. Conversely, low-dose glucocorticoids mighthave beneficial effects on the lipid profile [46].

There is no doubt that corticosteroids rapidly andeffectively suppress inflammation in RA and their use maybe justified as a short-term treatment, e.g. as a “bridging

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 43



Figure 1. Effects of T cell activation on plaque inflammation. Antigens presented by macrophages and dendritic cells (antigen-presenting cells) trigger the activation of antigen-specific T cells in the artery. Most of the activated T cells produce Th1 cytokines(e.g. IFN-γ), which activate macrophages and vascular cells, leading to inflammation. Regulatory T cells modulate the process bysecreting anti-inflammatory cytokines (such as IL-10 and TGF-β).

IFN-γ: interferon-γ; IL-10: interleukin-10; LDL: low-density lipoprotein; TGF-β: transforming growth factor-β; Th1: type 1 helper T cell. Redrawn with permission from [41].

Coronaryartery

Adhesion Migration Endothelium

Th1Antigen-

presenting cell

Inhibitorycytokines

Th1 cytokines

Processing

Processing

Smooth-muscle cell

Inflammation

RegulatoryT cell

Antigen-presenting cell

Antigens:oxidized LDL,

heat-shock proteins,microbes

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 44



therapy” during the period between the initiation of andresponse to DMARDs [47]. The debate regarding the use ofcorticosteroids continues.

Nonsteroidal anti-inflammatory drugsand COX-2-selective inhibitorsDuring the last decade there has been an ongoing debate asto whether or not COX-2-selective inhibitors (COXIBs) areassociated with an enhanced CV risk, particularly MIs.Several large-scale, placebo-controlled trials havedemonstrated that COXIBs are accompanied by anapproximately two-fold enhanced CV risk [48,49].

There are many observational database studies indicatingeither an increased on no increased CV risk withnonsteroidal anti-inflammatory drug (NSAID) therapy.Observational investigations have inherent methodologicalproblems that can only be addressed by the performance oflarge-scale, randomized investigations. However, suchstudies have not been conducted with NSAIDs. As analternative (albeit a second-best option), meta-analyses withmeta-regression techniques of comparative trials of NSAIDsand COXIBs can be employed in order to reach validconclusions regarding the CV risks associated with NSAIDuse. From one such study, and a large controlledinvestigation comparing a COXIB with a non-naproxenNSAID, it appears that non-naproxen NSAIDs confer asimilar CV risk to COXIBs [50,51].

DMARDsIn a prospective investigation of 1240 patients in which 190patients died during a follow-up of 18 years, the CVmortality rate was 70% lower in patients receivingmethotrexate than in those who did not receive the drug[52]. Another study revealed that patients who do notrespond to methotrexate treatment have a poor prognosis[53]. Not only is CV mortality reduced by methotrexate;from two case–control studies, it appears that the use ofmethotrexate is also associated with less CV morbidity[54,55].

TNF-blocking agentsJacobsson et al. linked a Swedish database – in which 921 of1430 patients received TNF-blockers – with a nationalmortality register. The adjusted HR for death was 0.65 whencomparing TNF blockade versus no TNF blockade [56]. In anearlier study, these investigators determined the incidence ofMIs in 983 RA patients, of whom 531 underwent treatmentwith TNF blockers [57]. There were 13 CV events in theanti-TNF-treated patients resulting in 14 CV events per 1000person-years, compared with 35 CV events per 1000person-years in RA patients not treated with anti-TNF. The

adjusted rate ratio was 0.46 in anti-TNF-treated comparedwith non-anti-TNF treated subjects. From a larger Britishcohort study, the risk of an MI was markedly reduced whencomparing anti-TNF responders with the non-responders(incidence ratio 0.36) [58].

Cardiovascular risk management in RA In summary, RA should be seen as a new CV risk factor, forwhich CV-risk management (CV-RM) is mandatory. CV-RM isgenerally performed on the basis of the 10-year absolute riskfor a (fatal) CV-event, which is derived from a CV risk formulabased on several CV risk factors. Examples include theFramingham risk calculator and the Systematic Coronary RiskEvaluation (SCORE). Treatment with statins and/orantihypertensives is then initiated at above a certain threshold,e.g. a 10-year CV-mortality risk of ≥10%. Thus far, CV riskfunction scores are not available for patients with RA and,therefore, existing CV risk functions such as SCORE should beadapted, for example by a multiplier, so that they can be usedto assess the CV risk in RA patients. Regular CV risk screeningappears to be appropriate for RA patients and lifestylerecommendations should be given to every patient. Treatmentwith statins and/or antihypertensives should be consideredwhen the 10-year CV risk is above a certain value. In addition,aggressive suppression of the inflammation is recommendedto further lower the CV risk [59].

DisclosuresThe author has no relevant financial interests to disclose.

References1. Van Doornum S, McColl G, Wicks IP. Accelerated atherosclerosis: an extraarticular feature

of rheumatoid arthritis? Arthritis Rheum 2002;46:862–73.

2. van Halm VP, Peters MJ, Voskuyl AE et al. Rheumatoid arthritis versus diabetes as a riskfactor for cardiovascular disease, a cross sectional study. The CARRE Investigation. AnnRheum Dis 2008; [Epub ahead of print].

3. Boers M, Dijkmans B, Gabriel S et al. Making an impact on mortality in rheumatoidarthritis: targeting cardiovascular comorbidity. Arthritis Rheum 2004;50:1734–39.

4. Panoulas VF, Douglas KM, Milionis HJ et al. Prevalence and associations of hypertensionand its control in patients with rheumatoid arthritis. Rheumatology (Oxford)2007;46:1477–82.

5. Ward MM. Recent improvements in survival in patients with rheumatoid arthritis: betteroutcomes or different study designs? Arthritis Rheum 2001;44:1467–9.

6. Gonzalez A, Maradit Kremers H, Crowson CS et al. The widening mortality gap betweenrheumatoid arthritis patients and the general population. Arthritis Rheum2007;56:3583–7.

7. Warrington KJ, Kent PD, Frye RL et al. Rheumatoid arthritis is an independent risk factorfor multi-vessel coronary artery disease: a case control study. Arthritis Res Ther2005;7:R984–91.

8. Van Doornum S, Brand C, King B et al. Increased case fatality rates following a first acutecardiovascular event in patients with rheumatoid arthritis. Arthritis Rheum2006;54:2061–8.

9. Södergren A, Stegmayr B, Lundberg V et al. Increased incidence of and impaired prognosisafter acute myocardial infarction among patients with seropositive rheumatoid arthritis.Ann Rheum Dis 2007;66:263–6.

10. Maradit-Kremers H, Crowson CS, Nicola PJ et al. Increased unrecognized coronary heartdisease and sudden deaths in rheumatoid arthritis: a population-based cohort study.Arthritis Rheum 2005;52:402–11.

11. del Rincón ID, Williams K, Stern MP et al. High incidence of cardiovascular events in arheumatoid arthritis cohort not explained by traditional cardiac risk factors. ArthritisRheum 2001;44:2737–45.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 45



12. Peters MJ, van Halm VP, Voskuyl AE et al. Prospective study reveals rheumatoid arthritis tobe an important independent risk factor for incident cardiovascular disease. Ann RheumDis 2008;67(Suppl II):309.

13. Gabriel SE, Crowson CS, O’Fallon WM. Comorbidity in arthritis. J Rheumatol1999;26:2475–9.

14. Wolfe F, Michaud K. Heart failure in rheumatoid arthritis: rates, predictors, and the effectof anti-tumor necrosis factor therapy. Am J Med 2004;116:305–11.

15. Gonzalez-Juanatey C, Testa A, Garcia-Castelo A et al. Echocardiographic and Dopplerfindings in long-term treated rheumatoid arthritis patients without clinically evidentcardiovascular disease. Semin Arthritis Rheum 2004;33:231–8.

16. Wolfe F, Freundlich B, Straus WL. Increase in cardiovascular and cerebrovascular diseaseprevalence in rheumatoid arthritis. J Rheumatol 2003;30:36–40.

17. Solomon DH, Goodson NJ, Katz JN et al. Patterns of cardiovascular risk in rheumatoidarthritis. Ann Rheum Dis 2006;65:1608–12.

18. Watson DJ, Rhodes T, Guess HA. All-cause mortality and vascular events among patientswith rheumatoid arthritis, osteoarthritis, or no arthritis in the UK General Practice ResearchDatabase. J Rheumatol 2003;30:1196–1202.

19. Han C, Robinson DW Jr, Hackett MV et al. Cardiovascular disease and risk factors inpatients with rheumatoid arthritis, psoriatic arthritis, and ankylosing spondylitis.J Rheumatol 2006;33:2167–72.

20. Liang KP, Liang KV, Matteson EL et al. Incidence of noncardiac vascular disease inrheumatoid arthritis and relationship to extraarticular disease manifestations. ArthritisRheum 2006;54:642–8.

21. del Rincón I, Haas RW, Pogosian S et al. Lower limb arterial incompressibility andobstruction in rheumatoid arthritis. Ann Rheum Dis 2005;64:425–32.

22. O’Leary DH, Polak JF, Kronmal RA et al. Carotid-artery intima and media thickness as arisk factor for myocardial infarction and stroke in older adults. Cardiovascular Health StudyCollaborative Research Group. N Engl J Med 1999;340:14–22.

23. Kumeda Y, Inaba M, Goto H et al. Increased thickness of the arterial intima-mediadetected by ultrasonography in patients with rheumatoid arthritis. Arthritis Rheum2002;46:1489–97.

24. Del Rincón I, Williams K, Stern MP et al. Association between carotid atherosclerosis andmarkers of inflammation in rheumatoid arthritis patients and healthy subjects. ArthritisRheum 2003;48:1833–40.

25. Gonzalez-Juanatey C, Llorca J, Martin J et al. Carotid intima-media thickness predicts thedevelopment of cardiovascular events in patients with rheumatoid arthritis. Semin ArthritisRheum 2008; [Epub ahead of print].

26. Greenland P, LaBree L, Azen SP et al. Coronary artery calcium score combined withFramingham score for risk prediction in asymptomatic individuals. JAMA 2004;291:210–5.

27. Chung CP, Oeser A, Raggi P et al. Increased coronary-artery atherosclerosis in rheumatoidarthritis: relationship to disease duration and cardiovascular risk factors. Arthritis Rheum2005;52:3045–53.

28. Kao AH, Krishnaswami S, Cunningham A et al. Subclinical coronary artery calcification andrelationship to disease duration in women with rheumatoid arthritis. J Rheumatol2008;35:61–9.

29. White D, Fayez S, Doube A. Atherogenic lipid profiles in rheumatoid arthritis. N Z Med J2006;119:U2125.

30. van Halm VP, Nielen MM, Nurmohamed MT et al. Lipids and inflammation: serialmeasurements of the lipid profile of blood donors who later developed rheumatoidarthritis. Ann Rheum Dis 2007;66:184–8.

31. Menotti A, Kromhout D, Blackburn H et al. Forty-year mortality from cardiovasculardiseases and all causes of death in the US Railroad cohort of the Seven Countries Study.Eur J Epidemiol 2004;19:417–24.

32. McMahon M, Grossman J, FitzGerald J et al. Proinflammatory high-density lipoprotein as abiomarker for atherosclerosis in patients with systemic lupus erythematosus andrheumatoid arthritis. Arthritis Rheum 2006;54:2541–9.

33. Nurmohamed MT. Atherogenic lipid profiles and its management in patients withrheumatoid arthritis. Vasc Health Risk Manag 2007;3:845–52.

34. Goodson NJ, Farragher TM, Symmons DP. Rheumatoid factor, smoking, and diseaseseverity: associations with mortality in rheumatoid arthritis. J Rheumatol 2008;35:945–9.

35. Gonzalez A, Maradit Kremers H, Crowson CS et al. Do cardiovascular risk factors conferthe same risk for cardiovascular outcomes in rheumatoid arthritis patients as in non-rheumatoid arthritis patients? Ann Rheum Dis 2008;67:64–9.

36. Gonzalez-Gay MA, Gonzalez-Juanatey C, Martin J. Rheumatoid arthritis: a diseaseassociated with accelerated atherogenesis. Semin Arthritis Rheum 2005;35:8–17.

37. Metsios GS, Stavropoulos-Kalinoglou A, Veldhuijzen van Zanten JJ et al. Rheumatoidarthritis, cardiovascular disease and physical exercise: a systematic review. Rheumatology(Oxford) 2008;47:239–48.

38. Redelmeier DA, Tan SH, Booth GL. The treatment of unrelated disorders in patients withchronic medical diseases. N Engl J Med 1998;338:1516–20.

39. Kroot EJ, van Gestel AM, Swinkels HL et al. Chronic comorbidity in patients with earlyrheumatoid arthritis: a descriptive study. J Rheumatol 2001;28:1511–7.

40. Ross R. Atherosclerosis – an inflammatory disease. N Engl J Med 1999;340:115–26.

41. Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J Med2005;352:1685–95.

42. Gonzalez-Gay MA, Gonzalez-Juanatey C, Pineiro A et al. High-grade c-reactive proteinelevation correlates with accelerated atherogenesis in patients with rheumatoid arthritis.J Rheumatol 2005;32:1219–23.

43. Chan AT, Manson JE, Albert CM et al. Nonsteroidal antiinflammatory drugs,acetaminophen, and the risk of cardiovascular events. Circulation 2006;113:1578–87.

44. Hinz B, Cheremina O, Brune K. Acetaminophen (paracetamol) is a selectivecyclooxygenase-2 inhibitor in man. FASEB J 2008;22:383–90.

45. Panoulas VF, Douglas KM, Stavropoulos-Kalinoglou A et al. Long-term exposure tomedium-dose glucocorticoid therapy associates with hypertension in patients withrheumatoid arthritis. Rheumatology (Oxford) 2008;47:72–5.

46. Boers M, Nurmohamed MT, Doelman CJ et al. Influence of glucocorticoids and diseaseactivity on total and high density lipoprotein cholesterol in patients with rheumatoidarthritis. Ann Rheum Dis 2003;62:842–5.

47. Vis M, Nurmohamed MT, Wolbink G et al. Short term effects of infliximab on the lipidprofile in patients with rheumatoid arthritis. J Rheumatol 2005;32:252–5.

48. Bresalier RS, Sandler RS, Quan H et al. Cardiovascular events associated with rofecoxib incolerectal adenoma chemoprevention trial. N Engl J Med 2005;352:1092–102.

49. Solomon SD, McMurray JJ, Pfeffer MA et al.; APC Study Investigators. Cardiovascular riskassociated with celecoxib in a clinical trial for colorectal adenoma prevention. N Engl JMed 2005;352:1071–80.

50. Kearney PM, Baigent C, Godwin J et al. Do selective cyclo-oxygenase-2 inhibitors andtraditional non-steroidal anti-inflammatory drugs increase the risk of atherothrombosis?Meta-analysis of randomised trials. BMJ 2006;332:1302–8.

51. Cannon CP, Curtis SP, FitzGerald GA et al.; MEDAL Steering Committee. Cardiovascularoutcomes with etoricoxib and diclofenac in patients with osteoarthritis and rheumatoidarthritis in the Multinational Etoricoxib and Diclofenac Arthritis Long-term (MEDAL)programme: a randomised comparison. Lancet 2006;368:1771–81.

52. Choi HK, Hernan MA, Seeger JD et al. Methotrexate and mortality in patients withrheumatoid arthritis: a prospective study. Lancet 2002;359:1173–7.

53. Krause D, Schleusser B, Herborn G et al. Response to methotrexate treatment is associatedwith reduced mortality in patients with severe rheumatoid arthritis. Arthritis Rheum2000;43:14–21.

54. van Halm VP, Nurmohamed MT, Twisk JW et al. Disease-modifying antirheumatic drugsare associated with a reduced risk for cardiovascular disease in patients with rheumatoidarthritis: a case control study. Arthritis Res Ther 2006;8:R151.

55. Suissa S, Bernatsky S, Hudson M. Antirheumatic drug use and the risk of acute myocardialinfarction. Arthritis Rheum 2006;55:531–6.

56. Jacobsson LT, Turesson C, Nilsson JA et al. Treatment with TNF blockers and mortality riskin patients with rheumatoid arthritis. Ann Rheum Dis 2007;66:670–5.

57. Jacobsson LT, Turesson C, Gulfe A et al. Treatment with tumor necrosis factor blockers isassociated with a lower incidence of first cardiovascular events in patients with rheumatoidarthritis. J Rheumatol 2005;32:1213–8.

58. Dixon WG, Watson KD, Lunt M et al.; British Society for Rheumatology Biologics RegisterControl Centre Consortium, Silman AJ, Symmons DP; British Society for RheumatologyBiologics Register. Reduction in the incidence of myocardial infarction in patients withrheumatoid arthritis who respond to anti-tumor necrosis factor alpha therapy: results fromthe British Society for Rheumatology Biologics Register. Arthritis Rheum2007;56:2905–12.

59. Peters MJ, Symmons DP, McCarey DW et al. Cardiovascular risk management (CV-RM) inpatients with rheumatoid arthritis and other types of inflammatory arthritis. Ann RheumDis 2008;67(Suppl II):310.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 46

LEAD

ING

ARTIC

LE


Rheumatoid arthritis (RA) is a common inflammatory diseaseof autoimmune pathogenesis. Since the 1940s, an antibodydirected against the Fc portion of immunoglobulin G (IgG),known as rheumatoid factor (RF), has been measured byvarious methods in the sera of RA patients, and has asensitivity of approximately 65–80%, but a limitedspecificity. RF may be positive in various autoimmunedisorders and chronic infections, and also in some healthysubjects, particularly in the elderly. However, RF is currentlythe only autoantibody included in the American College ofRheumatology 1987 classification criteria for the diagnosisof RA [1]. Positivity for RF has been associated with moresevere articular and extra-articular disease [2].

Serological markers other than RF have been investigatedin RA, but most have shown a poor diagnostic sensitivity andspecificity (Table 1). However, in the last decade, a new class ofautoantibodies directed against citrullinated proteins hasemerged as the most specific markers of RA identified to date,and have been shown to have important diagnostic andprognostic properties [3]. The present review discusses theautoantibodies other than RF described thus far in the sera ofpatients with RA and their clinical relevance, with particularemphasis on autoantibodies against citrullinated proteins.

Anti-citrullinated protein/peptide antibodies Anti-citrullinated protein/peptide antibodies (ACPA) are agroup of autoantibodies directed against peptides containing

citrulline, an amino acid generated by a post-translationalmodification process, which involves deimination of thenatural amino acid arginine by the enzyme peptidylargininedeiminase [4]. This group of autoantibodies is related to theantibodies recognized previously as antiperinuclear factor,antikeratin, or antifilaggrin antibodies. ACPA are regularlydetected by an enzyme-linked immunosorbent assay (ELISA)using synthetic peptides containing citrulline. The first ELISAtest using synthetic cyclic peptides derived from filaggrin(the CCP1 test) was improved to increase its sensitivity(CCP2 test) [5]. Other synthetic and non-synthetic antigenicsubstrates have been used to detect these antibodies,including citrullinated vimentin [6] and fibrin [7], and a morerecent anti-CCP assay (CCP3) has been developed [8].These tests have similar diagnostic (and probably prognostic)properties, but not the same reactivity, allowing thepossibility of discrepancies, which are observed morecommonly in non-RA samples and are due mostly todifferences in the substrates used rather than other technicalreasons [9–11].

Diagnostic value In the last 5 years, many reports have confirmed theimportant diagnostic value of ACPA. The second generationanti-CCP assay (CCP2) – the best known and the mostwidely used test – has a sensitivity similar to that of RF(60–80%), but has a higher specificity (>95%) [12].Approximately 30–40% of RF-negative RA patients testpositive in the CCP2 assay. In a recent meta-analysisincluding 37 studies of anti-CCP antibodies and 50 studieson RF, it was concluded that anti-CCP is more specific than

Serum Autoantibodies inRheumatoid Arthritis

Raimon Sanmartí, MD, Maria Victoria Hernández, MD, José A Gómez-Puerta, MD,Eduard Graell, MD, and Juan D Cañete, MD, PhDArthritis Unit, Rheumatology Service, Hospital Clinic of Barcelona, Barcelona, Spain

Various antibodies have been found in the sera of patients with rheumatoid arthritis (RA). At present, rheumatoid factor(RF) is the only autoantibody included in the American College of Rheumatology criteria for the classification of RA, but itsspecificity is limited. New autoantibodies with different degrees of sensitivity and specificity have been described in recentyears and a new class of autoantibodies directed against citrullinated peptides has also recently gained acceptance due totheir high specificity, and diagnostic and prognostic value. These, together with other autoantibodies found in RA, arediscussed in this review. Int J Adv Rheumatol 2008;6(2):47–52.

Address for correspondence: Raimon Sanmartí, Arthritis Unit,

Rheumatology Service, Hospital Clinic of Barcelona, Villarroel 170,

Barcelona, Spain. Email: [email protected]

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 47

RAIMON SANMARTÍ ET AL.


RF in the diagnosis of RA [13]. However, anti-CCPautoantibodies have been described in up to 10% ofpatients with other diseases [14], including psoriatic arthritis,Sjögren’s syndrome (SS), and systemic lupus erythematosus(SLE), emphasizing that they are not pathognomonic for RA.In these other diseases, anti-CCP may be a marker ofpersistent or more severe polyarthritis. Only in palindromicrheumatism is the frequency of anti-CCP similar to thatobserved in RA, suggesting that this disorder may beconsidered as an abortive form of RA [15].

ACPA have been shown to be useful in discriminatingthe evolution to RA (or to persistent and erosive arthritis) inpatients with early undifferentiated arthritis (UA). Thediagnostic utility of these tests in early arthritis clinics isclearly seen in a 3-year follow-up study of 318 patients withUA [16], in which 93% of patients who were anti-CCP-positive at inclusion evolved to having RA, in comparisonwith just 25% of anti-CCP-negative patients (odds ratio[OR] 38). Other studies confirm this potential predictive roleof anti-CCP, which is higher than that of RF [17]. Visser etal. found that anti-CCP was the highest-weighted factor ofseven variables that predicted persistent arthritis after 2 years of follow-up in a series of 524 patients with earlyarthritis [18].

Prognostic value Several reports have studied the predictive value of ACPA forjoint damage in RA and almost all have shown that these

autoantibodies predict a worse radiographic outcome in earlyRA. The presence of anti-CCP (determined by the CCP2 test)at baseline is associated with erosive disease and radiographicprogression to a greater extent than RF [5,12,19]. Otherstudies suggest that not only the presence, but also the levels,of these antibodies correlate with joint damage [20]. Levels ofanti-CCP seem to remain stable over the course of the disease,but may decrease (especially in those with early RA) due toantirheumatic therapy with disease-modifying antirheumaticdrugs (DMARDs) or tumor necrosis factor-α (TNF-α)antagonists, although the potential association of this decreasewith clinical response is not clear [21]. This prognostic value isalso seen in the RF-negative RA population. ACPA have notbeen associated with the frequency or severity of extra-articular features [2], although an association with serositis andpulmonary fibrosis was suggested in a Greek study [22].

Pathogenic significance The pathophysiological role of ACPA is unknown, althoughit is proven that the antibodies may be present years beforethe first clinical signs of RA [23]. In recent years, anassociation between ACPA and the rheumatoid (shared)epitope and the HLA-DRB04 genotype has beendemonstrated [24]. Smoking has also been identified as apredisposing factor for RA, but only in anti-CCP-positivepatients, and has a strong correlation with the DRB04genotype [25,26]. Citrullinated proteins have been found inthe synovium of patients with RA and other inflammatoryjoint diseases, probably as a consequence of apoptosis dueto inflammation [12,27]. However, the immunologicalresponse to the generation of ACPA seems to be veryspecific for RA, in particular for carriers of the shared epitope[10]. A direct role for ACPA in joint damage has recentlybeen suggested [28,29].

Anti-RA-33 antibodiesAnti-RA-33 antibodies, which are directed against theheterogeneous nuclear ribonucleoprotein A2 (hnRNP-A2)that is involved in mRNA transport and the regulation ofalternative splicing, have recently been identified as apotentially specific serological marker for RA [30]. They havebeen found in nearly 35% of patients with established RAbut not in control subjects [31]. In early RA, their sensitivityis approximately 30% [32], and in undifferentiatedpolyarthritis later classified as RA, anti-RA33 antibodies haveshown a higher specificity than RF [32], suggesting that theyare early antibodies in RA, and are more specific than RF.

Although anti-RA33 antibodies were initially thought tobe a useful early, specific marker for RA, they are not strictlyspecific for RA as they are also found in nearly 20% ofpatients with SLE and 40–60% of patients with mixed

Table 1. Sensitivity and specificity of autoantibodies in thesera of patients with RA according to different studies.

Autoantibodies Sensitivity (%)* Specificity (%)*

RF 65–80% 85%

ACPA 60–75% 95–98%

Anti-RA33 30–35% 96%

Anti-collagen type II 30% NE

Anti-GPI 15–64% 70%

Anti-BiP 63–73% 65–99%

Anti-calpastatin 5–83% 96%

Anti Ro (SSA) 3–15% NE

Antiphospholipid/ 12–48% NEanticardiolipin

ANCA 0–70% NE

ACPA: anti-citrullinated protein antibodies; ANCA: antineutrophil cytoplasmicantibodies; GPI: glucose-6-phosphatase isomerase; BiP: endoplasmic reticulumchaperone binding protein; NE: not established; RA: rheumatoid arthritis;RF: rheumatoid factor. *Values of sensitivity/specificity may differ according tothe cutoff values or rheumatoid or control populations used.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 48

SERUM AUTOANTIBODIES IN RA


connective tissue disease (MCTD) [33], mainly in those witherosive disease [34]. However, in SLE and MCTD, theyusually occur together with antibodies to U1-small nuclearribonucleoprotein particle (U1-snRNP) or Sm. Anti-RA33without concomitant anti-U1-snRNP antibodies have aspecificity for RA of 96% [33]. One study has found thatanti-RA-33 antibodies do not have a significant impact onradiographic outcomes [35].

Anti-collagen type II antibodiesThe possibility that an immune response directed againstjoint-related antigens such as type II collagen could play arole in the pathogenesis of RA has been hypothesized inrecent years. It has been suggested that an immuneresponse to a foreign antigen that shares some epitopespresent in type II collagen could produce antibodies not onlyto the antigen but also to the shared epitopes, producingjoint inflammation by the binding of antibodies to theepitope on the collagen cartilage. The specific expression oftype II collagen in articular cartilage and its ability to inducedestructive arthritis in animal models reinforces thishypothesis [36].

Anti-collagen type II (anti-CII) antibodies have beendemonstrated in serum and synovial fluid of approximately30% of patients with RA [37], although lower frequencieshave also been detected in patients with other rheumaticdiseases, including SLE and systemic sclerosis (SSc), amongothers. In RA patients, a direct correlation between serumand synovial fluid titers of anti-CII and acute phase reactantshas been found [38]. Anti-CII have been found mainly inearly RA [39], and seem to have the potential to destroycartilage in the early stages of RA. Besides the possibleimplication of anti-CII in the pathogenesis of RA, they couldbe a useful marker of cartilage destruction in some patientsand could offer early information, before the diseasebecomes clinically evident.

Although some reports have suggested that anti-CIIcould be predictive of a more severe outcome [40], nodifferences with controls have been found in patients beforethe onset of RA [41]. This may be because anti-CII mightonly be produced after the onset of articular inflammationcaused by other mechanisms. Thus, after breakdown ofcartilage, exposure of collagen to the immune system couldgenerate secondary antibodies against cartilage, and anti-CIIcould contribute to the perpetuation of inflammation.

Anti-GPI antibodiesAnti-GPI antibodies are directed against glucose-6-phosphate isomerase, a cytosolic enzyme that catalyzes theinterconversion of D-glucose-6-phosphate and D-fructose 6-phosphate, which are essential bodily reactions. GPI also has

cytokine and growth factor activities. Increased levels ofanti-GPI antibodies have been found in the serum andsynovial fluid of RA patients [42], with a prevalence rangingbetween 14.8% [43] and 64% [42].

Anti-GPI antibodies are not specific for RA relative toother rheumatic diseases [44,45], and no differencesbetween early and established RA have been identified[43,46]. However, increased concentrations of anti-GPI havebeen found in RA patients with extra-articularmanifestations such as nodules, vasculitis, or Felty’ssyndrome [46]. No correlation between anti-GPI and otherRA-associated antibodies, such as ACPA, has been found[43,45]. Furthermore, anti-GPI do not predict radiographicprogression in very early arthritis. Therefore, anti-GPIantibodies provide only weak discrimination of RA fromnon-RA rheumatic disorders and are not a predictive factorfor structural damage [45].

Anti-BIP antibodiesAutoantibodies against endoplasmic reticulum chaperonebinding protein (anti-BIP), also called p68, have been foundin the serum and synovial fluid of RA patients [47]. BIPstimulates synovial T cell proliferation, and its expressioncould be induced by a number of cellular stress mechanismssuch as ischemia, heat shock proteins, or cytokines [48].

The sensitivity of anti-BIP for established RA is 63–73%and the specificity is 71–99% [47,48]. In early RA patients,the sensitivity of anti-BIP is 66% and the specificity is 65%[48]. The sensitivity of anti-BIP to predict the onset of RA –before disease onset – is 45%, and the specificity 65%. Nocorrelation between anti-BIP and RF has been observed[47]. Anti-BIP antibodies have also been detected in patientswith SS [49]. Recently, a reduction in serum anti-BIP levelswas described in RA patients after biological therapy withTNF-α blockers [50]. However, no data on anti-BIP anddisease progression or severity in early RA are available.

Anti-calpastatinCalpastatin, the endogenous inhibitor of the intracellular,calcium-activated cysteine proteases calpain I and II is widelydistributed in the cytoplasmic fraction of almost allmammalian cells [51]. Elevated levels of extracellular calpainhave been reported in inflamed synovium, suggesting thatcalpain could be secreted by synovial cells and might play arole in cartilage degradation in RA [52]. Recently, variousreports have associated anti-calpastatin antibodies with RA[53], venous thrombosis [54], and autoimmune infertility[55], although epitope mapping in calpastatin remainscontroversial. Mimori et al. detected anti-calpastatinantibodies in 57% of RA patients and in lower percentagesof patients with SLE, polymyositis/dermatomyositis, or SSc,

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 49

and healthy controls [53]. Another study determined IgGanti-calpastatin antibodies in 58 RA patients and 24 subjectswith osteoarthritis (OA), 18 with SLE, 19 with SSc, and fivewith SS, using a modified ELISA [56]. IgG anti-calpastatinantibodies were found in 48 (83%) out of 58 patients withRA and in only two patients with OA. IgG anti-calpastatinantibodies are both sensitive (83%) and specific (96%) forRA. In contrast, sera from patients with other systemicrheumatic diseases showed lower sensitivities (6% in SLE,0% in SSc, and 20% in SS). Antibodies against C-terminalpeptide and calpastatin domain I have been found in only asmall proportion of RA patients and have not beenassociated with radiographic progression [57].

Recently, a strong association between anti-calpastatinantibodies, using an ELISA with purified synovial calpastatinas a substrate, and HLA-DRB alleles has been demonstrated[58]. The HLA-DRB1*0404 allele was very stronglyassociated with anti-calpastatin in RA sera. Eighty-threepercent of patients expressing both HLA-DRB1*0404 andHLA-DRB1*0401 were positive for antibodies againstsynovial calpastatin.

Anti-Ro (SSA) antibodies Anti-Ro (SSA) antibodies are associated with a variety ofautoimmune diseases including SS, SLE, subacute cutaneouslupus, neonatal lupus, SSc, and RA. The reported frequencyof anti-Ro in RA varies widely (3–15%), possibly due todifferences in methods and study groups [59–61]. Anti-Roantibodies have been associated with extra-articularmanifestations such as xerophthalmia, xerostomia, scleritis,oral ulcers, purpuric vasculitis, amyloidosis, and specificautoantibody profiles including hypergammaglobulinemia,cryoglobulins, anti-double stranded DNA, and anti-mitochondrial antibodies. Boire et al. found that patientswith anti-Ro antibodies had more severe disease and agreater requirement for immunosuppressive drugs [61],while their genetic profile showed a lower frequency ofHLA-DR4 than anti-Ro-negative patients, although theseresults were not confirmed by another study [60]. Anti-Roantibodies have also been identified as predictive markers ofpenicillamine and gold salt toxicity in RA [62].

Antiphospholipid antibodiesThe frequency of anticardiolipin antibodies (aCL) in RApatients ranges from 12% to 48% [63–66]. aCL haveshown a correlation with high levels of C-reactive proteinand repeated miscarriages, RF and antinuclear antibodies(ANA), extra-articular manifestations, nodules, andhemolytic anemia in patients with RA [63–66].

aCL have been linked with a higher risk of atherosclerosisin patients with RA. Pahor et al. evaluated internal carotid

artery intima-media thickness and antiphospholipid (aPL) ina selected group of 70 patients with RA (non-diabetic, non-hypertensive, premenopausal women) and compared themwith age- and sex-matched controls [67]. There was asignificantly higher internal carotid artery intima-mediathickness and a greater number of plaques in RA patientscompared with controls. IgG and IgM aCL were present in15.7% of RA patients compared with 5% of the controlgroup, whilst anti-β2GPI antibodies were positive in 30% ofRA patients compared with 7.5% of controls. Seriolo et al.evaluated aPL and plasma levels of protein S in 184 patientswith RA and extra-articular involvement [68]. Thirty-five(19%) had at least one type of aPL. Lupus anticoagulantwas present in seven patients with concomitant aCLpositivity. Thrombotic events were diagnosed in 34% ofaCL-positive patients with RA. Low free protein S levelswere found in 22 of 184 RA patients; 11 of those who hadlow free protein S levels and were positive for aCL. RApatients with positive aCL and a history of arterial and/orvenous thromboses showed lower levels of free protein Scompared with patients with positive aCL but no history ofthrombosis. aCL has also been found in a small proportionof patients with RA who were treated with TNF-αantagonists. The clinical implications of these drug-inducedaCL are unknown, but the majority of patients exhibit noclinical features related to antiphospholipid syndrome [69].

Antineutrophil cytoplasmic antibodiesAntineutrophil cytoplasmic antibodies (ANCA) are directedagainst lysosomal enzymes of human neutrophils andmonocytes and have been identified in various vasculitides,especially in Wegener’s granulomatosis and microscopicpolyangiitis. ANCA (perinuclear ANCA [pANCA] andcytoplasmic ANCA [cANCA]) have been detected in the seraof 0–70% of RA patients [70–72]. ANCA antibodies from RApatients recognize different antigens in the nucleus andcytoplasm, of which lactoferrin is the most common [73]. Ithas been suggested that ANCA occur especially in RApatients with longstanding, severe disease who are positivefor RF and ANA. An association between ANCA andvasculitic and pulmonary involvement has also beenproposed [70,71]. Mustila et al. evaluated the prevalence ofANCA in patients with early (<12 months) RA [70]. pANCAwere found in 40 (50%) patients and atypical cANCA inthree (4%) patients at study entry. pANCA were significantlymore frequent in RF-positive patients. There was nocorrelation between ANCA and clinical disease activity.During follow-up, radiographic erosions (Larsen score)advanced more rapidly in pANCA-positive patients. Anti-myeloperoxidase (MPO), a pANCA-specific antibody, wasdetermined in 97 patients with RA. Anti-MPO were detected



RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 50

SERUM AUTOANTIBODIES IN RA


in 12 (12%) patients and were associated with nodules, lunginvolvement, and a higher swollen joint count [72].

ConclusionSince the first description of RF >50 years ago, variousautoantibodies have been found in the sera of RA patients.The majority of these have been shown to be moreprevalent in RA than in other rheumatic diseases or healthysubjects although their sensitivity and specificity (Table 1)have consistently been shown to be lower than that of RF,and few appear to have prognostic significance. Variableresults have been obtained for some of theseautoantibodies, such as those directed against calpastatin,owing to technical difficulties and the characteristics of theantigenic substrate used.

However, in the last decade, ACPA have been identifiedas the most specific serological marker of RA, withinteresting diagnostic and prognostic properties. Theseautoantibodies are now a useful tool for the diagnosis of RAin clinical practice and their role in the disease process andjoint damage merits further investigation.

DisclosuresThe authors have no relevant financial interests to disclose.

References1. Arnett FC, Edworthy SM, Bloch DA et al. The American Rheumatism Association 1987

revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum1988;31:315–24.

2. De Rycke L, Peene I, Hoffman IE et al. Rheumatoid factor and anticitrullinated proteinantibodies in rheumatoid arthritis: diagnostic value, associations with radiologicalprogression rate, and extra-articular manifestations. Ann Rheum Dis 2004;63:1587–93.

3. van Venrooij WJ, Zendman AJ, Pruijn GJ. Autoantibodies to citrullinated antigens in (early)rheumatoid arthritis. Autoimmun Rev 2006;6:37–41.

4. Schellekens GA, de Jong BA, van den Hoogen FH et al. Citrulline is an essential constituentof antigenic determinants recognized by rheumatoid arthritis-specific autoantibodies. J Clin Invest 1998;101:273–81.

5. van Gaalen FA, Viser H, Huizinga TW. A comparison of the diagnostic accuracy andprognostic value of the first and second anti-cyclic citrullinated peptides autoantibody(CCP1 and CCP2) test for rheumatoid arthritis. Ann Rheum Dis 2005;64:1510–2.

6. Mathsson L, Mullazehi M, Wick MC et al. Antibodies against citrullinated vimentin inrheumatoid arthritis: higher sensitivity and extended prognostic value concerning futureradiographic progression as compared with antibodies against cyclic citrullinated peptides.Arthritis Rheum 2008;58:36–45.

7. Pérez ML, Gómara MJ, Ercilla G et al. Antibodies to citrullinated human fibrinogensynthetic peptides in diagnosing rheumatoid artritis. J Med Chem 2007;50:3573–84.

8. Luis Caro-Oleas J, Fernández-Suárez A, Reneses Cesteros S et al. Diagnostic usefulness ofa third-generation anti-cyclic citrulline antibody test in patients with recent-onsetpolyarthritis. Clin Chem Lab Med 2007;45:1396–401.

9. Vander Cruyssen B, Nogueira L, van Praet J et al. Do all anti-citrullinated protein/peptideantibody (ACPA) tests measure the same? Evaluation of discrepancy between anti-citrullinated protein/peptide antibody tests in patients with and without rheumatoidarthritis. Ann Rheum Dis 2008;67:542–6.

10. Bizzaro N, Tonutti E, Tozzoli R et al. Analytical and diagnostic characteristics of 11 2nd-and 3rd-generation immunoenzymatic methods for the detection of antibodies tocitrullinated proteins. Clin Chem 2007;53:1527–33.

11. Coenen D, Verschueren P, Westhovens R et al. Technical and diagnostic performance of 6assays for the measurement of citrullinated protein/peptide antibodies in the diagnosis ofrheumatoid arthritis. Clin Chem 2007;53:498–504.

12. Vossenaar ER, Walther J, van Venrooij. Anti-CCP antibodies, a highly specific marker for(early) rheumatoid arthritis. Clin Applied Immunol Rev 2004;4:239–62.

13. Nishimura K, Sugiyama D, Kogata Y et al. Meta-analysis: diagnostic accuracy of anti-cycliccitrullinated peptide antibody and rheumatoid factor for rheumatoid arthritis. Ann InternMed 2007;146:797–808.

14. Fabien N, Olsson NO, Goetz J et al. Prevalence of autoantibodies to cyclic citrullinatedpeptide in patients with rheumatic diseases other than rheumatoid arthritis: a Frenchmulticenter study. Clin Rev Allergy Immunol 2008;34:40–4.

15. Sanmartí R, Cañete JD, Salvador G. Palindromic rheumatism and other relapsing arthritis.Best Pract Res Clin Rheumatol 2004;18:647–61.

16. van Gaalen FA, Linn-Rasker SP, van Venrooij WJ et al. Autoantibodies to cyclic citrullinatedpeptides predict progression to rheumatoid arthritis in patients with undifferentiatedarthritis: a prospective cohort study. Arthritis Rheum 2004;50:709–15.

17. Vallbracht I, Rieber J, Oppermann M et al. Diagnostic and clinical value of anti-cycliccitrullinated peptide antibodies compared with rheumatoid factor isotypes in rheumatoidarthritis. Ann Rheum Dis 2004;63:1079–84.

18. Visser H, le Cessie S, Vos K et al. How to diagnose rheumatoid arthritis early: a predictionmodel for persistent (erosive) arthritis. Arthritis Rheum 2002;46:357–65.

19. Forslind K, Ahlmén M, Eberhardt K et al. Prediction of radiological outcome in earlyrheumatoid arthritis in clinical practice: role of antibodies to citrullinated peptides (anti-CCP). Ann Rheum Dis 2004;63:1090–5.

20. Bongi SM, Manetti R, Melchiorre D et al. Anti-cyclic citrullinated peptide antibodies arehighly associated with severe bone lesions in rheumatoid arthritis anti-CCP and bonedamage in RA. Autoimmunity 2004;37:495–501.

21. Bobbio-Pallavicini F, Caporali R, Alpini C et al. Predictive value of antibodies to citrullinatedpeptides and rheumatoid factors in anti-TNF-alpha treated patients. Ann N Y Acad Sci2007;1109:287–95.

22. Alexiou I, Germenis A, Koutroumpas A et al. Anti-cyclic citrullinated peptide-2 (CCP2)autoantibodies and extra-articular manifestations in Greek patients with rheumatoidarthritis. Clin Rheumatol 2008;27:511–3.

23. Nielen MM, van Schaardenburg D, Reesink HW et al. Specific autoantibodies precede thesymptoms of rheumatoid arthritis: a study of serial measurements in blood donors.Arthritis Rheum 2004;50:380–6.

24. van Gaalen FA, van Aken J, Huizinga TW et al. Association between HLA class II genesand autoantibodies to cyclic citrullinated peptides (CCPs) influences the severity ofrheumatoid arthritis. Arthritis Rheum 2004;50:2113–21.

25. Klareskog L, Stolt P, Lundberg K et al. A new model for an etiology of rheumatoid arthritis:smoking may trigger HLA-DR (shared epitope)-restricted immune reactions toautoantigens modified by citrullination. Arthritis Rheum 2006;54:38–46.

26. van der Helm-van Mil AH, Verpoort KN, le Cessie S et al. The HLA-DRB1 shared epitopealleles differ in the interaction with smoking and predisposition to antibodies to cycliccitrullinated peptide. Arthritis Rheum 2007;56:425–32.

27. Makrygiannakis D, af Klint E, Lundberg IE et al. Citrullination is an inflammation-dependent process. Ann Rheum Dis 2006;65:1219–22.

28. Kuhn KA, Kulik L, Tomooka B et al. Antibodies against citrullinated proteins enhance tissueinjury in experimental autoimmune arthritis. J Clin Invest 2006;116:961–73.

29. Hill JA, Bell DA, Brintnell W et al. Arthritis induced by posttranslationally modified(citrullinated) fibrinogen in DR4-IE transgenic mice. J Exp Med 2008;205:967–79.

30. Steiner G, Hartmuth K, Skriner K et al. Purification and partial sequencing of the nuclearautoantigen RA33 shows that it is indistinguishable from the A2 protein of theheterogeneous nuclear ribonucleoprotein complex. J Clin Invest 1992;90:1061–6.

31. Hassfeld W, Steiner G, Hartmuth K et al. Demonstration of a new antinuclear antibody (anti-RA33) that is highly specific for rheumatoid arthritis. Arthritis Rheum 1989;32:1515–20.

32. Hassfeld W, Steiner G, Graninger W et al. Autoantibody to the nuclear antigen RA33: amarker for early rheumatoid arthritis. Br J Rheumatol 1993;32:199–203.

33. Hassfeld W, Steiner G, Studnicka-Benke A et al. Autoimmune response to the spliceosome.An immunologic link between rheumatoid arthritis, mixed connective tissue disease andsystemic lupus erythematosus. Arthritis Rheum 1995;38:777–85.

34. Mediwake R, Isenberg DA, Schellekens GA et al. Use of anti-citrullinated peptide and anti-RA33 antibodies in distinguishing erosive arthritis in patients with systemic lupuserythematosus and rheumatoid arthritis. Ann Rheum Dis 2001;60:67–8.

35. Combe B, Dougados M, Goupille P et al. Prognostic factors for radiographic damage inearly rheumatoid arthritis. Arthritis Rheum 2001;44:1736–43.

36. Trentham DE, Townes AS, Kang AH. Autoimmunity to type II collagen an experimentalmodel of arthritis. J Exp Med 1977;146:857–68.

37. Clague RB. Autoantibodies to cartilage collagens in rheumatoid arthritis. Do theyperpetuate the disease or are they irrelevant? Br J Rheumatol 1989;28:1–5.

38. Kim WU, Yoo WH, Park W et al. IgG antibodies to type II collagen reflect inflammatoryactivity in patients with rheumatoid arthritis. J Rheumatol 2000;27;575–81.

39. Mullazehi M, Mathsson L, Lampa J et al. High anti-collagen type-II antibody levels andinduction of proinflammatory cytokines by anti-collagen antibody-containing immunecomplexes in vitro characterise a distinct rheumatoid arthritis phenotype associated withacute inflammation at the time of disease onset. Ann Rheum Dis 2007;66:537–41.

40. Cook AD, Rowley MJ, Mackay IR et al. Antibodies to type II collagen in early rheumatoidarthritis. Correlation with disease progression. Arthritis Rheum 1996;39:1720–7.

41. Möttönen T, Hannonen P, Oka M et al. Antibodies against native type II collagen do notprecede the clinical onset of rheumatoid arthritis. Arthritis Rheum 1988;31:776–9.

42. Schaller M, Stohl W, Tan SM et al. Raised levels of anti-glucose-6-phosphate isomeraseIgG in serum and synovial fluid from patients with inflammatory arthritis. Ann Rheum Dis2005;64:743–9.

43. Matsumoto I, Lee DM, Goldbach-Mansky R et al. Low prevalence of antibodies toglucose-6-phosphate isomerase in patients with rheumatoid arthritis and a spectrum ofother chronic autoimmune disorders. Arthritis Rheum 2003;48:944–54.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 51

44. Herve CA, Wait R, Venables PJ. Glucose-6-phosphate isomerase is not a specificautoantigen in rheumatoid arthritis. Rheumatology (Oxford) 2003;42:986–8.

45. Jouen F, Vittecoq O, Leguillou F et al. Diagnostic and prognostic values of anti glucose-6-phosphate isomerase antibodies in community-recruited patients with very early arthritis.Clin Exp Immunol 2004;137:606–11.

46. van Gaalen FA, Toes RE, Ditzel HJ et al. Association of autoantibodies to glucose-6-phosphate isomerase with extraarticular complications in rheumatoid arthritis. ArthritisRheum 2004;50:395–9.

47. Bläss S, Union A, Raymackers J et al. The stress protein BIP is overexpressed and is a majorB and T cell target in rheumatoid arthritis. Arthritis Rheum 2001;44:761–71.

48. Bodman-Smith MD, Corrigall VM, Berglin E et al. Antibody response to the human stressprotein BIP in rheumatoid arthritis. Rheumatology (Oxford) 2004;43:1283–7.

49. Bodman-Smith MD, Corrigal VM, Chan C et al. Anti-BIP antibodies in the serum ofpatients with autoimmune disease. Immunology 2003;107(Suppl 1):OP203.

50. Mavropoulos JC, Cuchacovich M, Llanos C et al. Anti-tumor necrosis factor-alpha therapyaugments dipeptidyl peptidase IV activity and decreases autoantibodies to GRP78/BIP andphosphoglucose isomerase in patients with rheumatoid arthritis. J Rheumatol2005;32:2116–24.

51. Rantapää-Dahlqvist S. Diagnostic and prognostic significance of autoantibodies in earlyrheumatoid arthritis. Scand J Rheumatol 2005;34:83–96.

52. Yamamoto S, Shimizu K, Shimizu K et al. Calcium-dependent cysteine proteinase (calpain)in human arthritic synovial joints. Arthritis Rheum 1992;35:1309–17.

53. Mimori T, Suganuma K, Tanami Y et al. Autoantibodies to calpastatin (an endogenousinhibitor for calcium-dependent neutral protease, calpain) in systemic rheumatic diseases.Proc Natl Acad Sci USA 1995;92:7267–71.

54. Schlosser U, Lackner KJ, Scheckenhofer C et al. Autoantibodies against the proteaseinhibitor calpastatin: a new risk factor for venous thrombosis? Thromb Haemost1997;77:11–3.

55. Wang LF, Wei SG, Miao SY et al. Calpastatin gene in human testis. Biochem Mol Biol Int1994;33:245–51.

56. Iwaki-Egawa S, Matsuno H, Yudoh K et al. High diagnostic value of anticalpastatinautoantibodies in rheumatoid arthritis detected by ELISA using human erythrocytecalpastatin as antigen. J Rheumatol 2004;31:17–22.

57. Vittecoq O, Pouplin S, Krzanowska K et al. Rheumatoid factor is the strongest predictor ofradiological progression of rheumatoid arthritis in a three-year prospective study incommunity-recruited patients. Rheumatology (Oxford) 2003;42:939–46.

58. Auger I, Roudier C, Guis S et al. HLA-DRB1*0404 is strongly associated withanticalpastatin antibodies in rheumatoid arthritis. Ann Rheum Dis 2007;66:1588–93.

59. Cavazzana I, Franceschini F, Quinzanini M et al. Anti-Ro/SSA antibodies in rheumatoidarthritis: clinical and immunologic associations. Clin Exp Rheumatol 2006;24:59–64.

60. Schneeberger E, Citera G, Heredia M et al. Clinical significance of anti-Ro antibodies inrheumatoid arthritis. Clin Rheumatol 2008;27:517–9.

61. Boire G, Ménard HA, Gendron M et al. Rheumatoid arthritis: anti-Ro antibodies define anon-HLA-DR4 associated clinicoserological cluster. J Rheumatol 1993;20:1654–60.

62. Tishler M, Nyman J, Wahren M et al. Anti-Ro (SSA) antibodies in rheumatoid arthritispatients with gold-induced side effects. Rheumatol Int 1997;17:133–5.

63. Fort JG, Cowchock FS, Abruzzo JL et al. Anticardiolipin antibodies in patients withrheumatic diseases. Arthritis Rheum 1987;30:752–60.

64. Keane A, Woods R, Dowding V et al. Anticardiolipin antibodies in rheumatoid arthritis. Br JRheumatol 1987;26:346–50.

65. Wolf P, Gretler J, Aglas F et al. Anticardiolipin antibodies in rheumatoid arthritis: theirrelation to rheumatoid nodules and cutaneous vascular manifestations. Br J Dermatol1994;131:48–51.

66. Merkel PA, Chang YC, Pierangeli SS et al. The prevalence and clinical associations ofanticardiolipin antibodies in a large inception cohort of patients with connective tissuediseases. Am J Med 1996;101:576–83.

67. Pahor A, Hojs R, Holc I et al. Antiphospholipid antibodies as a possible risk factor foratherosclerosis in patients with rheumatoid arthritis. Immunobiology 2006;211:689–94.

68. Seriolo B, Accardo S, Garnero A et al. Anticardiolipin antibodies, free protein S levels andthrombosis: a survey in a selected population of rheumatoid arthritis patients.Rheumatology (Oxford) 1999;38:675–8.

69. Bobbio-Pallavicini F, Alpini C, Caporali R et al. Autoantibody profile in rheumatoid arthritisduring long-term infliximab treatment. Arthritis Res Ther 2004;6:R264–72.

70. Mustila A, Paimela L, Leirisalo-Repo M et al. Antineutrophil cytoplasmic antibodies inpatients with early rheumatoid arthritis: an early marker of progressive erosive disease.Arthritis Rheum 2000;43:1371–7.

71. Tur BS, Süldür N, Ataman S et al. Anti-neutrophil cytoplasmic antibodies in patients withrheumatoid arthritis: clinical, biological, and radiological correlations. Joint Bone Spine2004;71:198–202.

72. Cambridge G, Williams M, Leaker B et al. Anti-myeloperoxidase antibodies in patients withrheumatoid arthritis: prevalence, clinical correlates, and IgG subclass. Ann Rheum Dis1994;53:24–9.

73 Brimnes J, Halberg P, Jacobsen S et al. Specificities of anti-neutrophil autoantibodies inpatients with rheumatoid arthritis (RA). Clin Exp Immunol 1997;110:250–6.



RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 52

CA

SE STUD

Y


Case A 45-year-old woman of Moroccan descent was admitted to the present authors’ hospital with persistent convulsions.She had a history of systemic lupus erythematosus (SLE) based on the presence of arthritis, discoid skin lesions, positive antinuclear antibodies (ANA), antibodies todouble-stranded DNA, and a positive Coombs test. In 1998,she suffered an intra-uterine fetal death due toantiphospholipid syndrome and in 2005, she was treated for biopsy-proven lupus nephritis (World HealthOrganization grade V). In 2006, she had a middle cerebralartery stroke with a grade four mitral valve insufficiency due to Libman–Sacks endocarditis; this resulted in left-sided spastic hemiparesis and a relapse of SLEglomerulonephritis for which she was treated with high-dose methylprednisolone and azathioprine. After the stroke the subject suffered epileptic fits for which valproic acid was started.

The current admission was due to sudden onset ofgeneralized convulsions preceded by 1 day of progressiveabdominal pain. There had been no fever or signs ofinfection, or neurological deficit. On admission, hermedication consisted of phenprocoumon, prednisone 5 mgtwice daily, azathioprine 50 mg three times daily,hydroxychloroquine 400 mg once daily, alendronic acid 70mg per week, valproid acid 500 mg twice daily, amlodipine10 mg once daily, peridonpril 8 mg once daily, atorvastatin10 mg once daily, and pantoprazole 20 mg once daily.

ExaminationOn admission, the patient had already received clonazepamfor generalized convulsions. She was unconscious with aGlasgow coma scale of 8. Her blood pressure was 180/100mmHg, pulse rate 130 beats/min (bpm), and hertemperature was 39.1ºC. Auscultation of the heart revealednormal heart tones with a murmur consistent with theknown mitral valve insufficiency. Examination of the lungsand abdomen revealed no abnormalities. Neurologicalexamination revealed no signs of meningitis or asymmetricdeficit; however, the head and eyes were fixed to the left.

The patient’s laboratory tests on admission showed anormal hemoglobin level (8.3 mmol/L), white blood cellcount (12.6x109/L), and platelet count (170x109/L), anerythrocyte sedimentation rate of 31 mm/h, C-reactiveprotein level of 3 mg/L, and an international normalizedratio (INR) of 4.0. Her sodium concentration was slightlydecreased (133 mmol/L) and the elevated creatinine level(due to her chronic renal insufficiency) remained stable at146 μmol/L. Liver enzymes were unremarkable. Spinal tapopening pressure was normal, and the spinal fluid leukocytecount was 2.0 cells per 3 μL with a total proteinconcentration of 1.19 g/L. Blood, urine, sputum, and spinalfluid cultures were all negative for abnormalities.

The electrocardiogram showed a sinus tachycardia of120 bpm without abnormalities. Chest radiography andultrasound of the abdomen also revealed no irregularities. Acardiac ultrasound did not indicate cardiac embolism oractive Libman–Sachs endocarditis. Computed tomography(CT) scanning and magnetic resonance imaging (MRI; bothstandard and with Gadolinium contrast) of the brainrevealed only old lesions consistent with the previous

Just Another Case of NeuropsychiatricSystemic Lupus Erythematosus?

Case Study presented by: Leroy R Lard, MD, PhD1, Darius Soonawala, MD2,Abraham Schoe, MD2, and Irene Speyer, MD3

1Departments of Rheumatology and 2Intensive Care Unit, Leiden University Medical Centre, Leiden, and3Department of Rheumatology, Bronovo Hospital, Hague, The Netherlands

Case Study Editor: Cornelia F Allaart, MD, PhDDepartment of Rheumatology, Leiden University Medical Centre, Leiden, The Netherlands

Int J Adv Rheumatol 2008;6(2):53–6.

Address for correspondence: LR Lard, Department of Rheumatology,

Leiden University Medical Centre, PO Box 9600, 2300 RC, Leiden,

The Netherlands. Email: [email protected]

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 53

LEROY R LARD ET AL.


ischemic stroke and old brain white matter lesions centrallyon both sides. No tumor or hematoma was found on thebrain scans. Thus far, no obvious causes for the persistentconvulsions and the fever were identified.

Management and further progressAt this point, our differential diagnosis was convulsions due toher past stroke currently triggered by high fever of unknown

infectious origin during immunosuppressive therapy, or byactive SLE. Clonazepam and carbamazepine were administeredintravenously and broad spectrum antibiotics were started.Over the next few days her fever subsided and she was awakeand responsive. The laboratory results remained unchanged.One week later she developed a flaccid paresis of the right armand leg together with phatic problems. New MRI of the brainand cervical spine (standard and with Gadolinium contrast)showed no signal abnormalities, no mass in the cervical cord,unchanged abnormalities in the brain, and no new ischemiclesions. This could neither prove nor rule out neuropsychiatricSLE (NPSLE). Due to the clinical deterioration in spite of thecurrent treatment, active NPSLE was considered the most likelydiagnosis. Therefore, methylprednisolone 1000 mg/day for3 days was initiated, followed by 60 mg/day of prednisone.The anti-epileptic drugs were continued intravenously. Initially,a mild clinical improvement seemed to occur; however, duringthe next 2 days she had two more seizures. The paresisprogressed to flaccid tetraparesis, soon followed by a criticalrespiratory failure for which an intensive care unit (ICU)admission was needed for mechanical ventilation. Afterexclusion of vascular or infectious causes, the neurologicaldeterioration was thought to be due to active NPSLE in theform of Guillain–Barré syndrome (GBS). However, serial nerveconduction studies showed axonal rather than demyelinatingfindings, and needle electromyography showed nospontaneous activity. In other words, the typical GBS patternwas not observed. Nevertheless, treatment was started withintravenous immunoglobulin and cyclophosphamide pulsetherapy (750 mg/m2) [1]; prednisone 60 mg/day and the anti-epileptic drugs were continued. Unfortunately, noimprovement was observed.

In the ICU, the patient’s urine was noted to have anorange/reddish color, and at that moment acute porphyria (AP)was considered. Testing revealed elevated urine levels of totalporphyria of 24 306 nmol/24 h (normal <210 nmol/24 h),porphobilinogen of 269 μmol/24 h (normal <9 μmol/24 h),and δ-aminolevulinic acid (δ-ALA) of 300 μmol/24 h (normal<53 μmol/24 h). Fecal coproporphyrin levels were significantlyelevated, with 86 nmol/g coproporphyrin I (normal <5.4 nmol/g) and 691 nmol/g coproporphyrin III (normal <2 nmol/g). This led to the diagnosis of an AP attack due tohereditary coproporphyria (HCP) and not that of NPSLE. All ofthe symptoms, namely the preceding abdominal pain, theconvulsions, fever, hypertension, sinus tachycardia, tetraparesis,and respiratory failure, could be attributed to the AP (Table 1)[2]. The initial attack may have been caused by the use ofmedication. Valproic acid and calcium channel blockers arecandidates for triggering an attack (Table 2) [2]. The clinicalcourse deteriorated by further administration of clonazepamand phenytoin, which are known to be porphyrinogenic drugs.

Table 1. Clinical manifestations of AP with the estimatedincidence [2].

Clinical manifestations Incidence (%)

Abdominal pain 85–95

Vomiting 43–88

Constipation 48–84

Diarrhea 5–12

Pain in extremities, back, chest, neck, or head 50–70

Paresis 42–68

Respiratory paralysis 9–20

Mental symptoms 40–58

Convulsions 10–20

Tachycardia 28–85

Hypertension 36–55

Fever 9–37

AP: acute porphyria.

Table 2. Common drugs that are unsafe and safe in patientswith porphyria [2].

Unsafe Safe

Alcohol Acetaminophen

Barbiturates Aspirin

Carbamazepine Gabapentin

Clonazepam Glucocorticoids

Diclofenac Insulin

Estrogens Narcotic analgesics

Phenytoin Penicillin

Progesterone Atropine

Calcium channel blockers Phenothiazines

Sulfonamide antibiotics Ranitidine

Valproic acid Streptomycin

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 54

CASE STUDY: JUST ANOTHER CASE OF NEUROPSYCHIATRIC SLE


Following the diagnosis of HCP, all porphyrinogenicmedication was stopped and hematin was startedintravenously [2]. Unfortunately, awaiting possible beneficialresponse, the patient developed cyclophosphamide-inducedleucopenia, making her susceptible to infections. The patientsuffered from multiple episodes of ventilator-associatedpneumonia and ultimately died due to a Gram-negative sepsis.

DiscussionHCP is one of four types of AP. It is a rare, autosomal-dominant metabolic disorder of the production of heme –the oxygen-binding prosthetic group of hemoglobin. It ischaracterized by a deficiency of the enzymecoproporphyrinogen oxidase (CPO). Heme synthesis beginsin the mitochondrion, proceeds into the cytoplasm, and endsin the mitochondrion (Fig. 1). Without CPO, heme synthesiscannot progress, and the metabolite porphyrin accumulatesin the cytoplasm, which is toxic. Accumulation of thesemetabolites causes the characteristic neurovisceral and/orphotosensitizing symptoms. The exact mechanism leading tothese symptoms remains unclear. Even with the geneticdisorder, in unchallenged conditions there still is a 50% CPOenzyme activity, which is sufficient to prevent an acute

attack of porphyria. However, infections, certain hormonesand drugs, and dietary changes can trigger an attack of APby inducing the synthesis of cytochrome P450, whichconsequently increases the heme oxygenase activity. Thisleads to an abnormal accumulation of the toxic porphyrins.The most common known drugs that can trigger AP areanti-epileptic drugs such as valproic acid, phenytoin, andclonazepam, which were administrated to our patient.

Treatment of mild AP attacks consists of a highcarbohydrate diet or a glucose 10% infusion (≥300 g daily),which reduces the activity of heme oxygenase. In severe APattacks, hematin must also be given intravenously at adosage of 3–4 mg/kg per day for ≥4 days [2,3]. These arenot curative drugs, but they can shorten attacks and reducethe intensity of an attack by inhibiting ALA synthase andthus the accumulation of toxic precursors. Side effects arerare but can be serious, for instance severe coagulopathyand anaphylactic reactions. Discontinuing porphyrinogenicdrugs is essential. As many anti-epileptic drugs exacerbatethis condition, the treatment of convulsions in these patientsis difficult. Some benzodiazepines are safe, and, when usedin conjunction with gabapentin, offer a possible regimen forconvulsion control [2,4].

Figure 1. Schematic representation of the heme synthesis. AP due to HCP is caused by a deficiency of the enzymecoproporphyrinogen oxidase.

AP: acute porphyria; CoA: co-enzyme A; HCP: hereditary coproporphyria.

Glycine + Succinyl Co-A D-aminolaevulinic acid

Coproporphyrinogen oxidase

Protoporphyrinogen IX

Protoporphyrinogen oxidase

Protoporphyrin IX

Heme

Ferrochelatase

Aminolaevulinate dehydratase

Prophobilinogen

Hydroxymethylbilane

Uroporphyrinogen III

Coproporphyrinogen III

Prophobilinogen deaminase

Uroporphyrinogen III synthase

Uroporphyrinogen decarboxylase

CytoplasmMitochondria

Aminolaevulinic acid synthase

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 55

Several reports have described the co-appearance of(cutaneous) porphyria in patients with SLE [5,6]. Allard et al.even describe a high prevalence of presence of ANA inpatients with AP [6]. However, this appears to be anunfortunate coincidence rather than a specificpathophysiological relationship. As the most commonsymptoms of AP (abdominal pain, vomiting, tachycardia,and fever) are not especially specific, and because SLE was alikely cause of the neurological symptoms in this case,porphyria was not initially considered in this patient. Manysymptoms could be attributed to an active NPSLE. Indeed,the lack of new lesions on a cerebral MRI does not excludean active NPSLE [7]. However, the poor response to high-dose immunosuppressive therapy, and the electromyographythat was not consistent with typical GBS, should havetriggered suspicions of an alternative, even more rare,diagnosis, at an earlier stage. The patient might still haveresponded to the treatment of AP; however, in themeantime the NPSLE treatment had made her even morevulnerable to the infectious complications, which ultimatelyresulted in her death. AP should be included in thedifferential diagnosis of atypical NPSLE, especially whenpresenting with acute neuropathy.

Case Study Editor’s commentsThis is a report of the unfortunate result of a missed diagnosis.It highlights that making a differential diagnosis needs to gobeyond the obvious and most likely, and should include, evenif as a final thought, the rare, the unlikely, and the relatively

unknown diseases, especially if the response to treatment isnot what may have been expected. It is also an illustration ofthe hazard of the doctor’s detachment/dissociation from thepatient’s bedside. Advanced technical tests may be consideredfor diagnoses and response to treatment, but in this case, theunusual color of the urine went unnoticed until the patient hada urinary catheter in the ICU. As one of the physicians involvedwith the care for this patient, I am entitled to be critical. Sincethis case occurred, we have considered AP and tested for it intwo other patients with atypical manifestations of NPSLE. Theyboth tested negative, and the diagnosis of NPSLE was finallyconfirmed. Given the rarity of AP, I hope that we will notforget this case when the next patient is admitted in severalyears time.

DisclosuresThe authors have no relevant financial interests to disclose.

References1. Gourley MF, Austin HA 3rd, Scott D et al. Methylprednisolone and cyclophosphamide,

alone or in combination, in patients with lupus nephritis. A randomized, controlled trial.Ann Intern Med 1996;125:549–57.

2. Anderson KE, Bloomer JR, Bonkovsky HL et al. Recommendations for the diagnosis andtreatment of the acute porphyrias. Ann Intern Med 2005;142:439–50.

3. Kochar DK, Pal M, Kochar SK et al. Acute intermittent porphyria presenting withneurological emergency: review of six cases. Neurol India 2007;55:413–5.

4. Kauppinen R. Porphyrias. Lancet 2005;365:241–52.

5. Filiotou A, Vaiopoulos G, Capsimali V et al. Acute intermittent porphyria and systemiclupus erythematosus: report of a case and review of the literature. Lupus 2002;11:190–2.

6. Allard SA, Charles PJ, Herrick AL et al. Antinuclear antibodies and the diagnosis of systemiclupus erythematosus in patients with acute intermittent porphyria. Ann Rheum Dis1990;49:246–8.

7. Jennings JE, Sundgren PC, Attwood J et al. Value of MRI of the brain in patients with systemiclupus erythematosus and neurologic disturbance. Neuroradiology 2004;46:15–21.

LEROY R LARD ET AL.


RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 56


CLINICAL REVIEWSCommentary and Analysis on Recent Key Papers

Clinical reviews were prepared by Tom Huizinga, MD, Peter Nigrovic, MD, Eric Ruderman, MD, and Hendrik Schulze-Koops, MD

TREATMENT STRATEGIES

Long-term follow-up results after autologoushaematopoietic stem cell transplantation for severesystemic sclerosisVonk MC, Marjanovic Z, van den Hoogen FH et al.Ann Rheum Dis 2008;67:98–104.

Since 1996, approximately 140 patients with systemic sclerosis(SSc) have undergone hematopoietic stem cell transplantation(HSCT). Early results of Phase I/II clinical trials showed thatHSCT is feasible in carefully selected patients with diffuse SSc.A follow-up study of 57 patients revealed a positive responseat 3 years in about 66% of the patients with a mortality rateof 8.7% [1]. Based on these data, the present authorsinitiated an analysis of the long-term follow-up results of theDutch and French patients from two Phase I/II trials.

The specific aims of this analysis were to evaluate thesurvival rates and the durability of responses at up to 7 years offollow-up in SSc patients treated with HSCT. A total of 26patients were followed up. The mobilization and collection ofperipheral blood stem cells (PBSC) was performed with 4 g/m2

of cyclophosphamide followed by human granulocyte-colonystimulating factor (G-CSF) 4–5 days later. At least 9.5 millionCD34+ cells per kg had to be collected by successive dailyapheresis to obtain 7 million stem cells after positive selection.Conditioning was performed at least 4 weeks later usingcyclophosphamide at 50 mg/kg/day from day –5 to day –2prior to PBSC injection. To assess the survival and diseaseresponse to treatment, clinical evaluation was conducted at

least every 6 months. After a median follow-up of 5.3 years(range 1–7.5 years), 81% (21 of 26) of the patientsdemonstrated a clinically beneficial response. The Kaplan–Meierestimated survival rate at 5 years was 96.2% (95% confidenceinterval [CI] 89–100%) and at 7 years, 84.8% (95% CI70.2–100%). The event-free survival rate, defined as survivalwithout mortality, relapse, or progression of SSc resulting inmajor organ dysfunction, was 64.3% (95% CI 47.9–86%) at 5years and 57.1% (95% CI 39.3–83%) at 7 years.

This study confirms that autologous HSCT in selectedpatients with severe diffuse SSc results in sustainedimprovement of skin thickening and stabilization of organfunction up to 7 years after transplantation.1. Farge D, Passweg J, van Laar JM et al.; EBMT/EULAR Registry. Autologous stem cell

transplantation in the treatment of systemic sclerosis: report from the EBMT/EULARRegistry. Ann Rheum Dis 2004;63:974–81.

Address for reprints: MC Vonk, Department of Rheumatology, RadboudUniversity Nijmegen Medical Centre, PO Box 9101, 6500 HB Nijmegen,The Netherlands. Email: [email protected]

The comparative one-year performance of anti-tumor necrosis factor αα drugs in patients withrheumatoid arthritis, psoriatic arthritis, andankylosing spondylitis: results from a longitudinal,observational, multicenter studyHeiberg MS, Koldingsnes W, Mikkelsen K et al.Arthritis Rheum 2008;59:234–40.

Therapy with anti-tumor necrosis factor (anti-TNF) agents isoften highly successful in patients with rheumatic illnesses, but

What is the “survival” of anti-tumor necrosis factor therapyin clinical practice, and what are the predictors of successfullong-term use of such therapy? The authors address thisquestion in an observational Norwegian cohort and findthat patients with rheumatoid arthritis tend to require achange in regimen sooner than patients with psoriaticarthritis or ankylosing spondylitis. Concomitantmethotrexate use was associated with greater longevityof therapy.

Despite modern therapies, the high mortality rate inpatients with systemic sclerosis (SSc; a 5-year mortality rateof 30%) requires the development of novel therapeuticapproaches. In the present study, the authors report theresults of intensive myelo- and immunosuppressionfollowed by autologous hematopoietic stem celltransplantation to treat severe SSc.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 57

CLINICAL REVIEWS


not all patients remain on therapy in the long-term. Reasonsfor discontinuation include inefficacy and adverse events,among others.

The present group used a Norwegian disease-modifyingantirheumatic drug registry to track the course of >1200courses of anti-TNF therapy to assess factors associated withearly discontinuation. They found that the underlyingdiagnosis was a powerful predictive factor, such that patientswith rheumatoid arthritis (RA) had significantly reducedtreatment longevity compared with patients with psoriaticarthritis (PsA) or ankylosing spondylitis (AS). The crude 1-year drug survival rates were 65.4%, 77.3%, and 77.5%respectively (although the difference between PsA and RAlost significance when adjusted for potential confounders).Among the three drugs (adalimumab, etanercept, andinfliximab), etanercept enjoyed the best crude survival rates,but this drug was often the first that was tried, and thesurvival advantage evaporated when only TNF-naïverecipients were considered. Factors associated with earliertermination of therapy included female gender and higherdisease activity. The use of methotrexate correlated with asubstantially reduced rate of discontinuation (hazard ratio0.53; 95% confidence interval 0.43–0.65); no subgroupanalysis was presented as to whether this advantage accrueddifferentially across anti-TNF agents. Interestingly, themethotrexate effect was absent in patients with AS.Improvement in health-related quality of life with TNFtherapy was greater in PsA and AS – particularly in the latter,accounting perhaps for the improved duration of therapy inthese patients.

Address for reprints: MS Heiberg, Department of Rheumatology,Diakonhjemmet Hospital, Box 23 Vinderen, 0319 Oslo, Norway. Email: [email protected]

Effect of glucosamine sulfate on hip osteoarthritis:a randomized trialRozendaal RM, Koes BW, van Osch GJ et al.Ann Intern Med 2008;148:268–77.

The utility of oral glucosamine for the management ofosteoarthritis (OA) remains uncertain, given conflictingresults of published clinical trials. Most of these studies havebeen conducted in knee OA. The present authors designed a

clinical trial to assess a therapeutic effect of glucosaminesulfate, 1500 mg daily for 2 years, in patients with hip OArecruited through primary care practices. Patients referredfor screening evaluation at the study center were eligible ifthey met American College of Rheumatology hip OA criteriawith Kellgren/Lawrence scores <4 (the most severe), werenot awaiting hip surgery, and were not receivingglucosamine therapy. Standardized radiographs wereobtained at baseline and 24 months, clinical questionnaireswere completed every 3 months, and compliance wasassessed by self report and pill counts. Patients and assessorswere blinded to the treatment group. Criteria for clinicallyimportant changes in pain and function (Western Ontarioand McMaster Osteoarthritis Index [WOMAC]) scores, andjoint narrowing, were defined in advance.

Out of 417 patients referred from general practitioners,222 patients with a mean age of approximately 63 yearswere enrolled in the study. Over 93% were available forfinal assessment. At 24 months, no significant differenceswere noted between glucosamine and placebo arms in pain,function, joint space narrowing, or pain medication use;95% confidence intervals excluded the pre-determinedclinically important difference values. Restriction of analysisto patients with excellent (>80%) adherence to therapy didnot change these results.

While it remains formally possible that this study misseda clinical subpopulation that responds well to glucosamine –because such patients might already be taking thesupplement and would, therefore, be excluded from the trial– the data are extremely convincing that no substantialbenefit for hip OA can be expected from glucosamine in thedose regimen assessed.

Address for reprints: RM Rozendaal, Department of General Practice,Erasmus Medical Center, Room Fg 351, PO BOX 2040, 3000 CARotterdam, The Netherlands. Email: [email protected]

Immunosuppressive therapy in lupus- and mixedconnective tissue disease-associated pulmonaryarterial hypertension: a retrospective analysis oftwenty-three casesJais X, Launay D, Yaici A et al.Arthritis Rheum 2008;58:521–31.

Connective tissue disease-associated pulmonary arterialhypertension may be successfully managed with atreatment protocol that begins with immunosuppressivetherapy and corticosteroids in less severely affectedpatients, reserving pulmonary vasodilators for those who progress or have severe functional limitations at presentation.

The role of glucosamine in the treatment of osteo-arthritis (OA) remains controversial. Only a few studieshave investigated its efficacy in hip OA. In thismethodologically pristine, 2-year, randomized, double-blind clinical trial free of industry support, patients withhip OA recruited from primary care offices demonstratedno benefit from therapy.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 58

TREATMENT STRATEGIES


Pulmonary arterial hypertension (PAH) can be a life-threatening complication of various collagen vascular diseases,including progressive systemic sclerosis (PSS), systemic lupuserythematosus (SLE), and mixed connective tissue disease(MCTD). Therapy for PAH has focused on the use ofpulmonary vasodilators to counteract vascular hyper-reactivity.However, recent evidence has pointed to the involvement ofan immune-mediated inflammatory process in thedevelopment and progression of PAH in patients with CTD, suggesting that immunosuppressive therapy may have a role in the management of PAH in at least some of these patients.

In a previous publication, the authors of this manuscriptreported on their treatment of a group of patients with PAHrelated to either SLE or MCTD [1]. They found that a subsetof these patients responded to immunosuppressants andglucocorticoids alone, without the addition of pulmonaryvasodilators. They proposed that patients who were NewYork Heart Association functional class I or II, as well asthose who were class II but had at least a near normalcardiac index, should be treated with immunosuppressantsand glucocorticoids initially, adding vasodilators only if theyfailed to respond. They also proposed that those patientswith worse functional class should be treated with acombination of immunosuppressants, glucocorticoids, andpulmonary vasodilators, although they acknowledged thelack of evidence that the first two were beneficial.

In the current manuscript, the authors report on the first 23 consecutive patients treated according to their proposedalgorithm, including 16 who met criteria for treatment withimmunosuppressive therapy alone and seven who weretreated with a combination of immunosuppressives andvasodilators. Eight of the first group met defined criteria forresponse (functional class I or II along with hemodynamicimprovement at the conclusion of therapy), and six of theeight non-responders subsequently improved with theaddition of pulmonary vasodilators. Four of the sevenpatients in the second group responded to the combination therapy.

The authors conclude that a subset of patients withCTD-associated PAH may respond to treatment aimed at theimmune mechanisms presumably responsible for theirdisease. They propose a sequence of therapy that appears tobe successful in their retrospective, uncontrolled series.Randomized, controlled trials in this population willdetermine whether their approach should become standardof care.1. Sanchez O, Sitbon O, Jais X et al. Immunosuppressive therapy in connective tissue

diseases-associated pulmonary arterial hypertension. Chest 2006;130:182–9.

Address for reprints: M Humbert, Service de Pneumologie etRéanimation Respiratoire, Hôpital Antoine-Béclère, 157 rue de la Portede Trivaux, 92140 Clamart, France. Email: [email protected]

Comparison of the clinical efficacy and safety ofsubcutaneous versus oral administration ofmethotrexate in patients with active rheumatoidarthritis: results of a six-month, multicenter,randomized, double-blind, controlled, Phase IV trialBraun J, Kästner P, Flaxenberg P et al.Arthritis Rheum 2008;58:73–81.

Low-dose methotrexate (MTX) therapy for rheumatoidarthritis (RA) may be administered orally or viasubcutaneous injection. While the latter method isassociated with better bioavailability of the medication andhas been used for patients with gastrointestinal toxicity fromoral dosing, there is uncertainty about whethersubcutaneous dosing of MTX generally results in greaterclinical efficacy.

In the current study, the authors identified 375 biologicalagent-naïve patients with active RA randomized to receive15 mg/week of MTX either orally or subcutaneously, in ablinded fashion for 16 weeks. At week 16, patients who hadnot achieved an ACR20 response had their dose escalatedfrom 15 mg orally to 15 mg subcutaneously for thoseoriginally on oral therapy, and from 15 mg subcutaneouslyto 20 mg subcutaneously for those originally onsubcutaneous therapy. The primary outcome of the studywas overall ACR20 response at 24 weeks. Higher levels ofresponse were also assessed.

At week 24, 78% of those initially treated withsubcutaneous MTX had achieved an ACR20 responsecompared with 70% of those initially treated with oral MTX,a difference that was statistically significant. The differencein ACR70 response was also statistically significant at 41%versus 33%, although the difference in ACR50 response wasnot. Fourteen per cent of the patients, approximatelyequally split between the two treatment groups, had anescalation of therapy at week 16 after not meeting anACR20 response. Interestingly, patients with diseaseduration of >1 year had the highest response tosubcutaneous MTX. Overall, adverse events were similarbetween the two groups; those treated with oral MTXreported more diarrhea, while those treated withsubcutaneous MTX reported more loss of appetite.

The authors conclude that subcutaneous MTX is moreeffective than oral MTX for RA, although the latter certainlyappears effective enough to support its initial use if it is moreconvenient for the patient. The results with both routes of

Subcutaneous methotrexate (MTX) is clinically moreeffective than oral MTX in MTX-naïve rheumatoid arthritispatients, although both methods of administration producesubstantial clinical benefits in this population.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 59

administration confirm the effectiveness of MTX as initialmonotherapy for RA.

Address for reprints: J Braun, Rheumazentrum Ruhrgebiet,Landgrafenstrasse 15, 44652 Herne, Germany. Email: [email protected]

Responsiveness of patient reported outcomesincluding fatigue, sleep quality, activity limitation,and quality of life following treatment withabatacept for rheumatoid arthritisWells G, Li T, Maxwell L et al.Ann Rheum Dis 2008;67:260–5.

Patient-reported outcomes (PROs) assess health, wellbeing,and treatment response from the perspective of the patientand may differ from the clinical manifestations. Therefore,the present authors evaluated the responsiveness of PROs,including the Short Form 36-item survey (SF-36), activitylimitation, fatigue, and sleep with respect to clinicalmeasures in rheumatoid arthritis (RA) patients. Data wereobtained from the ATTAIN (Abatacept Trial in Treatment ofAnti-TNF Inadequate Responders) study, which evaluatedthe efficacy and safety of abatacept on a background ofdisease-modifying antirheumatic drug (DMARD) treatmentin patients with active RA who were anti-tumor necrosisfactor (anti-TNF) treatment “failures”. Assessments includedtender joint count (TJC), health assessment questionnaire(HAQ), and clinical parameters. Multiple PROs includinghealth-related quality of life (measured by the SF-36) andsleep quality were assessed. The ability to detect a treatmenteffect was evaluated by the treatment difference (abataceptvs. placebo), percentage improvement relative to baselinescores, standardized response mean (SRM), and the relativeefficiency for assessing an outcome’s ability to detect atreatment effect relative to the TJC.

The PROs and quality of life (SF-36) indicated greatimpairment at baseline. The mean changes from baseline inthe study outcomes were larger for the abatacept group.The largest relative percent improvements were found foracute phase reactants and the PRO activity limitation,whereas the more psychological PROs (e.g. mental healthand mental component score) gave the smallest relativepercent improvement. Similar trends could be observed for

the SRMs, with a large SRM for physician global, HAQ, SF-36 physician component score, SF-36 bodily pain andfatigue. For the more physical parameters (physician global,SF-36 bodily pain, pain intensity, and HAQ), the relativeefficiencies in relation to the TJC for detecting a treatmenteffect were greater, in contrast to the assessment ofpsychological components. In general, most of the PROsshowed good concordance with both American College ofRheumatology criteria and European League AgainstRheumatism response criteria. Significant treatmentdifferences between abatacept and placebo were found forall core set measures and PROs. The differences weresubstantial in most cases, with significant but smallerdifferences noted for PROs associated with the morepsychological attributes.

Address for reprints: G Wells, Department of Epidemiology andCommunity Medicine, University of Ottawa, 451 Smyth Road, Ottawa,ON, KIH 8M5, Canada. Email: [email protected]

GENETICS

The additive effect of individual genes in predictingrisk of knee osteoarthritisValdes AM, Doherty M, Spector TD.Ann Rheum Dis 2008;67:124–7.

Genetic factors are determinants of osteoarthritis (OA), butmost individual genetic associations appear modest. Thegenotypes for 36 single nucleotide polymorphisms (SNPs) in17 candidate genes previously associated with OA wereanalyzed in 298 men and 305 women diagnosed with kneeOA, and in 297 male and 299 female control subjects. Theodds ratio (OR) for individuals in the top quartile of the“genetic risk” variable, compared with those in the bottomquartile was found to be approximately 9 (95% confidenceinterval [CI] 5.20–14.49) for women, and about 5 (95% CI3.10–8.27) for men.

Although these data indicate a relatively large effect, theodds between individuals in the lower versus the upper half of

Clinically relevant risk factors for osteoarthritis (OA) areobesity, with odds ratios (OR) of 3–18, and knee injury,with OR ranging from 5–16. The OR conferred bygenetic risk factors have been found to be small,indicating that genetics is perhaps more relevant in theelucidation of biologically relevant pathways than indisease management. With large-scale genetic testing,this may change. Indeed, the current study indicatessome additive effect of individual genetic risk factors inthe prediction of risk for OA.

On the basis of the ATTAIN (Abatacept Trial in Treatmentof Anti-TNF Inadequate Responders) study, the presentauthors validate that patient-reported outcomes forevaluating treatments for rheumatoid arthritis can beused to detect improvements that are important topatients. Physical assessments are more responsive to aneffective treatment than psychological assessments.

CLINICAL REVIEWS


RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 60

PATHOGENESIS


the distribution according to these known genetic risk factorswas only approximately 3. Nevertheless, the fact thatadditivity of genetic risk factors was observed is encouraging.

Address for reprints: AM Valdes, Twin Research Unit, St Thomas’Hospital Campus, Kings College London, Lambeth Palace Road,London, SE1 7EH, UK.

Association of systemic lupus erythematosus withC8orf13-BLK and ITGAM-ITGAXHom G, Graham RR, Modrek B et al.N Engl J Med 2008;358:900–9.

Systemic lupus erythematosus (SLE) is a chronic autoimmunedisease with very strong genetic and environmentalcomponents. A number of genes: interferon regulatoryfactor 5 (IRF5) and signal transducer and activator oftranscription 4 (STAT4), and some human leukocyteantigen-DR (HLA-DR) alleles have already been associatedwith the disease. In this report, the results of a genome-widesingle nucleotide polymorphism (SNP) scan in SLE patientsand control subjects are described.

A total of 1861 samples from control subjects weregenotyped on the Illumina Human Hap550 GenotypingBead-Chip (Illumina Inc., San Diego, CA, USA). A total of1465 samples (464 from cases and 1001 from controlsubjects) were genotyped on the HumanHap550v1 chip,and 1875 samples (1015 from cases and 860 from controls)were genotyped on the HumanHap550v3 chip. Anadditional, independent set of 1722 samples that weregenotyped on the HumanHap550 BeadChip was obtainedfrom public data repositories. Analysis of the genotypedsamples was performed in serial phases. After applying thedata-quality filters, series one consisted of 411 case subjectsand 1047 control subjects, series two consisted of 595 casesand 1516 controls, and series three comprised 305 casesubjects and 777 controls. A total of 502 033 SNPs wereadvanced into analyses. In a replication study consisting of793 case subjects and 857 control subjects from Sweden,two specific SNPs, namely rs11574637 and rs13277113,were genotyped by the use of fluorescent single-baseextension assays. To examine the functional consequencesof the finding, the authors analyzed gene expression of Blymphoid tyrosine kinase (BLK) and C8orf13 (chromosome8p23.1) in B cell lines transformed by the Epstein–Barr virus.

A minor allele of the SNP rs13277113 in the regionupstream from the transcription initiation site of the geneencoding BLK and C8orf13 was associated with a risk fordeveloping the disease both in the US and Swedishcase–control series (odds ratio [OR] 1.39; p=1x10–10), andalso with altered levels of messenger RNA in B cell lines. Inaddition, variants on chromosome 16p11.22, near the genesencoding integrin alpha M (ITGAM, or CD11b) and integrinalpha X (ITGAX), were associated with SLE in the combinedsample (rs11574637; OR 1.33; p=3x10–11). Both associationswere confirmed in the replication study.

Thus, two new genetic loci for SLE susceptibility wereidentified and associated with reduced expression of BLKand increased expression of C8orf13.

Address for reprints: TW Behrens, Genentech, 1 DNA Way, 45-3B, South San Francisco, CA 94080, USA. Email: [email protected]

PATHOGENESIS

Inhibition of monocyte chemoattractant protein-1ameliorates rat adjuvant-induced arthritisShahrara S, Proudfoot AE, Park CC et al.J Immunol 2008;180:3447–56.

The chemokines RANTES/CCL5 and monocytechemoattractant protein-1 (MCP-1)/CCL2 regulate thechemotaxis of monocytes and T lymphocytes. They are highlyexpressed in rheumatoid arthritis (RA) synovial tissue and arethought to be involved in the pathogenesis of the diseasethrough the promotion of leukocyte migration into thesynovial tissue. In this study, the effects of inhibition ofRANTES and MCP-1 were examined in rat adjuvant-inducedarthritis (AIA) – an animal model of RA. After the onset ofAIA, the modified chemokines P8A-MCP-1 and 44AANA47-RANTES, two inhibitors of endogenous MCP-1 and RANTES,respectively, were administered intraperitoneally, separately, orin combination, for 7 or 14 days. Measurement of anklecircumferences and articular index (AI) scores revealedimproved clinical signs of AIA after treatment with P8A-MCP-1. Furthermore, P8A-MCP-1 decreased jointinflammation, synovial lining, and bone destruction as

The therapeutic inhibition of monocyte chemoattractantprotein-1 (MCP-1)/CCL2 and RANTES/CCL5 wasexamined in rat adjuvant-induced arthritis. Treatment ofrats with an inhibitor of MCP-1 (P8A-MCP-1) resulted inamelioration of the disease, whereas inhibition ofRANTES or its corresponding receptors (using 44AANA47-RANTES and Met-RANTES) had no effect.

Identification of genetic factors involved in thepathogenesis of autoimmune diseases such as systemiclupus erythematosus (SLE) is of obvious importance. Inthe present study, the authors performed a genome-widescan of samples from patients with SLE and controlsubjects in order to identify disease-related genes.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 61

determined by histology, X-rays, and real-time reversetranscriptase-polymerase chain reaction (RT-PCR) for themarkers associated with bone destruction (receptor activatorof nuclear factor-κ ligand and matrix metalloproteinase-9).Downregulation of proinflammatory mediators and areduction in the number of macrophages in the synovial tissuecould be observed by enzyme-liked immunosorbent assay(ELISA) and immunostaining. As RANTES and MCP-1 inducechemotaxis through activation of p38 mitogen-activatedprotein kinase, the amount of phosphorylated p38 wasinvestigated by immunostaining and Western blot analysis.Reduced activation of p38 was detected in rats treated withP8A-MCP-1. In contrast, treatment with 44AANA47-RANTEShad no effect on the evaluated features. Furthermore, thecombination of P8A-MCP-1 and 44AANA47-RANTES was notmore effective than P8A-MCP-1 alone. Administration ofMet-RANTES, an antagonist for RANTES receptors, did notameliorate clinical signs either.

These results indicate that post-disease onset treatmentwith P8A-MCP-1 ameliorates AIA. The authors suggest thatMCP-1 is critical for the progression of AIA, whereas RANTESmay be of limited importance in the pathogenesis of RA.

Address for reprints: S Shahrara, Northwestern University FeinbergSchool of Medicine, Department of Medicine, Section of Rheumatology,McGaw Pavillion, 240 East Huron, Suite 2220, Chicago, IL 60611, USA.Email: [email protected]

Cutting edge: TNFR-shedding by CD4+ CD25+

regulatory T cells inhibits the induction ofinflammatory mediatorsVan Mierlo GJ, Scherer HU, Hameetman M et al.J Immunol 2008;180:2747–51.

Regulatory T cells (Treg cells) represent an important andheterogeneous population of lymphocytes that are capable ofblocking the action of other lymphocytes through directcontact as well as production of soluble mediators, such asinterleukin-10 (IL-10) and transforming growth factor-β(TGF-β). In general, these cells have been regarded as a“brake” on adaptive immunity, while leaving innate immunecells such as neutrophils, macrophages, and mast cells largely

alone. However, this group reports a novel mechanism bywhich Tregs could impact upon the effector arm of theimmune system as well. In this article, they show that Tregsexpress high surface levels of the p75 tumor necrosis factor(TNF) receptor TNFRII, the same molecule that forms the TNF-binding site of etanercept. Upon activation, Treg cells begin toshed this receptor via cleavage of the extracellular portion ofthe molecule, which then binds soluble TNF to prevent it fromreaching its targets on the surface of other cells. In a functionalassay, the authors show that Treg cells derived from micedeficient in TNFRII lack the ability to antagonize TNF. Further,pharmacological blockade of the enzyme responsible forcleaving TNFRII from the surface results in excess accumulationof this receptor on the surface of Treg cells.

No in vivo data are shown to demonstrate the importanceof TNFRII release by Treg cells in inflammatory diseases.Nevertheless, the authors introduce a compelling potentialmechanism by which Treg cells could limit autoimmunity at thelevel of tissue inflammation and injury, in addition to theircritical role in the maintenance of self-tolerance.

Address for reprints: R Toes, Department of Rheumatology, Postal ZoneC1-R, Leiden University Medical Center, PO BOX 9600, 2300 RC,Leiden, The Netherlands. Email: [email protected]

A followup study of antiphospholipid antibodiesand associated neuropsychiatric manifestations in137 children with systemic lupus erythematosusAvcin T, Benseler SM, Tyrrell PN et al.Arthritis Rheum 2008;59:206–13.

The pathogenesis of neuropsychiatric manifestations ofsystemic lupus erythematosus (SLE) remains uncertain. Sinceantiphospholipid (aPL) antibodies may have neurologicaleffects above and beyond the promotion of cerebrovascularthrombosis, this group examined a cohort of children withSLE tested regularly for these antibodies for a potentialassociation between aPL and neuropsychiatric lupus. Of 175patients seen at the Toronto Pediatric Lupus Clinic (TheHospital for Sick Children, Toronto, ON, Canada), 137 metthe entry criteria of aPL testing within the first 3 monthsafter diagnosis and longitudinal follow-up within the clinic.

Antiphospholipid (aPL) antibodies can induceneurological disease via cerebrovascular thrombosis andpotentially other mechanisms. The present authorsinvestigated the association of aPL antibodies withneuropsychiatric systemic lupus erythematosus (SLE) in137 children. Anti-β2-glycoprotein I antibodies weresomewhat more common among patients withneuropsychiatric SLE, and lupus anticoagulant was foundat an elevated frequency in patients with stroke.

Regulatory T cells (Treg cells) help to preventautoimmunity by limiting the activity of otherlymphocytes. Recently, experimental evidence in micehas suggested that Treg cells can also limit antigen-independent inflammation. In this article the authorsreport that activated Treg cells can shed substantialquantities of the tumor necrosis factor (TNF) receptorTNFRII, thereby dampening inflammation driven byinnate immune lineages such as macrophages.

CLINICAL REVIEWS


RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 62

INFECTIOUS AND MALIGNANT COMPLICATIONS


Overall, 66% of children had aPL antibodies at presentation,including anticardiolipin (aCL) in 65%, anti-β2-glycoprotein I(anti-β2GPI) in 41%, and lupus anticoagulant (LAC) in 26%.Approximately 70% of these patients remained persistentlypositive, while new positives were detected over the courseof observation in 20% of the remaining patients (meanfollow-up 31 months). Neuropsychiatric SLE wasdocumented in 26% of patients, ranging from headache(the most common) to transverse myelitis. However,evidence for an association between aPL status andneuropsychiatric disease was scarce. Evaluation of potentialassociations between aPL antibodies (considered individuallyor collectively, at presentation or over the course of disease)and different neuropsychiatric manifestations identified aplausible association between LAC positivity and stroke (fourof five stroke patients compared with 18 of 79 patientswithout stroke tested for LAC; p=0.015). Anti-β2GPIantibodies were positive at some point in 48% of patientswith neuropsychiatric manifestations compared with 25% ofpatients without (p=0.02). Chorea was noted in twopatients, both of whom were LAC-positive. General diseaseactivity, as assessed by SLE disease activity index, tended totrack with aCL titer but not with anti-β2GPI titer.

These findings are of potential interest but, in virtue ofthe very large number of hypotheses tested in thisexploratory study, will require validation in other cohorts.

Address for reprints: ED Silverman, Division of Rheumatology, The Hospitalfor Sick Children, University of Toronto, 555 University Avenue, Toronto,ON, MSG 1XA, Canada. Email: [email protected]

INFECTIOUS AND MALIGNANTCOMPLICATIONS

Prevalence of malignancy in psoriatic arthritisRohekar S, Tom BD, Hassa A et al.Arthritis Rheum 2008;58:82–7.

Rheumatoid arthritis (RA) and, potentially, antirheumatictherapies have been associated with a higher risk oflymphoma and potentially certain other malignancies. Inaddition, psoriasis has been associated with an increased riskfor skin malignancy in some studies, although the arearemains controversial. To examine whether patients with

adult psoriatic arthritis (PsA) develop malignancy at a rategreater than expected for the general population, theauthors examined a prospective, well-documented registryof >680 PsA patients for the incidence and nature of firstmalignancies occurring after entry into the cohort. Tocapture malignancies not recorded in the registry or afterloss to follow-up, the authors queried an Ontario (Canada)cancer registry for data on the same patients. Of 665patients included in the analysis, 68 (10.2%) developed amalignancy meeting study criteria. Comparison of the risk ofmalignancy overall, and of common types of malignancyconsidered separately, showed no change from malignancyincidence in the general Ontario population, as determinedthrough a provincial cancer registry (standard incidence ratio0.98, 95% confidence interval 0.77–1.24). Patients whodeveloped malignancy did not differ statistically from thosewithout across a wide spectrum of parameters, though useof biologics in the population was limited.

Together the data indicate that PsA, unlike RA, exhibitsno clear association with an overall elevated risk ofmalignancy, but the number of cases was too small toevaluate elevated risk in specific cancer subtypes.

Address for reprints: D Gladman, Centre for Prognosis Studies in the Rheumatic Diseases, Toronto Western Hospital, 1E-410B, 399 Bathurst Street, Toronto, ON, M5T 2S8, Canada. Email: [email protected]

Morbidity and mortality in rheumatoid arthritispatients with prolonged therapy-inducedlymphopenia: twelve-year outcomesLorenzi AR, Clarke AM, Wooldridge T et al.Arthritis Rheum 2008;58:370–5.

In an analysis of 53 rheumatoid arthritis (RA) patientstreated with the lymphocytotoxic monoclonal antibody

The current armamentarium of drugs against rheumatoidarthritis (RA) consists of agents that modulate the immunesystem at increasingly sophisticated levels, such as tumornecrosis factor blockade, costimulation blockade, and B celldepletion. Whether such interventions are safe can only be learned from long-term observations. In the early 1990s, patients were treated with the lymphocytotoxicmonoclonal antibody alemtuzumab (anti-CD52) with thehope that lymphocyte depletion and reconstitution wouldresult in autoreactive lymphocytes being replaced by atolerant immune system. In these patients, it was observedthat despite continued lymphopenia >11 years aftertherapy, no excess mortality or unusual infection-relatedmorbidity occurred. These data are reassuring for the long-term outcomes of current immunomodulatory therapies for RA.

Rheumatoid arthritis, particularly when highlyinflammatory, has been associated with an increased riskfor certain malignancies. These authors examined aprospective registry of >680 patients with adult psoriaticarthritis to determine whether a similar association exists.They found none.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 63

alemtuzumab (anti-CD52) and 102 control subjects selectedfrom the European League Against Rheumatism database ofpatients with rheumatic disorders (matched for diseaseseverity), lymphopenia was found to persist in the patientstreated with anti-CD52 after >11 years of follow-up.However, no adverse clinical consequences were observed,with a mortality rate ratio of 1.2 (95% confidence interval0.72–1.98). The causes of death were similar to those thatwould be expected in a hospital-based RA cohort. Noopportunistic infections were noted, and only threeinfections were documented following 36 electiveorthopedic procedures.

Address for reprints: JD Isaacs, Musculoskeletal Research Group,Institute of Cellular Medicine, Newcastle University, Newcastle-upon-Tyne, UK. Email: [email protected]

Hematologic malignant neoplasms after drugexposure in rheumatoid arthritisBernatsky S, Clarke AE, Suissa S. Arch Intern Med 2008;168:378–81.

This case–control study involved subjects from a cohort of 23 810 patients with rheumatoid arthritis assembled fromadministrative databases covering the population of the stateof Quebec, Canada, in the period 1980–2003. Cases that hadhematological malignant neoplasms were ascertained fromphysician billing and hospitalization records. Patients werematched for age and sex with 10 control subjects and clinicalvariables and concomitant medications were taken intoaccount, in order to diminish the possibility that theunderlying disease led to hematological malignancies. Duringthe study, hematological malignant neoplasms developed in619 patients, including lymphomas in 346 patients, leukemiain 178 patients, and multiple myelomas in 95 patients. Owingto the fact that biological agents first appeared in Quebec in

2002, there were too few exposures to these drugs to obtainmeaningful data.

Address for reprints: S Bernatsky, Division of Clinical Epidemiology,McGill University Health Centre, 687 Pine Avenue West, V Building,Montreal, QC, H3A 1A, Canada.

The risk of herpes zoster in patients withrheumatoid arthritis in the United States and theUnited KingdomSmitten AL, Choi HK, Hochberg MC et al.Arthritis Rheum 2007;57:1431–8.

Assessing and quantifying the underlying risk of specificinfections in rheumatoid arthritis (RA) is essential tounderstanding the added risk that may be attributable todisease-modifying antirheumatic drug (DMARD) therapy inthese patients. While herpes zoster has been suggested to bemore common in patients with RA, the evidence for this islimited, as is any firm evidence linking specific RA therapies toan increased risk of herpes zoster. In the current cohort study,the authors use data from a US managed-care database(PharMetrics claims database) and the UK General PracticeResearch Database (GPRD) to examine the rate of herpeszoster. The PharMetrics database included 122 272 patientswith RA between 1998 and 2002, and the GPRD included38 621 RA patients from 1990–2001. Compared with non-RApatients, the adjusted hazard ratio for herpes zoster was 1.91in the PharMetrics database and 1.65 in the GPRD, both ofwhich had 95% confidence intervals that did not cross 1.

Using nested, case–control analysis the authors were ableto show that treatment was associated with herpes zoster inboth databases. In the PharMetrics database, current use ofboth biological and non-biological DMARDs alone wasassociated with a statistically increased risk of herpes zoster,with odds ratios (ORs) of 1.52 and 1.34, respectively. Thecombination of biological and traditional DMARDs was alsoassociated with increased risk, although this was notstatistically significant, possibly because the number ofsubjects on both therapies was too low. In the GPRD, whichdid not include any patients taking biological DMARDs, theuse of non-biological DMARDs was associated with astatistically increased risk of herpes zoster (OR 1.27). Oralcorticosteroid use increased the risk of herpes zoster in bothdatabases, regardless of DMARD use.

These data from two large databases suggest that RA isindependently associated with an increased risk for the

Analysis of US and UK databases shows an increased riskof herpes zoster in rheumatoid arthritis patients, which isfurther increased with the use of disease-modifyingantirheumatic drugs or corticosteroids.

Hematological malignancies seem to occur morefrequently in patients with rheumatoid arthritis (RA).However, it is unclear whether there is a link withdisease-modifying antirheumatic drug (DMARD)exposure. In a large, case–control study of 619 cases (RAplus hematological neoplasms) and 6190 controls (RAalone, approximately matched in terms of diseaseseverity but without hematological neoplasms) thegreatest relative risk (about 2.2) for hematologicalmalignant neoplasms was noted after use ofcyclophosphamide. The only other significant risk ratios(RR) were for azathioprine use although this wasmarginal (the 95% confidence interval was 1.01–2.03).No increased RR were found for the use of methotrexateor antimalarial agents.

CLINICAL REVIEWS


RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 64

CARDIOVASCULAR RISK


development of herpes zoster, and that this risk may bemagnified by the use of corticosteroids, as well as bothbiological and non-biological DMARDs. This suggests thatRA patients may be candidates for the zoster vaccine, whichmay be particularly appropriate early in the disease coursebefore initiation of therapies that may preclude the use oflive vaccines.

Address for reprints: KA Chan, Department of Epidemiology, HarvardSchool of Public Health, 677 Huntington Avenue, Boston, MA 02115,USA. Email: [email protected]

CARDIOVASCULAR RISK

Accumulation of coronary artery disease riskfactors over three years: data from an internationalinception cohortUrowitz MB, Gladman D, Ibanez D et al.Arthritis Rheum 2008;59:176–80.

Coronary artery disease (CAD) affects up to 10% of patientswith systemic lupus erythematosus (SLE) and CAD risk factorsare predictive of myocardial infarction, adverse renaloutcomes, and mortality in SLE patients. Therefore, thepresent investigators monitored CAD risk factors in SLEpatients during the early phase of the disease, i.e. during thefirst years after the diagnosis. A total of 278 patients from theSystemic Lupus International Collaborating Clinics (SLICC)registry (a multicenter, international, inception cohort of newlydiagnosed SLE patients) were followed-up over a 3-yearperiod for their clinical, laboratory, and CAD risk factors,which were documented at enrolment and annually.

The SLE patients included in the study wererepresentative of the total SLICC cohort of 935 patients interms of their demographic features, CAD risk factordifferences, and medications. The remaining patients fromthe cohort had not yet completed 3 years of follow-up. Atenrolment, almost one-third of the patients had someclassical CAD risk factors such as hypertension (39.2%),hypercholesterolemia (36.3%), or ever smoking (37.4). Overthe 3 years of follow-up, all classical CAD risk factorsincreased by up to 60% of levels at enrollment: 48.6% forhypertension, 65.3% for hypercholesterolemia, 36.7% forsmoking, 55.6% for diabetes, and 37.2% for post-

menopausal state. Similarly, non-traditional CAD risk factors,present in up to 30% of patients at enrolment, accumulatedover the 3 years of follow-up: body mass index increased by54.3%, waist:hip ratio >0.8 by 79.5%, low physical activityby 71.6%, family history of CAD by 40.8%, and nephroticsyndrome by 82.8%. The Framingham 10-year risk profilewas higher in men than in women with SLE both at entryand at 3 years. It decreased in men over the 3-yearobservation period and remained constant in women. Withregard to the use of medications, the treatment ofhypertension and hypercholesterolemia, and the use ofantimalarials and immunosuppressives increased over 3years, whereas corticosteroid use increased only slightly.

The authors conclude that CAD risk factors accumulatein SLE patients from the early stages of the disease. Thus,they require careful monitoring, starting at diagnosis, inorder to be properly managed.

Address for reprints: MB Urowitz, Centre for Prognosis Studies, TorontoWestern Hospital, IE-409, 399 Bathurst Street, Toronto, ON, M5T 2S8,Canada. Email: [email protected]

Rheumatoid arthritis is associated with a highprevalence of hypothyroidism that amplifies itscardiovascular riskRaterman HG, van Halm VP, Voskuyl AE et al.Ann Rheum Dis 2008;67:229–32.

Hypothyroidism has been identified in some studies as acardiovascular risk factor, operating via factors such lipidprofile, blood pressure elevation, and effects on theendothelium. These authors examined a prospective cohortof 358 rheumatoid arthritis (RA) patients for the prevalenceof hypothyroidism and its potential effect on cardiovasculardisease (CVD) in a population already at an elevated riskbecause of RA. They found that 6.8% of patients had“clinical hypothyroidism” (defined as an establishedphysician diagnosis of hypothyroidism or the satisfaction ofa published criteria set), in excess of the 2.7% expected forthe Dutch population in general (p<0.001). Of these 16women, six (37.5%) had CVD, defined to include eithercoronary, cerebrovascular, or peripheral vascular disease,

Hypothyroidism has been linked to an increased risk ofcardiovascular disease (CVD), as has rheumatoid arthritis(RA). In the cohort of RA patients investigated by theseauthors, hypothyroidism was found to be more prevalentthan expected, and patients with both RA andhypothyroidism appeared to be at an elevated risk forCVD. The results are thought-provoking but arise from asmall sample and will require replication before definitiveconclusions can be drawn.

Risk factors for coronary artery disease (CAD) are present inone-third of patients with systemic lupus erythematosus(SLE) at the initial presentation of the disease, and increasegreatly during its early phase. Therefore, patients need tobe carefully monitored for proper management of CAD riskfactors, from the diagnosis of SLE.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 65

compared with 13% among the euthyroid members of theRA cohort (p=0.05). After adjusting for potentialconfounders – the hypothyroid patients were, on average, 6years older and were all female – the odds ratio for CVDwas around 3–4.

The authors suggest these data “demonstrate” a higherrisk for CVD in patients with co-incident RA andhypothyroidism, and recommend consideration of screeningfor hypothyroidism in patients with RA. However, theconfidence with which this conclusion is reached is quite outof proportion to the evidence presented. The authors’conclusions derive from four excess CVD diagnoses amongthese 16 older women – had there been only three, theeffect size would have been non-significant (p=0.047 →p=0.13; calculated by this Editor). The nature of the CVDevents is not described, nor is their temporal onset vis a viseither RA or hypothyroidism. Further, the definition of“clinical hypothyroidism” used in this series raises genuineconcerns for confounding, since patients might be morelikely to receive a formal diagnosis if under regular medicalcare for another problem, for example CVD. Accordingly,while this article remains provocative and should motivatefurther research, the conclusions are in no way sufficientlyrobust to justify present action by clinicians.

Address for reprints: MT Nurmohamed, VU University Medical Centre,departments of International Medicine and Rheumatology, PO BOX 7057,1007 MB Amsterdam, The Netherlands. Email: [email protected]

Recent corticosteroid use and recent diseaseactivity: independent determinants of coronaryheart disease risk factors in systemic lupuserythematosus?Karp I, Abrahamowicz M, Fortin PR et al. Arthritis Rheum 2008;59:169–75.

The pathogenesis of the markedly elevated risk of coronaryheart disease (CHD) in systemic lupus erythematosus (SLE)patients is unknown. In particular, the causal roles ofcorticosteroid therapy and SLE disease activity, and whethertheir putative effects are mediated through conventional riskfactors, are unclear. To this end, data were obtained from

the charts of >11 000 visits made by 310 patients with SLEto the Montreal General Hospital, Montreal, QC, Canada. Ahigher past-year corticosteroid dose was independentlyassociated with a significantly higher overall 2-year CHDrisk. Moreover, corticosteroid use was associated with higherlevels of conventional risk factors such as total serumcholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein cholesterol, apolipoprotein B (ApoB),triglycerides, systolic blood pressure (BP), body mass index,and blood glucose. Higher past-year lupus disease activitywas also independently associated with higher overall 2-yearCHD risk. Furthermore, it was associated with lower HDLcholesterol levels, and higher values for systolic BP, ApoB,triglycerides, and blood glucose.

Address for reprints: M Abrahamowicz, Division of ClinicalEpidemiology, Royal Victoria Hospital, 687 Pine Avenue West, V Building, Montreal, QC, H3A 1A1, Canada. Email: [email protected]

PROGNOSIS AND ASSESSMENT

Immunologic markers as potential predictors ofsystemic autoimmune disease in patients withidiopathic scleritisLin P, Bhullar SS, Tessler HH et al.Am J Ophthalmol 2008;145:463–71.

Scleritis manifests as a painful red or purple discoloration ofthe sclera, and can progress to sight-threatening scleralnecrosis. Causes include infections such as herpes andsyphilis as well as autoimmune diseases such as rheumatoidarthritis (RA), Wegener’s granulomatosis (WG), and lupus.The present authors conducted a retrospective review of 119scleritis patients seen at a university clinic, of whom 91 hadno evident cause at initial evaluation. Over subsequentfollow-up, 11 of these 91 patients were diagnosed with RAand five with WG. Since the general procedure in the clinicwas to assess both rheumatoid factor (RF) andantineutrophil cytoplasmic antibody (ANCA), the authorswere able to review the predictive value of these tests forthe subsequent development of disease. Of the 70 patientstested for RF, 19 (27%) were positive, of whom 10 went on

Scleritis is a painful inflammatory condition of the eye thatcan be infectious, idiopathic, or secondary to autoimmuneconditions such as rheumatoid arthritis and Wegener’sgranulomatosis. The authors investigated the utility oftesting for rheumatoid factor and antineutrophil cytoplasmicantibody in patients referred to a university-based referralclinic, and found that both tests performed well.

During the last decade, a dramatic increase in theincidence of coronary heart disease (CHD) has beenobserved in patients with systemic lupus erythematosus(SLE). In this retrospective study, it was observed thatrecent use of corticosteroids and recent lupus activity areindependently associated with higher values for severalwell-recognized CHD risk factors and with overall 2-yearCHD risk. These data indicate that novel steroid-sparingdrugs should be indicated in the care of SLE patients.

CLINICAL REVIEWS


RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 66

PROGNOSIS AND ASSESSMENT


to receive a diagnosis of RA, whereas only one of 51 RF-negative patients did so. Thus, the positive predictive valueof RF in this setting was 52.6%, with a negative predictivevalue of 98%. RF was predictive even among patients whoreported no joint symptoms. ANCA testing was performedin 70 patients and was positive in seven, of whom threedeveloped WG. Two ANCA-negative patients alsodeveloped WG, giving positive and negative predictivevalues of 42.9% and 96.8%, respectively.

The authors conclude that both tests are useful in thisclinical setting, but caution that the prevalence of RA andWG may be lower in non-tertiary-care clinics, changing thetest characteristics. Whether these tests would providesimilar benefits for patients seen in a rheumatology office,where clinical skills in the global assessment for RA and WGare presumably more advanced than in an ophthalmologyoffice, remains an open question.

Address for reprints: DA Goldstein, University of Illinois at Chicago,Department of Ophthalmology, 1905 West Taylor Street, L131 L.I.E.R.I.,M/C 648, Chicago, IL 60612, USA. Email: [email protected]

High anti-cyclic citrullinated peptide levels and analgorithm of four variables predict radiographicprogression in patients with rheumatoid arthritis:results from a 10-year longitudinal studySyversen SW, Gaarder PI, Goll GL et al.Ann Rheum Dis 2008;67:212–7.

A cohort of 238 patients with rheumatoid arthritis (RA) wasfollowed longitudinally for 10 years. For 125 of the patients,radiographs of the hands were available both at baselineand after 10 years. Baseline sera were analyzed for C-reactive protein, erythrocyte sedimentation rate (ESR), anti-cyclic citrullinated peptide (anti-CCP), immunoglobulin Arheumatoid factor (IgA RF), and IgM RF. Anti-CCP (OR 4.0)was the strongest independent predictor of radiographicprogression. Female gender (OR 3.3), a high ESR (OR 3.2),and a positive IgM RF (OR 3.1) were also independentpredictors. Patients with low levels of anti-CCP (OR 2.6) and

patients with high levels of anti-CCP (OR 9.9) were morelikely to develop radiographic progression than patients withno anti-CCP antibodies. No independent effects wereobserved for IgA RF, whereas after stratification for ESR itwas suggested that an additional effect was observed for RFand anti-CCP antibodies.

Address for reprints: SW Syversen, Department of Rheumatology,Diakonhjemmet Hospital, PB 23 Vindem, N-0319 Oslo, Norway. Email: [email protected]

Development and validation of a patient-baseddisease activity score in rheumatoid arthritis thatcan be used in clinical trials and routine practiceChoy EH, Khoshaba B, Cooper D et al.Arthritis Rheum 2008;59:192–9.

As the evidence grows to support the value of objectivedisease activity measures in rheumatoid arthritis (RA),patient-based disease activity scoring systems, once theprovince of clinical trials, are increasingly being used inclinical practice. In practice, the acceptability of thesemeasurements may be limited by the perception that theircomplexity limits the ease with which they can beincorporated into routine patient management. However, inclinical trials the key limitation of these measurements is thehigh degree of interobserver variability. While this can oftenbe addressed using a single clinician to score disease activity,this solution may be impractical in many situations.

In this paper the authors describe the development andvalidation data for an RA disease activity score that is driven bythe patients’ own assessments, a method that would addressboth the complexity and interobserver variability concerns.They developed two patient-based disease activity scores, onewith (PDAS1) and one without (PDAS2) a sedimentation rate.Both include the patients’ own assessments of swollen andtender joints, as well as visual analogue scales for pain andgeneral health and a health assessment questionnaire (HAQ).

The PDAS1 and the PDAS2 correlated well with theDAS28 in both the development cohort of 204 patients andthe validation cohort of 322 patients. The PDAS1, PDAS2,and the DAS28 all showed similar sensitivity to change,although there was a floor effect with the two PDAS scores,suggesting a decreased sensitivity to change at lower levelsof disease activity.

The authors conclude by suggesting that thesemeasurements may have utility in clinical trials,

Patient-based disease activity scores, either with orwithout the inclusion of a sedimentation rate, correlatewith the 28-joint Disease Activity Score (DAS28) andmay be useful both in clinical trials and clinical practice.

Prediction of joint damage is important for helping theclinician in terms of guiding treatment decisions. In a cohortstudy from Norway in which follow-up of 125 patients wasavailable for 10 years, the presence of autoantibodies (anti-cyclic citrullinated peptide [anti-CCP] and immunoglobulinM rheumatoid factor), erythrocyte sedimentation rate, andfemale gender were independent predictors of radiographicprogression. Interestingly, the patients with high levels ofanti-CCP were especially prone to radiographic progression,indicating that the anti-CCP levels may add moreinformation than the mere presence of anti-CCP.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 67

epidemiological studies, and clinical practice. They suggestthat these scores may be particularly valuable in the last caseby involving patients in the assessment of their own disease,an important factor in optimizing care. The one caveat theauthors note in the use of these scores is that they may notbe accurate in detecting remission or near-remission.

Address for reprints: EH Choy, Sir Alfred Baring Garrod Clinical TrialsUnit, Academic Department of Rheumatology, King’s College London,Weston Education Centre, Cutcombe Road, London SE5 9RJ, UK. Email: [email protected]

Differences in synovial tissue infiltrates betweenanti-cyclic citrullinated peptide-positiverheumatoid arthritis and anti-cyclic citrullinatedpeptide-negative rheumatoid arthritisvan Oosterhout M, Bajema I, Levarht EW et al.Arthritis Rheum 2008;58:53–60.

Anti-cyclic citrullinated peptide (anti-CCP) autoantibodies –highly specific serological markers for rheumatoid arthritis(RA) – often precede the onset of disease symptoms and arethought to be involved in disease pathogenesis. While theclinical presentation of RA patients with or without anti-CCPantibodies is similar at baseline, the progression to erosivedisease occurs predominantly in anti-CCP-positive patients.Since only sparse information is available on the pathologicalfeatures of synovitis in anti-CCP-positive compared withanti-CCP-negative patients, evaluation of synovial tissueinfiltrates might help to clarify the role of anti-CCPantibodies in synovial inflammation.

In this basic science study, synovial tissue specimensobtained from inflamed knee joints of 57 RA patients (34anti-CCP positive, 23 anti-CCP negative) who underwentarthroscopy were analyzed for several histological featuresalong with immunohistochemistry by two independentobservers. In addition, standard anteroposterior (AP) kneeradiographs obtained within 3 months before or after thesynovial tissue sampling were scored for the severity of jointdamage using the Kellgren/Lawrence (K/L) scale(range 0–4).

Synovial tissue samples from anti-CCP-positive patientsrevealed a higher number of infiltrating lymphocytes (61.6vs. 31.4 cells/high-power field [hpf]; p=0.01) expressingmore CD3, CD8, CXCL12, and CD45RO. The difference inthe mean number of infiltrating lymphocytes remaineddemonstrable after correction for disease activity using the28-jount count disease activity score (DAS28). While therewas no difference in vascularity, anti-CCP-positive patientsshowed a thicker synovial lining layer (mean score 2.1 vs.3.3; p=0.002) and a decreased extent of fibrosis (mean score1.2 vs. 2.0; p=0.04). No difference in absolute K/L scoresbetween the groups was observed, but more anti-CCP-positive patients had a K/L score of >1. The evaluation ofsynovial tissue of 31 patients (18 anti-CCP positive)obtained at earlier time points (mean of 3.8 years before theindex biopsy) revealed that the difference in the meanlymphocyte counts was also present at the time of theearlier arthroscopy, with 76.7 cells/hpf in anti-CCP-positivepatients compared with 26.7 cells/hpf in anti-CCP-negativepatients (p=0.008).

The authors concluded that inflamed synovial tissue inanti-CCP-positive RA patients is different from that in anti-CCP-negative patients, particularly with respect toinfiltrating lymphocytes, and shows a higher rate of localjoint destruction, which together might reflect differences inunderlying pathophysiological mechanisms.

Address for reprints: M van Oosterhout, Leiden University MedicalCenter, Department of Rheumatology, C4R, PO Box 9600, 2300Leiden, The Netherlands. Email: [email protected]

The authors of this study analyzed and comparedsynovial tissue infiltrates from patients with rheumatoidarthritis (RA) who were anti-cyclic citrullinated peptide(anti-CCP)-positive with those from anti-CCP-negativeRA subjects. The results demonstrate differences insynovial tissue infiltrates from patients with anti-CCP-positive versus anti-CCP-negative disease, particularlywith regard to lymphocyte infiltration.

CLINICAL REVIEWS


RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 68

MEETIN

G R

EPORT


The Annual European League Against Rheumatism (EULAR)Congress attracts rheumatologists from all over the worldand this year >13 000 delegates attended the meeting inParis, France. A total of 3435 abstracts were submitted forreview for presentation at the congress, representing a 9%increase from 2007, and reflecting a continued interest inclinical and scientific research in the field of rheumatology. Awide range of topics was covered at the Congress; thisreport discusses a few of the key presentations.

Remission in rheumatoid arthritisA number of presentations demonstrated that remission isan achievable goal in a significant proportion of rheumatoidarthritis (RA) patients. Diane van der Woude (LeidenUniversity Medical Center, Leiden, The Netherlands)presented data on disease-modifying antirheumatic drug(DMARD)-free disease remission, and independentpredictors for achieving such remission, in patients at theLeiden Early Arthritis Clinic [1]. During an average follow-upof 8.2 years and a total of 3817 patient-years, 69 of 454patients (15.2%) with recent-onset RA achieved DMARD-free remission, defined as persistent (>1 year) absence ofsynovitis without concurrent use of DMARDs. Independentpredictors for the achievement of DMARD-free remissionincluded old age, low body mass index, low erythrocytesedimentation rate, non-smoking, and the absence of anti-cyclic citrullinated peptide antibodies.

Another group of investigators from The Netherlandsperformed a prospective study involving a cohort of DMARD-naïve, recent-onset RA patients in daily clinical practice [2].Patients were treated using a step-up, tight-control approachwith methotrexate, followed by sulfasalazine, withadalimumab added to methotrexate at week 24 if remission(28-joint Disease Activity Score [DAS28] <2.6) had not beenachieved. The estimated median time to first remission in the169 patients available for the analysis was 25 weeks, with thefollowing percentages of patients in remission: 15.5% at week8, 22.2% at week 12, 30.7% at week 20, 38.8% at week 24,52.1% at week 36, and 51.0% at weeks 48–52.

One-year results from the COMET (Combination ofMethotrexate and Etanercept in Active Early RheumatoidArthritis) trial were presented by Professor Paul Emery(University of Leeds, Leeds, UK) [3]. This is the first majorclinical trial in patients with RA to use DAS28 remission asan endpoint, and involved patients with early, moderate-to-severe, active RA (DAS28 ≥3.2) who had not previouslyused methotrexate. Subjects were randomized tomethotrexate alone (n=268) or methotrexate plusetanercept 50 mg/week (n=274). At 1 year, 50% ofetanercept plus methotrexate recipients achieved remission(DAS28 <2.6) compared with 28% of those who receivedmethotrexate alone (p<0.001). In addition, a greaterproportion of patients in the combination therapy groupachieved low disease activity (DAS28 ≤3.2, 64% vs. 41%;p<0.001). Radiographic non-progression was achieved in80% of etanercept plus methotrexate recipients and 59% ofthe methotrexate-only group (p<0.001). Rates of severeadverse events were similar between the two treatmentgroups. These trial results have subsequently been publishedin the Lancet [4].

An interim analysis of a German, 5-year, multicenter,prospective study of the efficacy and safety of adalimumabin the setting of day-to-day rheumatology practice waspresented [5]. A total of 4640 patients with long-standingRA (mean disease duration 12 years) were included atbaseline. After 24 months of adalimumab therapy (data for824 patients were available), 20% of patients were in clinicalremission (DAS28 <2.6) and considerable improvements inphysical function were also achieved. Adalimumab wasgenerally well-tolerated among these patients.

The multicenter trial, BeSt (Behandel Strategieen),addressed the effect of tight control of disease with the aimof determining the optimal treatment strategy for early RApatients. Five-year data from this investigation werepresented by Dr Naomi Klarenbeek (Leiden UniversityMedical Center) [6]. Longitudinal data analysis showed thatthose treated with an initial combination of methotrexateand infliximab had significantly better Health Assessment

Annual European Congress ofRheumatology (EULAR 2008)

Paris, France, 11–14 June 2008

The Remedica Editorial Team

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 69

MEE

TIN

G R

EPO

RT


Questionnaire scores, and a slightly higher drug-freeremission rate at 5 years, compared with those who receivedsequential monotherapy or step-up combinationtherapy regimens.

Safety of anti-TNFsEtanerceptThe safety of long-term etanercept use was assessed in ananalysis of patients with DMARD-refractory RA or early RAwho were enrolled in open-label extensions of double-blind,controlled trials in Europe and North America [7]. The resultswere based on a total of 2054 patients and 9763 patient-years of etanercept exposure, with 57–71% and 35–43% ofpatients estimated to be continuing on etanercept at 3 yearsand 9 years (in ongoing North American studies),respectively. The overall rates of severe adverse events in theRA patients were similar to the severe adverse event rates incontrol patients from the double-blind phase of the includedrandomized controlled clinical trials. The overall rates ofserious infections were similar to those seen incontrol/methotrexate-treated patients in earlier trials. Thestandard incidence ratio (SIR) for malignancy was 0.98(95% confidence interval [CI] 0.78–1.22), calculated usingmatched data from the National Cancer InstituteSurveillance Epidemiology and End Results (SEER) database;however, the SIR for lymphoma was 3.53 (95% CI1.82–6.16). The researchers stated that “it is currentlyunknown if the higher-than-expected rate of lymphoma isrelated to TNF antagonist exposure or if it reflects theelevated risk of lymphoma in patients with RA”.

The risk of malignancy associated with etanercept therapyin RA was also investigated by Eric Matteson (Mayo Clinic,Rochester, MN, USA) and colleagues, who presented a meta-analysis of randomized controlled trials [8]. From a total of3316 patients, 2233 received etanercept and 1072 receivedcontrol therapy. Malignancies were detected in 26 etanerceptrecipients and in seven controls, resulting in a hazard ratio formalignancy of 1.84 (95% CI 0.79–4.28), suggesting anincreased risk of malignancy in etanercept-treated patients.However, this was not statistically significant and the authorsconcluded that the small number of cancers precluded theirability to reach definitive conclusions.

An analysis of RA patients included in a Germanbiologicals register assessed the use of biological agents indaily rheumatology practice in patients with a history ofcancer [9]. The investigators found that patients receivedbiological agents irrespective of prior malignancies, and thetime elapsed since the onset of the malignancy. The rate ofmalignancy recurrence was similar in patients treated withbiological agents and in those who received traditionalDMARD therapy.

AdalimumabReAlise (A Five-Year, Post-Marketing Observational Study toFollow-up Patients with Rheumatoid Arthritis FormerlyTreated in Study M02-497 [ReAct] and SubsequentlyPrescribed Humira) was undertaken to evaluate the long-term safety and effectiveness of adalimumab for up to 5years in patients with RA who had previously participated inReAct (A Study to Assess the Safety and Efficacy ofAdalimumab when Added to Inadequate Standard Anti-Rheumatic Therapy in Patients with Active RheumatoidArthritis) [10]. An interim analysis of ReAlise was presentedby Gerd Burmester (Charité University Hospital of the FreeUniversity and Humboldt University of Berlin, Germany)[11]. Data were available for 3421 patients. The rapid initialclinical improvements with adalimumab treatment observedin the ReAct study were sustained in ReAlise. Overall,11.7% of patients (n=400) withdrew from the study, 4.0%for adverse events and 5.7% for lack of efficacy. No newsafety signals were observed in this interim analysis, and therates of serious infections were low. The SIR, using the SEERdatabase for comparison, was 0.89 (95% CI 0.68–1.13) forall malignancies; however, similarly to that described abovefor etanercept, the SIR for malignant lymphoma was 3.81(95% CI 1.90–6.81).

An open-label extension of ATLAS (Adalimumab TrialEvaluating Long-term Efficacy and Safety for AnkylosingSpondylitis), presented by Desireé van der Heidje (LeidenUniversity Medical Center), demonstrated that thestatistically significant reduction in AS signs and symptoms(20% improvement in Assessment in Spondyloarthritisinternational Society criteria) observed at 24 weeks (in thedouble-blind phase) was sustained at up to 3 years oftreatment [12]. Comparison of adverse events and severeadverse events rates between the double-blind and 3-yearexposure periods, identified no significant differencesbetween the two treatment periods. In addition, there wereno cases of tuberculosis or demyelinating disorders.

Data on radiographic progression in patients entered intoan open-label extension of the 24-week, Phase III trial,ADEPT (Adalimumab Effectiveness in Psoriatic Arthritis Trial),were presented by Philip Mease (Seattle RheumatologyAssociates, Seattle, WA, USA) [13]. In initial adalimumabresponders in the ADEPT study, continued treatment withopen-label adalimumab led to sustained inhibition ofradiographic progression at up to 144 weeks.

Novel agentsGolimumabData from Phase III trials of the anti-TNF antibody,golimumab, were presented at the Congress. The results ofGO-FORWARD (Golimumab for Subjects with Active RA

REMEDICA EDITORIAL TEAM

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 70

EULAR 2008


Despite methotrexate), presented by Edward Keystone(Toronto University, Toronto, ON, Canada), showed thatgolimumab plus methotrexate was significantly moreefficacious than methotrexate alone in subjects with activeRA [14]. In GO-AFTER (Golimumab after Former anti-TNFTherapy Evaluated in RA), 461 patients with active RA whohad previously received at least one anti-TNF agent (65%had failed one, 25% failed two, and 10% failed three) wererandomized to placebo, golimumab 50 mg subcutaneously,or 100 mg subcutaneously every 4 weeks. Josef Smolen(University of Vienna, Vienna, Austria) presented data fromthe study [15]. Prior anti-TNF therapy had beendiscontinued due to lack of efficacy in approximately 60%of patients. Among this subgroup, 35.7% and 42.7% ofpatients in the golimumab 50 mg and golimumab 100 mggroups had a 20% improvement in American College ofRheumatology criteria (ACR20 response) at week 14compared with 17.7% of the placebo group (p=0.006 andp<0.001, respectively). Adverse event and severe adverseevent rates were similar across the treatment groups. Thisstudy is the first to demonstrate that a patient who fails onup to three anti-TNF-α agents may be successfully treatedwith another.

Arthur Kavanaugh (University of California–San Diego,La Jolla, CA, USA) presented data from GO-REVEAL (AMulticenter, Randomized, Double-Blind, Placebo-ControlledTrial of Golimumab, a Fully Human Anti-TNF-α MonoclonalAntibody, Administered Subcutaneously in Subjects withActive Psoriatic Arthritis) [16]. Golimumab was significantlymore efficacious in improving the signs and symptoms ofpsoriatic arthritis (PsA) at week 24 compared with placebo,and its efficacy was maintained at week 52. At week 24,2.4% of golimumab-treated patients experienced severeadverse events compared with 6.2% of placebo recipients.In a separate presentation of data from GO-REVEAL [17],treatment with golimumab significantly improved psoriaticnail changes, dactylitis score, and enthesitis score in patientswith active PsA (all p<0.001 compared with placebo).

Tocilizumab Several Phase III trials of tocilizumab, an anti-interleukin-6(anti-IL-6) receptor antibody, were presented at this year’sCongress. The AMBITION (Actemra™ [Tocilizumab] versusMethotrexate Double-Blind Investigative Trial inMonotherapy) investigators assessed the efficacy and safetyof tocilizumab monotherapy versus methotrexatemonotherapy in patients with active RA who had not failedprevious methotrexate or biological treatment. Tocilizumabmonotherapy (8 mg/kg every 4 weeks) was found to besuperior to a standard escalating dose of methotrexate, withhigher proportions of patients achieving ACR20, ACR50, and

ACR70 responses at week 24 [18]. Professor Graeme Jones(University of Tasmania, Tasmania, Australia) and colleaguesalso reported safety data from the study. Severe adverseevents and serious infections were slightly higher in thosetreated with tocilizumab compared with methotrexate (4%vs. 3% and 1.4% vs. 0.7%, respectively). In contrast,elevation in liver enzymes occurred more frequently in themethotrexate group than in the tocilizumab arm (4% vs. 2%).

The results of RADIATE (A Randomized, Double-BlindStudy of Safety and Reduction in Signs and SymptomsDuring Treatment With Tocilizumab Versus Placebo, inCombination With Methotrexate, in Patients With Moderateto Severe Active Rheumatoid Arthritis and InadequateResponse to Anti-TNF Therapy) were presented by ProfessorEmery. Tocilizumab plus methotrexate therapy wasdemonstrated to be efficacious in the treatment of refractoryRA patients, irrespective of the number of, or most recentlyfailed anti-TNF treatment [19].

INCB18424 William Williams (Incyte Inc., Wilmington, DE, USA) andcolleagues presented results from a 28-day proof-of-concepttrial of INCB18424, an orally available janus-associatedkinase (JAK) inhibitor, in eight patients with active RA [20].Six subjects received the inhibitor, while two receivedplacebo. Four of the six INCB18424-treated subjectsachieved at least an ACR20, three achieved at least anACR50, and two achieved an ACR90 response. Threepatients achieved a DAS28 score <2.6, and the mean DAS28score decreased from 6.0±0.32 to 3.29±0.99 by day 28.This is the first demonstration of clinical activity of aselective JAK inhibitor in RA.

References1. van der Woude D, Brand R, Toes RE et al. Risk factors for DMARD-free remission in RA,

lessons from an inception cohort. Ann Rheum Dis 2008;67(Suppl II):48.

2. Kuper I, Hoekstra M, ten Klooster P et al. Remission can be achieved in 50% of earlyrheumatoid arthritis patients after 25 weeks in daily clinical practice. Ann Rheum Dis2008;67(Suppl II):48.

3. Emery P, Breedveld F, Hall S et al. Clinical remission, radiographic non-progression, andnormalized function with the combination of etanercept and methotrexate in thetreatment of early active rheumatoid arthritis: 1-year results of the COMET trial. AnnRheum Dis 2008;67(Suppl II):50.

4. Emery P, Breedveld FC, Hall S et al. Comparison of methotrexate monotherapy with acombination of methotrexate and etanercept in active, early, moderate to severerheumatoid arthritis (COMET): a randomised, double-blind, parallel treatment trial. Lancet2008; [Epub ahead of print].

5. Tony H, Wassenberg S, Krause A et al. Adalimumab (Humira®) is effective and well-tolerated in patients with rheumatoid arthritis (RA): disease activity and physical functionresults from a German database of RA patients receiving adalimumab. Ann Rheum Dis2008;67(Suppl II):188.

6. Klarenbeek NB, Güler-Yüksel M, van der Kooij SM et al. Clinical outcomes of four differenttreatment strategies in patients with recent-onset rheumatoid arthritis: 5-years results ofthe BEST-study. Ann Rheum Dis 2008;67(Suppl II):187.

7. Klareskog L, Moreland LW, Cohen SB et al. Safety and efficacy of over 10 years ofcontinuous etanercept therapy in patients with rheumatoid arthritis in North America andEurope. Ann Rheum Dis 2008;67(Suppl II):175.

8. Bongartz T, Warren FC, Mines D et al. Etanercept therapy in rheumatoid arthritis and therisk of malignancies. Ann Rheum Dis 2008;67(Suppl II):51.

9. Strangfeld A, Listing J, Herzer P et al. RA patients with prior malignancy under treatmentwith biologics. Ann Rheum Dis 2008;67(Suppl II):332.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 71

REMEDICA EDITORIAL TEAM


10. Bombardieri S, Ruiz AA, Fardellone P et al.; Research in Active Rheumatoid Arthritis(ReAct) Study Group. Effectiveness of adalimumab for rheumatoid arthritis in patients witha history of TNF-antagonist therapy in clinical practice. Rheumatology (Oxford)2007;46:1191–9.

11. Burmester GR, Matucci Cerinic M, Mariette X et al. Safety and effectiveness of adalimumab(Humira®) is maintained in patients with rheumatoid arthritis — three-year results of realise, apost-marketing observational study. Ann Rheum Dis 2008;67(Suppl II):176.

12. van der Heijde D, Sieper J, Dijkmans BA et al.; for the ATLAS Study Group. Three-year safetyand efficacy results from the adalimumab (Humira®) trial evaluating long-term efficacy andsafety in ankylosing spondylitis (ATLAS). Ann Rheum Dis 2008;67(Suppl II):519.

13. Mease PJ, Ory P, Sharp JT et al. Adalimumab (Humira®) sustains inhibition of joint damagefor more than 2 years in patients with psoriatic arthritis (PsA). Ann Rheum Dis2008;67(Suppl II):99.

14. Keystone E, Genovese MC, Klareskog L et al. Golimumab, a new human anti-TNF-alphamonoclonal antibody, administered subcutaneously every 4 weeks in patients with activerheumatoid arthritis despite methotrexate: week 24 results of the randomized, double-blind, placebo-controlled, GO-FORWARD study. Ann Rheum Dis 2008;67(Suppl II):185.

15. Smolen J, Kay J, Doyle MK et al. Golimumab, a new human anti-TNF-alpha monoclonalantibody, subcutaneously administered every 4 weeks in patients with active rheumatoidarthritis who were previously treated with anti-TNF-alpha agent(s): results of therandomized, double-blind, placebo-controlled, GO-AFTER study. Ann Rheum Dis2008;67(Suppl II):50.

16. Kavanaugh A, Mease P, Krueger GG et al. Golimumab, a new, human, TNF alpha antibodyadministered subcutaneously every 4 weeks in psoriatic arthritis patients: 52-week efficacyand safety results of the randomized, placebo-controlled GO-REVEAL study. Ann RheumDis 2008;67(Suppl II):99.

17. Gladman D, Kavanaugh A, McInnes I et al. Golimumab, a new human TNF-alphaantibody, administered every 4 weeks as a subcutaneous injection in psoriatic arthritis: nail,enthesitis, and dactylitis response in the randomized, placebo-controlled, GO-REVEALstudy. Ann Rheum Dis 2008;67(Suppl II):526.

18. Jones G, Gu JR, Lowenstein M et al. Tocilizumab monotherapy is superior to methotrexatemonotherapy in reducing disease activity in patients with rheumatoid arthritis: theambition study. Ann Rheum Dis 2008;67(Suppl II):89.

19. Emery P, Keystone E, Tony H et al. Tocilizumab (TCZ) significantly improves diseaseoutcomes in patients with rheumatoid arthritis whose anti-TNF therapy failed: the radiatestudy. Ann Rheum Dis 2008;67(Suppl II):127.

20. Williams WV, Scherle P, Shi J et al. Initial efficacy of INCB018424, a selective janus kinase1& 2 (JAK1&2) inhibitor in rheumatoid arthritis (RA). Ann Rheum Dis 2008;67(Suppl II):62.

RT404_2_Rheum_6.2_05.qxd 10/9/08 16:44 Page 72

Reader Survey – Let Us Know What You Think!Please visit the website and complete the survey online (registration online is free for healthcare professionals):www.advancesinrheumatology.com Or you can photocopy this page, complete the survey below, and fax it toRemedica on (312) 372 0217 from the US or on +44 (0)20 7759 2951 from Europe and the rest of the world.

1. We are aiming to provide practical information for rheumatologists and associated healthcare professionals.

How would you rate the information presented in this issue? STRONGLY DISAGREE STRONGLY AGREE

a) The technical quality of information included in the International

Journal of Advances in Rheumatology was high: 1 2 3 4 5

b) The information was relevant to my practice: 1 2 3 4 5

c) The information was presented clearly: 1 2 3 4 5

d) The leading articles provided new information regarding the

understanding and treatment of rheumatological disease: 1 2 3 4 5

e) The literature analysis was helpful and I would

like to see analyses in future issues: 1 2 3 4 5

2. Did you learn anything from the International Journal of Advances in Rheumatology

that will change the way you practice? ■■ Yes ■■ No

If so, what? ............................................................................................................................................................................................................

3. What specific topics do you think should be covered in future issues?

................................................................................................................................................................................................................................

4. Have you ever visited our website www.advancesinrheumatology.com? ■■ Yes ■■ No

If yes, is there anything you think we could do to improve it?

................................................................................................................................................................................................................................

5. Would you like to recommend the International Journal of Advances in Rheumatology to a colleague? ■■ Yes ■■ No

My colleague’s email is: ........................................................................................................................................................................................

Name .......... .......................................................................................... Job title ..........................................................................................

Institution .... ..........................................................................................................................................................................................................

Address ....... ..........................................................................................................................................................................................................

Country ..................................................................................................... Post/zip code .................................................................................

Email ........... ..........................................................................................................................................................................................................

INTERNATIONAL JOURNAL OF ADVANCES IN

RHEUMATOLOGY

RT404_2_Rheum_6_2_COV_US_02.qxd 8/29/08 5:11 PM Page 3

www.advancesinrheumatology.com

RT404_2_Rheum_6_2_COV_US_02.qxd 8/29/08 5:11 PM Page 4

rt404 2 rheum 6.2 05 - uk healthcare cecentral · raimon sanmartí, maria victoria hernández,...

Documents