Clin Drug Investig (2013) 33 (Suppl 2):S113–S132 DOI 10.1007/s40261-012-0036-y

Section 3: Rheumatoid Arthritis

Ó Springer International Publishing Switzerland 2013

The Complex Interplay Between Rheumatoid Arthritis and Atopy: Focus on Anti-TNFa Treatment Maria Sole Chimenti,1 Gioia Di Muzio,1 Maria Domenica Guarino,1 Paola Conigliaro,1 Elisabetta Greco,1 Roberto Perricone1 1 Rheumatology, Allergology and Clinical Immunology; Department of Internal Medicine, University of Rome Tor Vergata, Via Montpellier 1, 00133 Rome, Italy

Name and address for correspondence: Maria Sole Chimenti, Department of Internal Medicine, Unit of Rheumatology, University of Rome Tor Vergata, Via Montpellier 1, 00163 Rome, Italy. Tel: +39 06 72596287; Fax: +39 06 20900358; e-mail: maria.sole.chimenti@ uniroma2.it

It is well known that balance between CD4+ helper T cell subsets (Th1 and Th2) is crucial for a healthy immune response and plays a role in the pathological mechanisms underlying autoimmune diseases and atopy. In particular, rheumatoid arthritis (RA) can be considered a Th1-mediated disease, while Th2 cell responses predominate in atopic disorders [1]. As many studies support the concept of a mutual Th1/Th2 antagonism, in recent years there has been considerable interest in searching for a relationship between the development of allergic diseases and the course of autoimmune diseases as RA. Rudwaleit et al. [2] studied the prevalence of atopic disorders in RA and in ankylosing spondylitis (AS) observing that they exhibit divergent T helper (Th) cytokine patterns. They found that atopic disorders are decreased in RA but non-significantly increased in SA, and hypothesized that atopy may suppress the severity but not necessarily the onset of autoimmunity. Other authors suggest a possible coexistence of atopic disorders and

rheumatological processes, despite the different cytokine patterns. Kero et al. [3] for example, stated that Th1 and Th2 diseases could be compatible, indicating a common environmental factor behind the disease process. On the other hand, Hartung et al. [4] demonstrated that RA and atopy antagonize each other and that a change in cytokine patterns could provide an indication for curative effects on RA. Thus, patients with atopic diseases may develop a less severe form of RA. This phenomenon can be explained by the identification of other lymphocyte subsets, such as Th17 T cells and regulatory T cells (Treg) [5]. Th17 cells differentiate from naı¨ve CD4+ T cells, and Th1 and Th2 can antagonize the differentiation of Th17. Furthermore, Th17 plays a central role in the pathogenesis of several autoimmune diseases as RA. The increasing prevalence of both autoimmune and atopic disease are probably due to dysfunction of regulatory T cells that may control either type of pathology [6]. In addition, it as been demonstrated that mature Th1 cells cannot be switched and the transient attenuation of RA in atopic subjects during the pollen season, for example, may be due to suppressed Th1 inflammation, rather than a change in phenotype [7]. There is evidence to indicate that interleukin (IL)-4 secreted by mast cells may not only trigger their apoptotic death, but also suppress the secretion of tumour necrosis factor (TNF) a and IL-1b by synovial macrophages [8, 9]. The use of TNFa antagonists (infliximab, etanercept and adalimumab) in the treatment of RA and spondyloarthritis (SpA) has changed the course of these diseases. They have changed the treatment paradigm by making remission achievable. The next therapeutic goal will be maintaining remission, restoring a state of immune tolerance: an example is that patients with clinical improvement after Infliximab therapy have an increase in the percentage of Treg cells [10].

S114

The role of immune reactions in adverse events that occur during treatment with anti-TNFa agents is well known. The hypothesis is that during anti-TNFa treatment, patients could develop antibodies against anti-TNFa. Monoclonal antibodies (mAb) against TNFa in particular can induce anti-drug antibodies despite having been shown to be generally effective and safe. The presence of these anti-drug antibodies is associated with low-to-absent serum mAb concentrations, with a reduction of clinical response and drug efficacy. This has been described for both infliximab and adalimumab but not for etanercept, probably due to its different structure [11]. The mechanism leading this process is the clearance of mAb by immune complex (IC) formation (drug–anti-drug antibody complexes) [12]. Furthermore, the incidence of anti-drug antibodies with or without ICs could be associated with adverse events. The presence of infliximab–anti-infliximab complexes is associated with infusion-related allergic reactions, and higher concentrations of anti-infliximab antibodies were associated with a higher risk of infusion reactions [11]. A review of the literature showed that hypersensitivity reactions to TNFa-blocking agents are not uncommon [13]. These manifestations should be considered as an immunoglobulin E (IgE)-mediated reaction or a delayed reaction mediated by IgG and complement. This can be activated because of a complex interplay occurring in RA patients treated with anti-TNFa due to changes of lymphocyte subset that, although still not demonstrated, does not appear to be due to the atopic status of the patient [14]. In clinical practice, the presence of anti-drug antibodies is not monitored; therefore, many patients continue mAb treatment in the presence of anti-drug antibodies, resulting in the ongoing formation of ICs and the possible hazard of continuous stimulation of the immune system by these ICs. This could explain the relationship between Th1 and Th2 and the intricate immunological basis of associated side effects [9, 11]. Moreover, we noted, in a large cohort of RA patients referred to our out-patient department of Rheumatology of the University of Rome ‘‘Tor Vergata’’, a reduction in NK absolute cell number and frequency in the PB compared with healthy subjects. Indeed, an early and permanent increase of NK cells was demonstrated in RA patients treated with etanercept, suggesting the attempt to restore the innate immune compartment (unpublished results). Although NK cells represent an arm of the innate immunity, recent data underpin a role in the responses of acquired immunity [15]. In fact, in inflammatory arthritis, NK cells may produce pro-inflammatory Th1 cytokines, creating and/or maintaining the inflammatory response. We can conclude that highlighting the need to better understand the prevalence of allergy in patients with

inflammatory arthritis (RA, SpA and psoriatic arthritis) by studying patients’ clinical and laboratory findings as hay fever, urticaria and asthma, total and specific serum IgE levels, and skin and patch tests. The relevance of antibodies against anti-TNFa (specific Ig isotypes) should be correlated with levels of complement fragments and complement cleavage products and with activation of mast cells, both in patients with atopic disease and in patients with loss of efficacy due to the presence of ICs with anti-drug antibodies. Furthermore, in clinical practice we should consider the correlation between anti-TNFa treatment and evidence of modification of subpopulations of lymphocytes as efficacy of the treatment in re-establishing the immune tolerance. There is evidence that blocking an individual inflammatory cytokine affects immune tolerance, either directly or indirectly [10]. Acknowledgments Native English editing was provided by Raewyn Poole of inScience Communications, Springer Healthcare. This editorial support was funded by Pfizer. The authors declare no conflicts of interest.

References 1. Panayi GS, Corrigall VM, Pitzalis C. Pathogenesis of rheumatoid arthritis. The role of T cells and other beasts. Rheum Dis Clin North Am. 2001;27(2):317–34. 2. Rudwaleit M, Andermann B, Alten R, et al. Atopic disorders in ankylosing spondylitis and rheumatoid arthritis. Ann Rheum Dis. 2002;61(11):968–74. 3. Kero J, Gissler M, Hemminki E, Isolauri E. Could TH1 and TH2 diseases coexist? Evaluation of asthma incidence in children with coeliac disease, type 1 diabetes, or rheumatoid arthritis: a register study. J Allergy Clin Immunol. 2001;108(5):781–3. 4. Hartung AD, Bohnert A, Hackstein H, et al. Th2-mediated atopic disease protection in Th1-mediated rheumatoid arthritis. Clin Exp Rheumatol. 2003;21(4):481–4. 5. Steinman L. A brief history of Th17, the first major revision in Th1/Th2 hypothesis of T cell-mediated tissue damage. Nat Med. 2007;13(2):139–45. 6. Verhoef CM, van Roon JA, Vianen ME, et al. Mutual antagonism of rheumatoid arthritis and hay fever; a role for type 1/type 2 T cell balance. Ann Rheum Dis. 1998;57(5):275–80. 7. Oskeritzian CA, Wang Z, Kochan JP, et al. Recombinant human (rh)IL-4-mediated apoptosis and recombinant human IL-6-mediated protection of recombinant human stem cell factor-dependent human mast cells derived from cord blood mononuclear cell progenitors. J Immunol. 1999;163(9):5105–15. 8. Shelburne CP, Ryan JJ. The role of Th2 cytokines in mast cell homeostasis. Immunol Rev. 2001;179:82–93. 9. Rabin RL, Levinson AI. The nexus between atopic disease and autoimmunity: a review of the epidemiological and mechanistic literature. Clin Exp Immunol. 2008;153(1):19–30. 10. Albani S, Koffeman EC, Prakken B. Induction of immune tolerance in the treatment of rheumatoid arthritis. Nat Rev Rheumatol. 2011;7(5):272–81. 11. Korswagen LA, Bartelds GM, Krieckaert CL, et al. Venous and arterial thromboembolic events in adalimumab-treated patients with antiadalimumab antibodies: a case series and cohort study. Arthritis Rheum. 2011;63(4):877–83.

S115 12. Bartelds GM, Krieckaert CL, Nurmohamed MT, et al. Development of antidrug antibodies against adalimumab and association with disease activity and treatment failure during long-term follow-up. JAMA. 2011;305(14):1460–8. 13. Posadas SJ, Pichler WJ. Delayed drug hypersensitivity reactions—new concepts. Clin Exp Allergy. 2007;37(7):989–99. 14. Bennett AN, Wong M, Zain A, et al. Adalimumab-induced asthma. Rheumatology (Oxford). 2005;44(9):1199–200. 15. Vivier E, Raulet DH, Moretta A, et al. Innate or adaptive immunity? The example of natural killer cells. Science. 2011;331(6013):44–9.

The Pathogenic Role of the Complement System in Rheumatoid Arthritis and Its Relationship with TNFa Inhibitors

interaction between TNFa and the complement system may contribute to the pathogenesis of these diseases. Reduction of complement activation could be one of the mechanisms by which TNFa inhibitors exert their effects in inflammatory arthritides [7]. Because of these findings, the complement system could be an attractive therapeutic target in RA. The objectives of this study were to investigate the complement system in patients with RA and to evaluate the relationships between complement components and autoantibodies (RF and ACPA), inflammatory markers and clinical response in a subgroup of RA patients treated with anti-TNFa agents [8].

Methods Gioia Di Muzio,1 Eleonora Ballanti,1 Maria Sole Chimenti,1 Gianfranco Gigliucci,1 Lucia Novelli,1 Barbara Kroegler,1 Roberto Perricone1 Rheumatology, Allergology and Clinical Immunology, Department of Internal Medicine, University of Rome Tor Vergata, Rome, Italy Address for correspondence: Rheumatology, Allergology and Clinical Immunology, Department of Internal Medicine, University of Rome Tor Vergata, Rome, Italy. Tel: +39 06 72596287; Fax +39 06 20900358; e-mail: [email protected]

Introduction The complement system is an integral part of the immune defence system, essential against pathogens and in immune complex clearance [1]. It comprises more than 30 plasmaand membrane-bound proteins and can be activated through three pathways: the classical, the alternative and the lectin pathways. Its activation contributes to the pathogenesis of several autoimmune and inflammatory conditions. Synthesis of complement components occurs in chronically inflamed tissues such as the rheumatoid synovial tissue (lining cells, fibroblasts, mononuclear phagocytes and endothelial cells) [2]. Levels of native complement proteins are generally depressed in the synovial fluid of patients with rheumatoid arthritis (RA), reflecting consumption of complement [3]. On the other hand, elevated levels of several complement cleavage (split) products and low levels of complement inhibitors have been observed in the synovial fluid and synovial tissue of patients with RA. Autoantibodies (rheumatoid factor, RF; anti-citrullinated protein antibodies, ACPA) and the complement system are involved in the pathogenesis of RA [4]. ACPA and antiTNFa agents are capable of in vitro modulation of complement activity [5], but there is a paucity of in vivo data. It is well known that anti- TNFa agents are effective for the treatment of RA [6], and some studies suggest that the

This retrospective, observational prospective, observational study included patients with RA diagnosed according to 1987 American College of Rheumatology (ACR) criteria, and healthy, age- and gender-matched volunteers as controls. This study only considered data obtained from patients of the UOC of Rheumatology of Policlinico Tor Vergata, University of Rome, Tor Vergata, Italy, after the approval of the ethics committee. All enrolled patients gave informed consent. The statistical analysis was performed using GraphPad 5.0 statistical software (GraphPad Prism, San Diego, CA, USA).Two-tailed P values were reported, P values B0.05 were considered significant [8]. All enrolled patients were naı¨ve for biologic therapy at enrolment, as required for the study. Serological analysis included erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), complement C3, C4 and CH50, and autoantibodies RF (by nephelometry, cut-off [40 UI/mL) and ACPA (by ELISA, cut-off [20 U/mL). Split products of C3 and complement factor B were assessed by immunoelectrophoresis and counterimmunoelectrophoresis at baseline and at weeks 14 and 22. A subgroup of patients with RA started anti-TNFa treatments because inadequate responders to conventional DMARDs therapy and were also assessed after 14 and 22 weeks. Disease activity was measured with DAS28ESR and the response to therapy was assessed according to European League Against Rheumatism (EULAR) criteria.

Results The study included 114 patients with RA (89 women, 78%) and 30 healthy volunteers as controls. All the enrolled patients were naı¨ve for biological therapy at the time of the study; 96% of them were treated with at least one diseasemodifying antirheumatic drug (DMARD) with or without

S116

prednisone, and 4% were treated with prednisone alone. Seventy-six patients started anti-TNFa treatments (27 adalimumab and 49 etanercept). At baseline, C3 and C4 levels were significantly higher in patients with RA than in the control group (P values of 0.0012 and 0.0003 by paired t-test, respectively for C3 and C4), and CH50 levels were similar. No demographic or clinical differences were observed between ACPA-positive (n = 76) and ACPA-negative (n = 38) patients at baseline (data not shown), and no significant differences in complement (C3, C4 and CH50) levels were seen (Table 1). Even after stratifying for the presence or absence of RF, no significant differences were observed (data not shown). No split products of C3 and factor B were detected in RA patients either at baseline, or at 14 and 22 weeks.

Table 1 baseline

Laboratory parameter of controls versus RA patients at Controls (n = 30)

ESR (0–20 mm/h)

/

30.6 ± 22.2

/

CRP (0–5 mg/L)

/

13.7 ± 17.3

/

RF (0-40 U/L) ACPA (0–20 U/L)

/ /

235.9 ± 553.5 132.2 ± 251.1

/ /

C3 (mg/dL)

110 ± 25

127.9 ± 26.5

0.0012

C4 (mg/dL)

22.7 ± 8.3

29.7 ± 10.2

0.0003

CH50 (U/mL)

130.0 ± 31.2

DAS28

/

 

Baseline of RA cohort (n = 114)

P value 

Laboratory parameter (mean ± SD)

138.0 ± 44.4 4.8 ± 1.2

NS /

Paired t-test (controls vs. baseline of the 114 RA patients)

In patients undergoing anti-TNFa therapy, C3, C4 and RF levels were significantly reduced after 22 weeks (p \ 0.003, p \ 0.005 and p \ 0.04, respectively), while CH50 levels remained unchanged. There was a statistically significant decrease in C3 and C4 levels at both 14 and 22 weeks. Moreover, we observed that after 14 and 22 weeks of therapy with TNFa inhibitors, the levels of C3, C4 and CH50 were not significantly different from those of healthy controls. No associations between C3, C4 or CH50 levels and ACPA levels were observed at 14 or 22 weeks of follow-up (Table 2). Levels of RF showed a statistically significant decrease at 22 weeks of follow-up (Pc = 0.04, Table 2), independent from the anti-TNF agent administered. When stratifying for ACPA or RF, no significant differences were observed in mean complement C3, C4 or CH50 levels at baseline and the decrease in C3 and C4 was similar between the groups (data not shown). No statistical analysis was performed to establish any correlation between complement, ESR and CRP. Disease activity significantly improved at both 14 and 22 weeks of anti-TNF therapy showing an average of a moderate EULAR response (P \ 0.0001). Eighteen per cent achieved a good response, 36% a moderate response, and 47% low or no response at 22 weeks. Variation in complement levels did not correlate with changes in disease activity measured by DAS28ESR. A tendency of a higher reduction in complement C3 levels was observed in patients with good EULAR response. Patients with baseline lower (although normal) C3 levels appeared to be also those who had a better EULAR response.

Table 2 Laboratory parameters of patients with rheumatoid arthritis receiving anti-TNFa therapy at baseline and after 14 and 22 weeks of treatment Parameter (cut-off), mean ± SD

Baseline (n = 76)

14 weeks (n = 76)

ESR (0–20 mm/h)

30.3 ± 22.5

20.7 ± 16.4

CRP (0–5 mg/L)

12.7 ± 16.7

5.3 ± 8.1

RF (0–40 U/L) ACPA (0–20 U/L)

290.6 ± 663.9 129.5 ± 208.1

212.2 ± 457.2 151.0 ± 214.5

C3

127.6 ± 27.3

C4

29.2 ± 9.9 134.3 ± 27.0

134.7 ± 32.7

4.8 ± 1.3

4.0 ± 1.2

CH50 DAS28ESR

Pc value* vs. baseline 0.001 \0.0001

22 weeks (n = 76)

Pc value** vs. baseline

19.5 ± 17.1

\0.0001

5.3 ± 7.5

\0.0001

NS NS

135.7 ± 256.1 147.9 ± 217.0

115.7 ± 22.4

0.003

114.6 ± 25.2

0.001

25.8 ± 8.3

0.001

24.8 ± 8.3

0.005

NS \0.0001

138.6 ± 32.0 4.1 ± 1.4

0.04 NS

NS \0.0001

One-way analysis of variance (ANOVA) was used to compare laboratory parameters among the baseline, 14 and 22 weeks of follow-up. Bonferroni post-test was used accordingly (Pc) ACPA anti-citrullinated protein antibodies, CRP C-reactive protein, DAS28ESR disease activity score using erythrocyte sedimentation rate, ESR erythrocyte sedimentation rate; Pc corrected P value, RF rheumatoid factor, SD standard deviation * One-way ANOVA corrected with Bonferroni (baseline anti-TNF treated patients vs. 14 weeks follow-up) ** One-way ANOVA corrected with Bonferroni (baseline anti-TNF-treated vs. 22 weeks follow-up)

S117

Discussion In our RA cohort we observed relatively high levels of C3, C4 and CH50 at baseline compared with controls. This evidence has been frequently observed in plasma of RA patients. It has been previously reported that in RA, as well as in other inflammatory conditions, the presence of abnormally, pathologically elevated complement C3 levels may reflect the presence of an underlining inflammatory process [9]. We can suggest that the relatively high amount of plasma C3 could be due to spill over from the joints probably reflecting an in situ overexpression or even to an abnormal production of complement proteins due to inflammation [10, 11]. At baseline, no differences were observed between the anti-TNF treated and non-treated patients, and treated patients had significantly higher mean baseline C3 and C4 levels than controls (P \ 0.001). No demographic or clinical differences were observed between ACPA+ and ACPA- patients. The relevance of complement levels in RA is unknown as opposed to low complements reflecting disease activity in SLE. Our data showed that in our cohort of RA patients, there was a baseline correlation between complement activity (CH50) and RF. In fact, we noticed a higher number of patients with CH50 below the normal values also having elevated or very elevated RF levels. Conversely, we didn’t find any association between C3, C4 and CH50 levels and the presence (elevated/normal/decreased) or absence of ACPA. This was observed at baseline, regardless of the DMARD used. We investigated the presence of split products of complement at baseline and after treatment to understand if changes in C3 and C4 levels after anti-TNF therapy could be due to complement activation or rather to normalization of previously high levels of these complement components. The eventual release of C3 and B, that were never detected in the patients either at baseline, or at 14 and 22 weeks, was evaluated to better address this issue. In the subgroup of patients receiving anti-TNFa therapy, we observed relatively high levels of C3 and C4 at baseline (no difference between biologic treated and non-treated patients) and a significant reduction in both after antiTNFa treatment. The decrease in C3 and C4 complement levels with treatment could be explained by complement activation or normalization of their previously high levels. We considered reduction of complement native components as an improvement of a pre-existing pro-inflammatory status, due to anti-TNFa drugs especially, because most of the complement activation occurs locally in the synovium. After 22 weeks of anti-TNF therapy, CH50 levels increased and negatively correlated with RF levels. Also, the number of patients with positive RF was significantly reduced at 22 weeks of anti-TNF therapy.

Furthermore, after therapy with TNF antagonists, changes in C3, C4 or CH50 levels did not correlate with variations in ACPA levels that remained stable during the 22 weeks of follow-up. Considering the relationship between complement and disease activity, we noticed that higher levels of C3 at baseline were significantly associated with worse EULAR response, so patients who had a lower reduction in C3 levels from baseline during the follow up were also those who achieved the lowest EULAR score. All treated patients showed an inverse correlation between C3 and EULAR response, suggesting that persistently elevated C3 levels may be considered a negative predictive factor influencing the outcome of anti-TNF therapy for RA. RF levels rather than ACPA seem to correlate with complement activity. In patients with RA receiving antiTNFa therapy, levels of complement components appear to correlate with disease activity, but not with ACPA. The observed decrease of C3 and C4 complement levels in patients treated with biological DMARDs could be explained as complement activation or normalization of previously high levels of these components. The presence of high levels of C3 in RA patients might be a negative prognostic factor ascribed to a pro-inflammatory status reversed by anti-TNFa therapy. Autoantibodies have the potential to activate the complement in vitro, and this was demonstrated for RF and recently for ACPA [6, 12]. However, the behaviour of complement system respect to these autoantibodies in vivo has not been fully addressed [5]. Trouw et al. [6] from in vitro observations, suggested that complement activation could be involved in RA pathogenesis in ACPA positive patients. This was not confirmed by our in vivo results that provided evidence that RF rather than ACPA may play a role in complement modulation. It has been previously reported that in RA, as well as in other inflammatory conditions, the presence of abnormally, pathologically elevated complement C3 levels may reflect the presence of an underlining inflammatory process [13]. We can suggest that the relatively high amount of plasma C3 could be due to spill over from the joints probably reflecting an in situ overexpression or even an abnormal production of complement proteins due to inflammation [14, 15]. This could explain the reduction in complement C3 and C4 levels in the absence of complement activating cleavage fragments observed by us after treatment with TNF antagonists.

Conclusion Reduction of native complement components should be considered an improvement of a pre-existing pro-inflammatory state. Anti-TNFa drugs have an outstanding

S118

anti-inflammatory potential that may be responsible for part of their mechanism in disease control. Complement C3 could be of prognostic value in the assessment of response to anti-TNFa therapy. These in vivo observational data may contribute to the determination of the mechanisms exerted by complement on RA pathogenesis. To conclude, specific modulation and inhibition of local complement production could be an attractive target for RA therapy. Acknowledgments Native English editing was provided by Raewyn Poole of inScience Communications, Springer Healthcare. This editorial support was funded by Pfizer.

12.

13.

14.

The authors declare no conflicts of interest. Authors’ contributions GDM designed the study, supervised the project and wrote the manuscript. EB and MSC performed the statistical analysis. GG, LN and BK contributed in selecting clinical and laboratory values of the patients. RP supervised the study and helped to write the manuscript.

References 1. Sjoberg AP, Trouw LA, Blom AM. Complement activation and inhibition: a delicate balance. Trends Immunol. 2009;30:83–90. 2. Moffat GJ, Lappin D, Birnie GD, et al. Complement biosynthesis in human synovial tissue. Clin Exp Immunol. 1989;78:54–60. 3. Swaak AJ, Van Rooyen A, Planten O, et al. An analysis of complement components in the synovial fluid in rheumatic diseases. Clin Rheumatol. 1987;6(3):350–7. 4. Brodeur JP, Ruddy S, Schwartz LB, et al. Synovial fluid levels of complement SC5b-9 and fragment Bb are elevated in patients with rheumatoid arthritis. Arthritis Rheum. 1991;34:1531–7. 5. Smolen JS. Autoantibodies in rheumatoid arthritis. In: van Venrooij WJ, Maini RN, editors. Manual of biological markers of disease. Dordrecht, The Netherlands: Kluwer Academic Publishers; 1996. 6. Trouw LA, Haisma EM, Levarht EW, et al. Anti-cyclic citrullinated peptide antibodies from rheumatoid arthritis patients activate complement via both the classical and alternative pathways. Arthritis Rheum. 2009;60:1923–31. 7. Maini R, St Clair EW, Breedveld F, et al. Infliximab (chimeric anti-tumour necrosis factor alpha monoclonal antibody) versus placebo in rheumatoid arthritis patients receiving concomitant methotrexate: a randomised phase III trial. ATTRACT Study Group. Lancet. 1999;354:1932–9. 8. Di Muzio G, Perricone C, Ballanti E, et al. Complement system and rheumatoid arthritis: relationships with autoantibodies, serological, clinical features, and anti-TNF treatment. Int J Immunopathol Pharmacol. 2011;24(2):357–66. 9. Firestein GS, Paine MM, Littman BH. Gene expression (collagenase, tissue inhibitor of metalloproteinases, complement, and HLA–DR) in rheumatoid arthritis and osteoarthritis synovium: quantitative analysis and effect of intraarticular corticosteroids. Arthritis Rheum. 1991;34:1094–105. 10. Sampietro T, Bigazzi F, Dal Pino B, et al. Up regulation of C3, C4, and soluble intercellular adhesion molecule-1 co-expresses with high sensitivity C reactive protein in familial hypoalphalipoproteinaemia: further evidence of inflammatory activation. Heart. 2004;90:1438–42. 11. Sze´plaki G, Proha´szka Z, Duba J, et al. Association of high serum concentration of the third component of complement (C3) with

15.

pre-existing severe coronary artery disease and new vascular events in women. Atherosclerosis. 2004;177:383–9. Watson WC, Cremer MA, Wooley PH, et al. Assessment of the potential pathogenicity of type II collagen autoantibodies in patients with rheumatoid arthritis: evidence of restricted IgG3 subclass expression and activation of complement C5 to C5a. Arthritis Rheum. 1986;29:1316–21. Firestein GS, Paine MM, Littman BH. Gene expression (collagenase, tissue inhibitor of metalloproteinases, complement, and HLA–DR) in rheumatoid arthritis and osteoarthritis synovium: quantitative analysis and effect of intraarticular corticosteroids. Arthritis Rheum. 1991;34:1094–105. Sampietro T, Bigazzi F, Dal Pino B, et al. Up regulation of C3, C4, and soluble intercellular adhesion molecule-1 co-expresses with high sensitivity C reactive protein in familial hypoalphalipoproteinaemia: further evidence of inflammatory activation. Heart. 2004;90:1438–42. Sze´plaki G, Proha´szka Z, Duba J, et al. Association of high serum concentration of the third component of complement (C3) with pre-existing severe coronary artery disease and new vascular events in women. Atherosclerosis. 2004;177:383–9.

Tumour Necrosis Factor Alpha (TNFa) Cardiomyocyte Preconditioning: Clues to Explain Anti-TNFa Agents Contraindication in Severe Heart Failure Fabio Cacciapaglia,1,2 Emanuela Salvatorelli,2,3 Luca Navarini,2,4 Antonella Afeltra,2,4 Pierantonio Menna,2,3 Giorgio Minotti2,3 1 Internal Medicine and Rheumatology, ‘‘Ninetto Melli’’ Hospital, San Pietro Vernotico, Brindisi, Italy 2Center for Integrated Research, Campus Bio-Medico University, Rome, Italy 3Drug Sciences, Campus Bio-Medico University, Rome, Italy 4Clinical Medicine, Campus Bio-Medico University, Rome, Italy

Name and address for correspondence Fabio Cacciapaglia, MD, Department of Internal Medicine and Rheumatology, ‘‘Ninetto Melli’’ Hospital, San Pietro Vernotico, Brindisi, Italy. Tel/Fax: +39 0831 670377; e-mail: [email protected]

Introduction Over the past decade, anti-tumour necrosis factor alpha (TNFa) agents have shown activity in many rheumatic diseases, changing patients’ quality of life and paving the way to disease remission [1]. Main contraindications to anti-TNFa agents include active infections, latent tuberculosis infection (LTBI), current or recent malignancy, and moderate-to-severe congestive heart failure (HF). Screening for risk factors, patient-tailored prophylactic therapies, and careful clinical follow-up are highly recommended to avoid or minimize untoward effects of antiTNFa agents. With regard to the risk of treatment-related cardiac events, it is worth recalling that rheumatoid arthritis (RA) patients are inherently exposed to increased morbidity and mortality for myocardial infarction (MI) and HF compared

S119

with the general population [2]. Data from national registries seem to demonstrate the safety of anti-TNFa agents in these patients, and a measurable reduction in the incidence of cardiovascular events has been documented in those patients who respond to TNFa inhibitor treatment [3]. Nevertheless, a contraindication to using anti-TNFa agents in RA patients with New York Heart Association (NYHA) class III–IV HF has been maintained. Contraindication to anti-TNFa agents in RA patients with HF comes from two randomized clinical trials of non-RA patients who developed HF with high serum levels of TNFa [4]. The rationale for testing anti-TNFa agents in significant heart failure rests with the assumption that high circulating levels of TNFa and HF worsening were linked by a causeand-effect relationship [5]. Surprisingly, however, the two studies failed to demonstrate an effect of anti-TNFa agents on HF deterioration; the dose-related increase in hospitalization or fatal events could in fact be demonstrated in patients with NYHA class III–IV HF. On the basis of these negative outcomes, infliximab and etanercept were labelled as contraindicated in non-RA patients with HF, while a cautionary label against using any anti-TNFa agent in rheumatological, dermatological and gastroenterological patients with HF was also issued. Caution was extended to any new anti-TNFa drug that was granted approval for clinical use. In clinical practice, physicians are faced with the problem that anti-TNFa agents may cause both detrimental and beneficial effects on the heart. In RA patients, anti-TNFa agents seem to delay disease-associated deterioration of cardiac function toward HF or MI; however, RA patients with pre-existing HF may show cardiac sequelae if they are given anti-TNFa drugs. Several hypotheses have been put forward to explain this Janus-like behaviour of TNFa [6]: direct cytotoxicity with complement fixation on cardiomyocytes, TNFa rebound elevations that result in a prolonged circulation of biologically active TNFa, and dysregulation of the cellular receptor machine that relays life- or death-oriented messages from TNFa. There are in fact two TNFa receptors: TNFa receptor 1 (TNF-R1, also referred as p55) and TNFa receptor 2 (TNF-R2 or p75). TNF-R1 seems to mediate activation or increased expression of proapoptotic proteins, while TNF-R2 seems to activate the transcription of genes related to cellular survival and proliferation. One more distinction between the two receptors is posed by their different affinity for TNFa: whereas TNF-R1 shows low affinity, TNF-R2 shows high affinity and may ‘‘pass’’ TNFa to TNF-R1 only upon saturation [7].

The aforesaid differences between TNF-R1 and TNFR2 anticipate that low-level TNFa would convey salutary effects to the heart by binding preferentially to TNF-R2, while high levels of TNFa would convey negative effects by binding to TNF-R1. Priming the heart with low-level TNFa would protect the heart against subsequent exposure to high-level TNFa if this hypothesis is proved correct.

Methods In agreement with this premise, the objective of our study was to demonstrate that low-dose TNFa cause cardiac preconditioning that protects against re-challenging with high-dose TNFa. For this, we used H9c2 cardiomyocytes, a cell line that is widely used to probe cardiotoxicity from noncardiovascular drugs [8]. H9c2 cells were grown to subconfluence and cell viability was assessed by the mitochondrion-specific assay based on antimycin A suppression of 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium-bromide reduction. After exploratory experiments of 24-h exposure of cardiomyocytes to 0.01–1000 ng/mL of TNFa, we identified a 50% inhibitory concentration (IC50), of 0.06 ± 0.03 ng/mL (n = 10), and a nontoxic concentration of 0.01 ng/mL that could be used to precondition cardiomyocytes before they were rechallenged with high-dose TNFa. Preconditioned H9c2 cardiomyocytes were rechallenged with TNFa, or with other cytotoxic substances: hydrogen peroxide (H2O2) as oxidative stress, or etoposide (ETO) as apoptotic agent. To adjust the survival effect of TNFa preconditioning to the different concentrations of the stressors, the net gain in viability was plotted against the ratio between the stressor’s concentrations at which survival occurred, and the IC50 value of that stressor in unpreconditioned cardiomyocytes. Results We observed that low-dose TNFa protected cardiomyocytes against re-exposure to high-dose TNFa, increasing the IC50 some 60-fold (from 0.06 ± 0.03 ng/mL to 3.6 ± 2.2 ng/mL, n = 5; p \ 0.01) and causing survival effects that propagated over a broad range of otherwise toxic TNFa concentrations (Fig. 1). Quantikine assays of soluble TNF-R1 and TNF-R2 in cell supernatants showed that cardiomyocytes exposed to

S120

beneficial effects. The untoward cardiac effects of antiTNFa drugs would be accommodated by such a dual behaviour of TNFa. In HF patients, detrimental cardiac effects of anti-TNFa drugs might be caused by abrogation of cardiomyocyte preconditioning that is driven by low levels of TNFa and protects against high TNFa and related death messages. Acknowledgments This work was supported by a research grant from Schering-Plough for laboratory costs. Native English editing was provided by Mary Hines of inScience Communications, Springer Healthcare. This editorial support was funded by Pfizer.

Fig. 1 Effects of tumour necrosis factor alpha (TNFa)-preconditioning on H9c2 cardiomyocyte survival against tumour necrosis factor alpha (TNFa), hydrogen peroxide (H2O2) and etoposide (ETO). IC50 = half maximal (50%) inhibitory concentration. The abscissor (x-axis coordinate) shows the ratio obtained from the concentrations of the stressor (TNFa, H2O2 or ETO) in preconditioned experiments divided by the mean IC50 of each stressor in unpreconditioned experiments; the ordinate (y-axis coordinate) shows the survival effect induced by preconditioning, expressed as the percentage of cells still alive versus unpreconditioned cells. Ratios above unity demonstrate increased cardiomyocyte survival at concentrations higher than unpreconditioned IC50 (see also text)

low TNFa and then rechallenged with TNFa at its IC50, released a two-fold excess of TNF-R2 compared with TNFR1. This was consistent with low TNFa activating TNF-R2 signals. We exposed unpreconditioned or preconditioned cardiomyocytes to a H2O2 bolus (0.01–500 lM) or to a ROS-independent proapoptotic agent ETO (0.01–1000 lM). In either case we observed that TNFa preconditioning made cardiomyocytes more resistant to H2O2 or ETO, increasing the IC50 values, respectively, from 90 ± 35 to 205 ± 22 lM or from 1 ± 0.35 to 191.7 ± 2.8 lM (n = 5, p B 0.01).

Discussion TNFa exerts its effects by increasing cellular levels of reactive oxygen species (ROS) and oxidative stress, or by inducing both ROS-dependent or -independent apoptosis [9]. In comparison with survival effects induced by low TNFa against high TNFa, survival effects were more evident when low TNFa protected against ETO but were less evident when low TNFa protected against H2O2 (see also Fig. 1). This suggests that in cardiomyocytes, TNFa toxicity was caused primarily by ROS-independent apoptosis. Our results call for revisiting the axiom ‘‘TNFa therapy is synonymous with cardiac dysfunction’’: one should appreciate that TNFa may convey both detrimental and

References 1. Rosenblum H, Amital H. Anti-TNF therapy: safety aspects of taking the risk. Autoimmun Rev. 2011;10:563–8. 2. Symmons DP, Gabriel SE. Epidemiology of CVD in rheumatic disease, with a focus on RA and SLE. Nat Rev Rheumatol. 2011;7:399–408. 3. Cacciapaglia F, Navarini L, Menna P, et al. Cardiovascular safety of anti-TNF-alpha therapies: facts and unsettled issues. Autoimmun Rev. 2011;10:631–5. 4. Coletta AP, Clark AL, Banarjee P, Cleland JGF. Clinical trials update: RENEWAL (RENAISSANCE and RECOVER) and ATTACH. Eur J Heart Fail. 2002;4:559–61. 5. Mann DL. Inflammatory mediators and the failing heart: past, present, and the foreseeable future. Circ Res. 2002;91:988–98. 6. Sarzi-Puttini P, Atzeni F, Shoenfeld Y, et al. TNF-alpha, rheumatoid arthritis, and heart failure: a rheumatological dilemma. Autoimmun Rev. 2005;4(3):153–61. 7. Flaherty MP, Guo Y, Tiwari S, et al. The role of TNF-alpha receptors p55 and p75 in acute myocardial ischemia/reperfusion injury and late preconditioning. J Mol Cell Cardiol. 2008;45: 735–41. 8. Menna P, Salvatorelli E, Minotti G. Cardiotoxicity of antitumor drugs. Chem Res Toxicol. 2008;21:978–89. 9. Cacciapaglia F, Menna P, Navarini L, et al. Matters of the heart: the case of TNFalpha-targeting drugs. Mol Interv. 2011;11:79–87.

Rapid Remission in Response to Etanercept Therapy Results in Better Outcome in Patients with Rheumatoid Arthritis Bernd Raffeiner,1,3 Costantino Botsios,2 Francesca Ometto,2 Livio Bernardi,2 Cristiana Vezzari,2 Silvano Todesco,2 Paolo Sfriso,2 Leonardo Punzi2 1 Rheumatology Unit, Department of Medicine, University of Padova, Padova, Italy 2Rheumatology Unit, Clinical and Experimental Medicine, University of Padova, Padova, Italy 3Rheumatology, Internal Medicine, General Hospital of Bolzano, Bolzano, Italy

Name and address for correspondence: Bernd Raffeiner, Rheumatology Unit, Department of Medicine, University of Padova, via Giustiniani 2, 35128 Padova, Italy. Tel: +39 0498212190; Fax: +39 0498212191; e-mail: [email protected]

S121

Introduction Functional outcome of patients with rheumatoid arthritis (RA) depends on the rapid achievement of disease remission to prevent radiological damage. The use of biological agents early in the course of RA has been found to be more effective than their use in established RA [1]. One of the characteristics of tumour necrosis factor alpha (TNFa) blockers is the speed of clinical response, with responses seen in the first 2 weeks of treatment; this has been demonstrated with adalimumab in the PREMIER study, with infliximab in the ASPIRE study, with etanercept in the ERA study, and recently with certolizumab pegol [2–4]. Inhibition of radiological progression can be observed as early as 16 weeks after starting anti-TNFa therapy [5]. Early responders (those with responses at 6 weeks after initiation of therapy) had a higher probability of achieving improved quality of life and significantly better symptom control than patients who were responders at week 12 [6]. The rapid induction of remission is associated with better long-term improvements than delayed response [6]. The objective of this study was to establish the proportions of patients achieving clinical remission within the first month (fast remission responders) and after the first month (slow remission responders) after commencing treatment with etanercept, and to determine the effects of these two groups on the maintenance of clinical and radiological remission.

Methods A retrospective case control study was performed at Rheumatology Unit, University of Padova, on patients with RA who started etanercept therapy (25 mg twice a week subcutaneously) from 2004 to 2010 because of moderate-tosevere disease activity despite treatment with disease-modifying anti-rheumatic drugs (DMARDs). Approval of ethical committee and informed patient consent were obtained. All patients having available rheumatologic control data at the first and third month after the start of etanercept treatment were included. Patients with disease duration of B4 years were classed as having early RA, those with longer disease durations were considered to have established RA. Patients achieving DAS28 remission (DAS28\2.6) by the end of the first month were classified as fast remission responders and patients reaching DAS28 remission at three months were classed as slow remission responders. Patients who did not reach remission within 3 months or who did not maintain remission for at least 6 months were considered unresponsive and excluded from the study. Fast remission responders were compared with slow remission responders regarding the maintenance of clinical

remission on etanercept in long-term follow-up. The effect of treatment on radiological progression was determined by van der Heijde-modified Total Sharp Score (TSS) calculated for X-rays performed at baseline and after 1 year. Characteristics including age, gender, disease duration, levels of rheumatoid factor and anti-cyclic citrullinated peptide (anti-CCP) antibodies, DAS28 score, erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) at baseline and the use of previous and concomitant treatments were compared between the two remission groups. Statistical analysis was performed by Student T-test and Pearson test as appropriate. Parts of data have been previously presented [7].

Results The study considered a total of 176 patients who reached remission at one of the two time points. 37 patients were excluded because not reaching remission at any of the time points. Sixty-eight patients (38.6%) reached DAS28 remission within the first month of treatment with etanercept and 108 (61.4%) reached remission between 1 month and 3 month after treatment commenced. Baseline characteristics and therapies were not different between both groups upon statistical analysis. Seven of 68 fast remission responders (10.3%) lost disease control during the follow-up period (mean time to loss of remission 3.5 years), compared with 23.1% of slow remission responders (25 out of 108; p \ 0.05). When patients with early RA (26.1% of all patients included) were considered separately, the difference in relapse rates between the two groups was even more significant (16.6% vs. 64.7%; p \ 0.05). Radiological progression (defined as a change [D] in TSS of [1) occurred in 6.1% of fast remission responders, compared with 13.3% of slow remission responders (p \ 0.05).

Discussion In patients with early RA, American College of Rheumatology (ACR) core set measures after 12 weeks of DMARD treatment predicted articular destruction 2 years later, demonstrating the importance of obtaining rapid control of disease activity in order to prevent long-term irreversible damage [8]. Studies have shown that a response to treatment by week 12 increases the possibility of low disease activity (LDA) or remission after 1 year of anti-TNFa therapy in patients with RA [9]. It was demonstrated that there are long-term clinical and radiographical differences between patients who show response by week 12 and those who do not. Similarly, it was demonstrated that attainment of remission or LDA at 3 months led to no or minimal progression of joint damage over a

S122

1-year period in patients treated with methotrexate and infliximab [10]. The effects of very rapid response to antiTNFa treatment have been addressed recently. Response to treatment as early as week 6 predicts continuation of drug in the long term [11]. Furthermore, patients treated with certolizumab who had a clinical response at week 6 demonstrated significantly greater ACR20, ACR50 and ACR70 responses, significantly lower DAS28, and higher LDA and remission, and improved patient-reported outcomes (pain, physical function, and fatigue) at the end of 1 year of treatment relative to patients who had a response at week 12. The greater improvement in physical function in the week 6 responders is of particular significance as the Health Assessment Questionnaire Disability Index (HAQ-DI) has been demonstrated to be a strong predictor of disability and mortality [12]. Our study clearly shows that fast remission response by the end of the first month after treatment initiation does occur. Fast remission response determines better outcome in early and established disease by greater maintenance of clinical and radiological remission compared with slow remission response [6]. In patients who show a slower or an incomplete response, careful monitoring and inspection of the rate of improvement during the first 12 weeks of therapy may help to predict the potential benefit of continuing treatment. Furthermore, the ability to predict long-term outcomes based on the onset of initial clinical response has the potential to reduce costs, decrease unnecessary drug exposure, and allow prompt access to alternative therapy.

4. Bathon JA, Martin RW, Fleischmann RM, et al. A comparison of etanercept and methotrexate in patients with early rheumatoid arthritis. N Engl J Med. 2000;343:1586–93. 5. Smolen J, Landewe´ RB, Mease P, et al. Efficacy and safety of certolizumab pegol plus methotrexate in active rheumatoid arthritis: the RAPID 2 study. A randomised controlled trial. Ann Rheum Dis. 2009;68:797–804. 6. Keystone EC, Curtis J, Fleischmann R, et al. A more rapid clinical response following certolizumab pegol treatment is associated with better 52-week outcomes in patients with rheumatoid arthritis. Ann Rheum Dis. 2009;68(3):225. 7. Raffeiner B, Botsios C, Ometto F, et al. Fast remission response to etanercept predicts maintenance of clinical and radiological remission in rheumatoid arthritis. Ann Rheum Dis. 2011;70(Suppl 3):250. 8. Ichikawa Y, Saito T, Yamanaka H, et al. Clinical activity after 12 weeks of treatment with nonbiologics in early rheumatoid arthritis may predict articular destruction 2 years later. J Rheumatol. 2010;37:723–9. 9. Aletaha D, Funovits J, Keystone EC, et al. Disease activity early in the course of treatment predicts response to therapy after one year in rheumatoid arthritis patients. Arthritis Rheum. 2007;56: 3226–35. 10. Smolen JS, Han C, van der Heijde DM, et al. Radiographic changes in rheumatoid arthritis patients attaining different disease activity states with methotrexate monotherapy and infliximab plus methotrexate: the impacts of remission and tumour necrosis factor blockade. Ann Rheum Dis. 2009;68:823–7. 11. Gu¨lfe A, Kristensen LE, Geborek P, et al. Six and 12 weeks treatment response predicts continuation of tumor necrosis factor blockade in rheumatoid arthritis: an observational cohort study from southern Sweden. J Rheumatol. 2009;36:517–21. 12. Keystone EC, Curtis JR, Fleischmann RM, et al. Rapid improvement in the signs and symptoms of rheumatoid arthritis following certolizumab pegol treatment predicts better longterm outcomes: post-hoc analysis of a randomized controlled trial. J Rheumatol. 2011;38(6):990–6.

Conclusion Rapid achievement of remission with etanercept is associated with better maintenance of disease control in patients with RA, particularly in those with early RA. Acknowledgments Native English editing was provided by Raewyn Poole of inScience Communications, Springer Healthcare. This editorial support was funded by Pfizer.

References 1. Villaverde Garcia V, Balsa Criado A. Does early treatment of rheumatoid arthritis lead to a better long-term prognosis? Reumatol Clin. 2010;6(2):106–10. 2. Breedveld FC, Weisman MH, Kavanaugh AF, et al. The PREMIER study: a multicenter, randomized, double-blind clinical trial of combination therapy with adalimumab plus methotrexate 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. 3. St Clair EW, van der Heijde DM, Smolen JS, et al. Combination of infliximab and methotrexate therapy for early rheumatoid arthritis: a randomized, controlled trial. Arthritis Rheum. 2004; 50: 3432–43.

Discontinuation of Anti-TNFa Therapy Due to Remission in Rheumatoid Arthritis : A Retrospective Study Maria Sole Chimenti,1 Dario Graceffa,1 Gioia Di Muzio,1 Eleonora Ballanti,1 Maurizio Rinaldi,1 Roberto Perricone1 1 Rheumatology, Allergology and Clinical Immunology; Department of Internal Medicine, University of Rome Tor Vergata, Via Montpellier 1, 00133 Rome, Italy

Name and address for correspondence: Maria Sole Chimenti, Department of Internal Medicine, Unit of Rheumatology, University of Rome Tor Vergata, Via Montpellier 1, 00163 Rome, Italy Tel: +39 06 72596287; Fax: +39 06 20900358; e-mail: maria.sole.chimenti@ uniroma2.it

Introduction Rheumatoid arthritis (RA) is a chronic inflammatory disease that affects approximately 1% of the general population [1]. Increased TNFa levels have been shown to play a

S123

key role in the early stages of the inflammatory process, during which TNFa can stimulate T-cell activation and induce the expression of interleukin (IL)-2, interferon (IFN)c, inflammatory cytokines (IL-1 and IL-12) and chemokines (such as IL-8) [1, 2]. Recent evidence suggests that TNFa blockers may produce rapid and sustained suppression of inflammatory disease, preserve function and attenuate joint damage. In recent years, the TNFa inhibitors adalimumab, infliximab and etanercept have become widely used in the treatment of RA because of their favourable efficacy and safety profiles. However, clinical practice guidelines and consensus statements on patient selection criteria, treatment duration and interruption of anti-TNFa therapy are under constant revision [3, 4]. International Registry Results Results from clinical trials have demonstrated that adalimumab, infliximab and etanercept are equally effective in RA [2]. Population-based cohort studies have compared treatment retention rates between these agents and determined the specific causes of treatment discontinuation. The Swiss Clinical Quality Management for Rheumatoid Arthritis (SCQM-RA) included 2364 RA patients who were treated with etanercept, infliximab and adalimumab during the period 1997–2006 [5]. Treatment discontinuation was reported in 803 of the 2364 (34%) patients; lack of efficacy of the drug was the most frequent reason for discontinuation (55.8% of patients who discontinued treatment). Only a small proportion of patients (1%) suspended treatment for disease remission. The median treatment duration with anti-TNFa therapy was 37 months; however, discontinuation rates differed between the three anti-TNFa agents, with a shorter retention rate for infliximab. This might be explained by a higher risk of infusion reactions or acute systemic reactions with infliximab compared with etanercept and adalimumab. Data from the British Society for Rheumatology Biologics Register (BSRBR), which included 6739 patients with RA who were treated with etanercept, infliximab, adalimumab and had a mean followup duration of 15 months, confirmed that the major cause of treatment discontinuation is inefficacy of the treatment [6]. The Danish nationwide DANBIO registry collected data from 2326 RA patients, observed from 2000 to 2007 [7]. Rates of improvements, considered as evidence to gain satisfactory treatment responses (70% improvement according to the American College of Rheumatology criteria, European League Against rheumatism good response or remission expressed by Disease Activity Score in 28 joints \2.6), and causes of treatment discontinuation were similar to those observed in the other registries. However, a clinically relevant difference emerged in the treatment

response rates was seen, with response rates being lowest for infliximab, intermediate for etanercept and highest for adalimumab. Treatment maintenance rates, which may be a marker for drug efficacy, were lowest for infliximab, intermediate for adalimumab and highest for etanercept. Moreover, adalimumab had the highest rates of treatment response and disease remission (19% of patients receiving adalimumab, 17% of patients receiving etanercept and 11% of those receiving infliximab achieved an ACR70 response, and 41%, 34% and 27% respectively, achieved a EULAR response after 6 months), and etanercept had the longest drug survival rates. In particular, at 48 months, the drug adherence rates were 56% for etanercept, 52% for adalimumab and 41% for infliximab. In general, when all reasons for treatment discontinuation were considered in a national register, baseline comorbidities and the use of infliximab were associated with higher discontinuation rates [8]. Data regarding response rates from the BSRBR considered older age, low functional status, and concomitant use of prednisone as negative predictors of clinical response and remission [9]. Predictors of drug discontinuation included female gender, patients who were smokers, patients having higher baseline health assessment questionnaire (HAQ) scores, an associated baseline co-morbidity and the use of infliximab rather than etanercept [10]. Good drug survival rates were observed in patients receiving concomitant synthetic disease-modifying antirheumatic drugs (DMARDs) and in those with higher baseline disease activity [8]. Discontinuation of Anti-TNFa After Disease Remission Numerous studies suggest that withdrawal of anti-TNFa therapy is almost invariably followed by disease recurrence. This has been demonstrated in the ATTRACT (AntiTumour Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy) study, in which long-standing RA patients were treated for 24 months with infliximab plus methotrexate and subsequently with methotrexate alone [11]. Disease relapse was observed after a mean period of 14 weeks, after which infliximab was restarted and ACR (American College of Rheumatology) responses comparable to those prior to treatment withdrawal were achieved. Patients with established RA who respond to treatment (ACR 20 response) require ongoing therapy to maintain their response in the long term. Notably, the drug can be restarted after an interval of several months without observed problems. To our knowledge, there are no reported data about etanercept discontinuation after RA disease remission. The objective of our study was therefore to assess the maintenance of efficacy and characteristics of relapse in patients with RA following discontinuation of etanercept.

S124

Methods We designed a retrospective study to assess maintenance of efficacy and relapse characteristics (timing and severity) of clinical and serological parameters of remission in patients with RA who had moderate-to-severe disease activity and were treated with a long-term etanercept regimen as monotherapy. Clinical notes of patients referring to our out-patient department of Rheumatology of the University of Rome ‘‘Tor Vergata’’ from January 2006 until February 2010 were analysed to select subjects affected by RA treated with etanercept, administered at a dose of 50 mg subcutaneously once weekly or 25 mg twice weekly, who obtained a clinical remission. A total of 37 unrelated consecutive patients (26 females, aged between 18 and 64 years) affected by RA (mean disease duration of 11.31 ± 8.33 years) were studied. Informed consent was obtained from all patients before starting the treatment and good clinical practice guidelines were followed. Patients fulfilling remission criteria, based on DAS28 remission (DAS28 \2.6), Tender joint count \1, Swollen joint count \1, C-reactive protein (CRP) B1 mg/ dL and Patient Global Assessment B1 (on a 0–10 scale), continuously for 36 weeks underwent treatment discontinuation. Clinical relapse was defined by the presence of one of the following criteria: disease activity score (DAS) 28-erythrocyte sedimentation rate (DAS28-ESR) C2.6 associated with elevated CRP, or tender or swollen joint count C1. Data from the laboratory analyses were collected in a Windows-based database (Microsoft Excel 2007) and analysed with GraphPad Prism 5 statistical software (GraphPad Software, San Diego, CA). Data were expressed as mean ± SEM. The significance of differences in mean values obtained at T0, T12 and T24 weeks of treatment was assessed with Student’s t-test (statistical significance was set at p B 0.05).

Table 1 Demographic and clinical characteristics at baseline Patients (n)

37

Gender (M/F)

11/26

Age (mean ± SD)

51 ± 11.8

Age of onset (mean ± SD)

28.9 ± 12.6

Mean disease duration (mean ± SEM)

11.3 ± 8.3

Mean etanercept treatment period (mean ± SEM)

70.0 ± 44.9

DAS28-ESR (mean ± SEM)

6.18 ± 0.9

ESR (mean ± SEM)

33.9 ± 19.0

CRP (mean ± SEM)

5.9 ± 1.6

Previous methotrexate treatment

36/37 (97%)

Previous leflunomide treatment

21/37 (57%)

Previous sulphasalazine treatment

19/27 (51%)

CRP C-reactive protein, DAS disease activity score, ESR erythrocyte sedimentation rate, SD standard deviation, SEM standard error of the mean

The most frequently reported adverse event, during the treatment, was weight gain (minimum 3 kg, maximum 5 kg) observed in four patients, while in five cases an injection site reaction was reported. In two patients, a urinary tract infection was diagnosed and resolved with an antimicrobial oral treatment, without requiring etanercept discontinuation. After a period of treatment, each patient achieving satisfactory disease control and fulfilling the previously defined criteria of remission were recommended to discontinue the treatment. After treatment withdrawal, an accurate follow up of patients was performed in order to investigate the length of remission and the possible recurrence of the disease. The mean DAS28-ESR reduced from 6.18 ± 0.93 to 2.15 ± 0.10 from baseline to treatment interruption, mean ESR reduced from 33.9 ± 19.0 to 19.0 ± 8.9 and mean CRP 5.88 ± 1.6 to 1.0 ± 0.4. Mean time to disease relapse after the discontinuation of etanercept was 16.5 ± 1.6 weeks. At relapse, the mean DAS28-ESR value increased to 4.8 ± 0.1, the mean ESR increased to 32.7 ± 20.1 and the mean CRP increased to 6.8 ± 2.5.

Results A total of 37 patients, affected by RA treated with etanercept satisfied the analysis inclusion criteria (Table 1). In this study group, after a mean treatment period of 67.0 ± 44.9 weeks (range: 45–220 weeks) at a dose of 50 mg subcutaneously once weekly or alternatively 25 mg twice weekly, a good disease control corresponding to a clinical remission was achieved. Etanercept efficacy was consistent as expressed by the improvement of disease activity parameters, and the safety profile was excellent.

Discussion Our results confirmed that the efficacy of etanercept was consistent with respect to clinical and laboratory disease parameters in patients with RA, and its safety profile was good. Although a gradual worsening of disease was observed in all patients who achieved remission, the length of the disease-free period was consistent, lasting a mean of 16.5 weeks after treatment discontinuation.

S125

Notably, a disease-free period of 16.5 weeks after treatment discontinuation results in a consistent medication sparing and reduction of side effects risks, since patients during etanercept withdrawal didn’t use other medications such as prednisone or NSAIDs. Moreover, in clinical practice this finding could be useful in several circumstances, for example in patients with health-related scheduled events who need treatment interruption (e.g. pregnancy planning, surgery, or live attenuated virus vaccination). To our knowledge, this is the first study evaluating the outcome of windrowing etanercept after disease remission in patients with RA. An important consideration arising from these results is that the definition of remission should be limited to the treatment period, given that treatment interruption is invariably followed by a disease relapse. We believe that until specific clinical-laboratory tools to define clinical and radiological remission are available, a clinician should consider personalizing the treatment scheme to individual patient characteristics. This study is small and consider only etanercept treatment but adds value as most studies do not stop TNF inhibitor medications when remission has occurred in patients with RA.

Conclusion These data suggest that a mean period of 16.5 weeks, without treatment, can occur without a rebound of signs and symptoms. New assessments to define remission should be developed and incorporated into clinical trials and longitudinal registries. Further studies are also needed to validate intermittent therapeutic strategies in early RA and the clinical behaviour of disease after the re-introduction of anti-TNFa treatment. Acknowledgments Native English editing was provided by Raewyn Poole of inScience Communications, Springer Healthcare. This editorial support was funded by Pfizer. The authors declare no conflicts of interest. Authors’ contributions MSC designed the study, supervised the project and wrote the manuscript. DG and GDM performed the statistical analysis. EB and MR contributed in selecting clinical and laboratory values of the patients. RP supervised the study and helped to write the manuscript.

References 1. Ro¨nnelid J, Wick MC, Lampa J, et al. Longitudinal analysis of citrullinated protein/peptide antibodies (anti-CP) during 5 year follow up in early rheumatoid arthritis: anti-CP status predicts worse disease activity and greater radiological progression. Ann Rheum Dis. 2005;64:1744–9. 2. Weinblatt ME, van Riel PL. Targeted therapies: summary clinical trials working group. Ann Rheum Dis. 2006;65:iii89.

3. Punzi L, Podswiadek M, Sfriso P, et al. Pathogenetic and clinical rationale for TNF-blocking therapy in psoriatic arthritis. Autoimmun Rev. 2007;6:524–8. 4. Van Aken J, Lard LR, le Cessie S, et al. Radiological outcome after four years of early versus delayed treatment strategy in patients with recent onset rheumatoid arthritis. Ann Rheum Dis. 2004;63:274–9. 5. Du Pan SM, Dehler S, Ciurea A, et al. Comparison of drug retention rates and causes of drug discontinuation between antitumor necrosis factor agents in rheumatoid arthritis. Arthritis Rheum. 2009;15:560–8. 6. Hyrich KL, Watson KD, Silman AJ, Symmons DP. Predictors of response to anti-TNF-alpha therapy among patients with rheumatoid arthritis: results from the British Society for Rheumatology Biologics Register. Rheumatology (Oxford). 2006;45:1558–65. 7. Hetland ML, Christensen IJ, Tarp U, et al. Direct comparison of treatment responses, remission rates, and drug adherence in patients with rheumatoid arthritis treated with adalimumab, etanercept, or infliximab: results from eight years of surveillance of clinical practice in the nationwide Danish DANBIO registry. Arthritis Rheum. 2010;62:22–32. 8. Hyrich KL, Watson KD, Silman AJ. British Society for Rheumatology Biologics Register. Predictors of response to anti-TNF-alpha therapy among patients with rheumatoid arthritis: results from the British Society for Rheumatology Biologics Register. Rheumatology (Oxford). 2006;45:1558–65. 9. Hyrich KL, Lunt M, Watson KD, British Society for Rheumatology Biologics Register, et al. Outcomes after switching from one anti-tumor necrosis factor agent to a second anti-tumor necrosis factor agent in patients with rheumatoid arthritis. Arthritis Rheum. 2007;56:13–20. 10. Hyrich KL, Watson KD, Lunt M, Symmons DP. Changes in disease characteristics and response rates among patients in the United Kingdom starting anti-tumour necrosis factor therapy for rheumatoid arthritis between 2001 and 2008. Rheumatology (Oxford). 2011;50:117–23. 11. Maini R, St Clair EW, Breedveld F, et al. Infliximab (chimeric anti-tumour necrosis factor alpha monoclonal antibody versus placebo in rheumatoid arthritis patients receiving concomitant methotrexate: a randomised phase III trial. ATTRACT Study Group. Lancet. 1999;354:1932–9.

Long-Term Safety of Etanercept in Elderly Patients with Rheumatoid Arthritis in Clinical Practice Alfredomaria Lurati,1 Luca Bertani,1 Katia Angela Re,1 Mariagrazia Marrazza,1 Daniela Bompane,1 Magda Scarpellini1 1

Rheumatology Unit, Fornaroli Hospital Magenta Italy

Name and address for correspondence: Dr Alfredomaria Lurati, Rheumatology Unit, Fornaroli Hospital, Via Donatore Sangue 50, Magenta, 20013, Italy. Tel: +39 02 9796 3833 Fax: +39 02 9796 3409; e-mail: [email protected]

Introduction Despite the high prevalence of rheumatoid arthritis (RA) in the elderly, and the likelihood that older

S126

patients will have adverse reactions to medications as a result of age-related changes in drug metabolism and the presence of comorbidities requiring concomitant medications, patients older than 65 years tend to be inadequately represented in RA clinical trials. Moreover, the tolerability of, and response to therapy in these patients as a group have not been studied in detail. In one retrospective analysis of disease-modifying anti-rheumatic drugs (DMARDs) including gold, Dpenicillamine, azathioprine, and methotrexate (MTX), a substantially higher withdrawal rate due to toxicity was observed in patients older than 65 years compared to those younger than 65 years [1, 2]. The development of biological agents during the last decade, in particular tumour necrosis factor (TNF) inhibitors represents a major breakthrough in the treatment of RA [3–5]. There may be unmet need with less treatment in the elderly due to perceived risk of side effects when using DMARDs and biologics. The objective of this study was to assess whether the use of etanercept in ‘real life’ clinical practice is associated with a different rate of adverse events, primarily infections and serious side effects such as death, malignancies, demyelinating diseases and cardiac failure, in elderly versus younger patients with RA, and whether the drug survival curves were different in the two groups.

Methods Patients with RA, diagnosed according to American College of Rheumatology (ACR) criteria [6], who started treatment with etanercept between November 2005 and November 2006 referring to our rheumatology unit were included in this study and prospectively followed for 5 years. Informed consent was obtained from every patient enrolled. The etanercept dose was 50 mg subcutaneously once a week. Patients were allowed to continue treatment with non-steroidal anti-inflammatory drugs (NSAIDs) and corticosteroids if needed, and MTX or other previous DMARDs if tolerated and deemed necessary. Safety profile endpoints included incidence rates of all side effects (SEs) defined as: adverse events (AE, with a temporary therapy suspension), serious AE (SAE, defined as an AE that required the permanent discontinuation of the therapy), and infectious adverse events (IAE, with temporary therapy suspension or discontinuation of treatment associated with or without intravenous antibiotic treatment). All patients were evaluated every 2 months. All data collected during the study were recorded in electronic form for subsequent analysis.

Statistical Methods Pairwise comparisons were based on the Wilcoxon matched pairs signed-rank test. The safety profile in the two groups was estimated by survival analysis, taking any AE as an event of interest using the Kaplan-Meier method. Comparisons between the resultant curves were made by log-rank test and correlations performed with the Mantel Cox equation. All values of p \ 0.05 were considered to indicate statistical significance (two-tailed test).

Results Eighty-two patients were enrolled in this study from November 2005 to November 2006 and followed until November 2011: 31 patients (28 females) aged B65 years and 52 patients (40 females) aged [65 years. In patients aged [65 years the safety profile of etanercept was similar to that observed in patients aged B65 years (p [ 0.05), and the survival curves were also similar (p [ 0.05). Rates of infection were comparable across the groups (p [ 0.05) (Table 1). Urinary tract infections and upper respiratory tract infections (including nasopharyngitis) were the most frequently reported IAEs. Rates of SAEs were also comparable in the two groups (p [ 0.05) (Table 2). The mean survival time before the occurrence of a first treatment-related AE with etanercept was 29.6 ± 2.42 months in Group B65 years, compared with 31.3 ± 2.6 months in Group [65 years (P = 0.67). The median survival time was 31.7 months in Group B65 years and 34.4 months in Group [65 years (p = 0.33). No significant correlation was found between the survival time until the first adverse event and age (p = 0.63). The mean survival time before discontinuation of treatment with etanercept due to SAEs was 46.2 ± 2.99 months in Group B65 years and 47.7 ± 4.1 in Group [65 years (p = 0.27). No significant correlation has been found between therapy duration when SAE occurred and the age used as a covariate in the Mantel Cox equation (p = 0.728). Finally, 5-year retention rates were 64.7% in Group B65 years and 63.0 % in Group [65 years (p = 0.34) (Fig. 1).

Discussion Epidemiological studies have shown that RA is most prevalent in those 65 years of age or older [7]. The disease now has a mean age of onset of 55 years in ERA cohorts so

S127 Table 1 Infections occurring during treatment with etanercept in patients aged B65 years and those aged [65 years Total events observed

p value

UTI

31

47

78

0.23

Upper RTI (including influenza)

18

22

40

0.13

COPD

3

2

5

0.11

Discitis

2

1

3

0.57

Mycotic infection

1

1

2

0.54

HBV/HCV reactivation

0

0

0

0.44

TB reactivation

0

0

0

0.32

Herpes virus infection

3

0

0

0.15

58

73

131

p = 0.55

Total IAE

Cumulative survival rate

Age Age B65 years [65 years (n = 31) (n = 51)

a

COPD chronic obstructive pulmonary disease, HBV hepatitis B virus, HCV hepatitis C virus, IAE infectious adverse event, NS not significant, RTI respiratory tract infection, SAE serious adverse event, TB tuberculosis, UTI urinary tract infection

Months of therapy before first AE

b Table 2 Serious adverse events occurring during treatment with etanercept in patients aged B65 years and those aged [65 years Total events observed

p-value

Malignancies

0

1 (NHL)

1

0.17

Squamocellular carcinoma

1

1

2

0.15

MGUS

2

1

3

0.09

Cardiac failure

0

1

1

0.43 0.33

Lower RTI

2

3

5

UTI [3/year

8

5

13

0.23

Neurological disorders including MS

0

0

0

0.15

Myocardial infarction Total SAEs

0

1

1

0.09

13

13

26

0.12

MGUS monoclonal gammopathy of unknown significance, MS multiple sclerosis, NHL non-Hodgkin’s lymphoma, NS not significant, RTI respiratory tract infection, SAE serious adverse event, UTI urinary tract infection

the age is increasing and, 1/3 of patients may be diagnosed after the age of 60 years. In general, older patients are more likely to have concomitant illnesses, including malignancies and cardiovascular disease [8]. Medication compliance, drug interactions, and decreased tolerability complicate the management of older patients [9]. The older RA patient is likely to require multiple medications and have difficulties with drug compliance, and may be prone

Cumulative survival rate

Age Age B65 years [65 years (n = 31) (n = 51)

Months of therapy before discontinuation due to SAE

Fig. 1 Cumulative survival rates in the two groups: months of etanercept therapy (a) before occurrence of the first AE and b before discontinuation of treatment due to SAE. Group A = age B65 years; Group B = age [65 years. AE adverse event, SAE serious adverse event

to commit dosage errors [10, 11]. Moreover RA is associated with increased mortality and comorbidity from a number of causes compared with the general population, including infection [12]. Data on the safety of anti-rheumatic drugs in elderly patients are somewhat lacking, particularly with respect to biological agents [13–15]. This

S128

study demonstrates that etanercept has a good safety profile in older patients with RA in the ‘real life’ clinical practice setting, with no observed increases in the rates of infections, reactivation of tuberculosis (and other opportunistic infections), malignancies, heart failure or demyelinating diseases.

14. Fleischmann R, Baumgartner SW, Weisman MH, et al. Long term safety of etanercept in elderly subjects with rheumatic diseases. Ann Rheum Dis. 2006;65(3):379–84. 15. Bathon JM, Fleischmann RM, Van der Heijde D, White B, et al. Safety and efficacy of etanercept treatment in elderly subjects with rheumatoid arthritis. J Rheumatol. 2006;33(2):234–43.

Conclusion

Adopting Low-Dose Etanercept Strategy in the LongTerm Management of Rheumatoid Arthritis Patients

In patients with RA in the clinical practice setting, the safety profile and durability of etanercept in those aged [65 years was similar to that in younger patients.

Bernd Raffeiner,1,3 Costantino Botsios,2 Silvano Todesco,2 Livio Bernardi,2 Francesca Ometto,2 Cristiana Vezzari,2 Paolo Sfriso,2 Leonardo Punzi2

Acknowledgments Native English editing was provided by Raewyn Poole of inScience Communications, Springer Healthcare. This editorial support was funded by Pfizer.

1 Rheumatology Unit, Department of Medicine, University of Padova, Padova, Italy, 2Rheumatology Unit, Clinical and Experimental Medicine, University of Padova, Padova, Italy, 3Rheumatology, Internal Medicine, General Hospital of Bolzano, Bolzano, Italy

References 1. Dahl SL, Samuelson CO, Williams HJ, et al. Second-line antirheumatic drugs in the elderly with rheumatoid arthritis: a post hoc analysis of three controlled trials. Pharmacotherapy. 1990;10:79–84. 2. van Schaardenburg D, Breedveld FC. Elderly-onset rheumatoid arthritis. Semin Arthritis Rheum. 1994;23:367–78. 3. Moreland LW, Baumgartner SW, Schiff MH, et al. Treatment of rheumatoid arthritis with a recombinant human tumor necrosis factor receptor (p75)-Fc fusion protein. N Engl J Med. 1997;337:141–7. 4. Moreland LW, Schiff MH, Baumgartner SW, et al. Etanercept therapy in rheumatoid arthritis. A randomized, controlled trial. Ann Intern Med. 1999;130:478–86. 5. Bathon JM, Martin RW, Fleischmann RM, et al. A comparison of etanercept and methotrexate in patients with early rheumatoid arthritis. N Engl J Med. 2000;343:1586–93. 6. Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 1988;31:315–24. 7. US Senate Special Committee on Aging, the American Association of Retired Persons, the Federal Council on the Aging, and the US Administration on Aging. Aging America: Trends and projections. Washington, DC: US Government Publication; 1991. 8. Villa-Blanco JI, Calvo-Ale`n J. Elderly onset rheumatoid arthritis: differential diagnosis and choice of first-line and subsequent therapy. Drugs Aging. 2009;26(9):739–50. 9. Olivieri I, Palazzi C, Peruz G, Padula A. Management issues with elderly-onset rheumatoid arthritis: an update. Drugs Aging. 2005;22(10):809–22. 10. Filipowicz-sosnowska A, Rupinski R. Elderly onset rheumatoid arthritis. Pol Arch Med Wewn. 2008;118(Suppl):36–42. 11. Tutuncu Z, Reed G, Kremer J, Kavanaugh A. Do patients with older onset rheumatoid arthritis receive less aggressive treatment? Ann Rheum Dis. 2006;65(9):1226–9. 12. Deal CL, Meenan RF, Goldenberg DL, et al. The clinical features of elderly-onset rheumatoid arthritis. A comparison with younger-onset disease of similar duration. Arthritis Rheum. 1985;28(9): 987–94. 13. Lurati A, Marrazza M, Scarpellini M, et al. Safety of etanercept in elderly subjects with rheumatoid arthritis. Biologics. 2010;4:1–4.

Name and address for correspondence: Dr Bernd Raffeiner, Rheumatology Unit, Department of Medicine, University of Padova, via Giustiniani 2, 35128 Padova, Italy. Tel: ?39 0498212190; Fax: ?39 0498212191; e-mail: [email protected]

Introduction Rheumatoid arthritis (RA) leads to disability, work loss and large direct and indirect costs to society [1]. The introduction of biological agents has the changed history of the disease and improved patient outcomes. Biological agents have higher annual treatment costs than conventional treatments for RA; however some studies have proven their cost-effectiveness for certain patients [2]. Increasing the dose to achieve disease control is a common practice for infliximab more than for etanercept or adalimumab. More recently the focus has been to reduce the dose of a biologic in those in sustained clinical remission to cut treatment costs. We have been recently reported some dose reduction data [3, 4]. The tumour necrosis factor alpha (TNFa) inhibitor etanercept appears particularly suitable for dose reduction once patients have achieved remission. Etanercept has similar serum concentration-time curve profiles whether 25 mg or 50 mg is administered [5]. Furthermore, it neutralizes TNFa with more than 20-fold greater potency than other TNFa blockers at low concentrations of TNFa, as would be expected in disease remission [6]. The objectives of this study were to elucidate the clinical efficacy of low-dose etanercept in maintaining remission induced by standard dose; to compare the arrest of radiological progression and safety of low-dose and standard-dose etanercept; and to estimate the cost savings deriving from a low dose etanercept treatment strategy.

S129

Methods This was a prospective non-randomized cohort study performed at Rheumatology Unit, University of Padova, from January 2004 to December 2009. The study was conducted within Italian registry Antares/MonitorNet. Approval from local ethics committee was obtained. Inclusion criteria were patients with RA who had achieved stable DAS28 remission (DAS28 score \2.6) during treatment with standard-dose etanercept (25 mg twice a week). Informed consent was obtained from each patient involved. After 12 months of DAS28 remission, patients continued treatment with a low-dose (25 mg once a week) or standard-dose etanercept regimen according to their DAS28 activity prior to commencing treatment with etanercept. All patients with moderate disease activity before biologic therapy (DAS28 [3.5 and \5.1) continued low dose etanercept. Patients with severe disease activity (DAS28 C5.1) prior to treatment were consecutively chosen in 1:1 manner to reduce or continue standard dose. Patients were monitored every 3 months for maintenance of clinical remission (DAS28 \2.6). Patients who did not maintain remission following dose reduction returned to standard-dose etanercept once DAS28 remission was lost. A Kaplan Meier survival curve was calculated for patients receiving low-dose etanercept. A van der Heijde modified Total Sharp Score (TSS) was calculated for hand and feet X-rays at baseline and after one year for all three groups. Progression was reported as absolute progression (change [D] in TSS[0) and as real progression (DTSS C5) according to the smallest detectable difference (SDD). The proportion of those with no radiographic progression was compared between the three groups. The annual progression of erosions prior to treatment with a biological agent was estimated for each patient using X-rays preceding the start of biological agent therapy and was codified as DTSS/ year. The incidences of infections per 1,000 patient-years of low- and standard-dose etanercept were calculated using the Mid-P exact test modified by Miettinen 7]. Cost savings for the low-dose strategy were calculated based on Italian national drug pricing including tax of 2009 through the subtraction of missed doses from expected full dose price. All data were compared for statistical significance by Mann–Whitney or exact Fisher tests, as appropriate.

moderate disease activity; 54 patients with severe disease activity continued to receive standard-dose etanercept. There were no differences with respect to age, gender, disease duration, rheumatoid factor or anti-cyclic citrullinated peptide (CCP) antibody status, presence of vasculitis, concomitant and past therapy with disease-modifying antirheumatic drugs (DMARDs), steroids or nonsteroidal anti-inflammatory drugs (NSAIDs), comorbidities, Health Assessment Questionnaire (HAQ) and DTSS/year between patients with severe disease activity in the low- and standard-dose groups. Both severe groups were found to have more aggressive disease in X-rays taken prior to biological therapy (DTSS/year 12.1 ± 8.9) than the moderate group (8.4 ± 5.3; p \ 0.05) and were more likely to be treated with concomitant at least one concomitant DMARD (76.7% vs. 59.5%; p \ 0.05). At the time of this analysis, 89 patients (81.6%) had maintained DAS28 remission with low-dose etanercept for a mean of 2.59 (± 1.3) years. Patients who did not maintain remission after dose reduction regained remission after reinstatement of standard-dose etanercept therapy, except one patient who was switched to adalimumab. The majority of relapses occurred in the first year of low-dose therapy (60%), with 30% in the second year and 10% in the third year. Successful maintenance of remission was more frequent in younger patients taking less corticosteroids and NSAIDs, but was not different between the severe and moderate pre-treatment disease activity groups. Radiological progression occurred in only a minority of patients receiving low-dose etanercept: DTSS [0 in 13.7% and DTSS C5 in 0.9% compared to 13% and 1.8% in full dose patients for DTSS [0 and DTSS C5 respectively. The ability of etanercept to arrest radiological progression was not dissimilar in both disease activity groups receiving low-dose treatment, in patients receiving standard-dose treatment, and in patients who failed dose reduction. Low-dose etanercept recipients had fewer infections than the standard-dose group; the incidence ratio of severe infections was 3.7 per 1000 patient-years for the low-dose group and 9.4 per 1000 patient-years for the standard-dose group; p \ 0.001. No further differences were found for other common adverse events such as local injection reactions and constitutional symptoms. The low-dose etanercept strategy produced cost savings of €1,583,273 from January 2004 to December 2009 for all patients translating to a cost saving per patient of €6,315 per year in low dose etanercept.

Results The study included 163 patients, 110 with severe disease activity prior to biological therapy, and 53 with moderate disease activity. Low-dose etanercept was administered to 56 patients with severe disease activity and all 53 patients with

Discussion New strategies for long-term treatment of patients with RA are required, especially in light of evidence that high rates

S130

of relapses occur with treatment discontinuation. This is particularly intriguing because of the need of long-term treatment to be balanced with the associated treatmentrelated adverse effects and costs. Low-dose etanercept has been proven by our study to sustain clinical and radiological remission in most patients. Low-dose etanercept was also found to have a favourable safety profile, with a lower incidence of infections than seen with standard-dose therapy. Adopting a low-dose etanercept strategy produces large cost savings, particularly in established RA where therapy discontinuation is very difficult, but also in earlier disease. Results of a recent study combining clinical trial and practice data to explore different treatment scenarios in early RA indicated that, in a situation where a considerable proportion of patients achieve remission, dose reduction will increase the cost-effectiveness of etanercept treatment [8].

http://www.OpenEpi.com. Updated 19 Sep 2010. Accessed 2 Oct 2010. 8. Kobelt G, Lekander I, Lang A, et al. Cost-effectiveness of etanercept treatment in early active rheumatoid arthritis followed by dose adjustment. Int J Technol Assess Health Care. 2011;27(3):93–200.

Conclusion

Name and address for correspondence: Bernd Raffeiner, Rheumatology Unit, Department of Medicine, University of Padova, via Giustiniani 2, 35128 Padova, Italy. Tel: ?39 0498212190; Fax: ?39 0498212191; e-mail [email protected]

This observational study has demonstrated that low-dose treatment with etanercept maintains remission in patients with RA and is associated with a lower incidence of infections and lower costs than continuation of standarddose etanercept therapy. Acknowledgments Native English editing was provided by Raewyn Poole of inScience Communications, Springer Healthcare. This editorial support was funded by Pfizer.

References 1. Lundkvist J, Kasteng F, Kobelt G. The burden of rheumatoid arthritis and access to treatment: health burden and costs. Eur J Health Econ. 2008;8(2):49–60. 2. Doan QV, Chiou CF, Dubois RW. Review of eight pharmacoeconomic studies of the value of biologic DMARDs (adalimumab, etanercept, and infliximab) in the management of rheumatoid arthritis. J Manag Care Pharm. 2006;12:555–6. 3. Botsios C, Furlan A, Ostuni P, et al. Effects of low dose etanercept in maintaining DAS remission previously achieved with standard dose in patients with rheumatoid arthritis. Ann Rheum Dis. 2007;66(2):54. 4. Raffeiner B, Botsios C, Sfriso P, et al. Efficacy of low dose etanercept in maintaining clinical and radiological remission in rheumatoid arthritis. Ann Rheum Dis. 2010; 69(Suppl 3):102. 5. Kawai S, Sekino H, Yamashita N, et al. The comparability of etanercept pharmacokinetics in healthy Japanese and American subjects. J Clin Pharmacol. 2006;46(4):418–23. 6. Tracey D, Klareskog L, Sasso EH, et al. Tumour necrosis factor antagonist mechanisms of action: a comprehensive review. Pharmacol Ther. 2008;117(2):244–79. 7. Dean AG, Sullivan KM, Soe MM. OpenEpi: Open Source Epidemiologic Statistics for Public Health, Version 2.3.1.

Use of Power Doppler to Identify Subclinical Synovitis and Risk of Radiological Progression in Rheumatoid Arthritis Despite TNFa Blocker Remission Bernd Raffeiner,1,3 Marwin Gutierrez,4 Viviana Ravagnani,5 Costantino Botsios,2 Mariagrazia Canova,2 Silvano Todesco,2 Paolo Sfriso,2 Leonardo Punzi2 1 Rheumatology Unit, Department of Medicine, University of Padova, Padova, Italy, 2Rheumatology Unit, Clinical and Experimental Medicine, University of Padova, Padova, Italy 3Rheumatology, Internal Medicine of Bolzano, Bolzano, Italy 4Rheumatology Unit, Polytechnic University of the Marche, Ancona, Italy 5Rheumatology, Internal Medicine Carlo Poma, Mantova, Italy

Introduction Functional outcome in patients with rheumatoid arthritis (RA) depends on control of disease activity and arrest of radiological damage. Remission of clinical activity can nowadays be achieved in the majority of patients with RA by treatment with combinations of glucocorticoids, synthetic disease-modifying antirheumatic drugs (DMARDs) and tumour necrosis factor a (TNFa) blockers [1]. Recent imaging studies have identified subclinical inflammation in many patients receiving treatment with synthetic DMARDs and considered to be in clinical remission. Active synovitis was demonstrated by magnetic resonance imaging (MRI) and ultrasound. Power Doppler (PD) positivity correlated best with radiological progression despite clinical remission induced by synthetic DMARDs [2]. TNFa blockers have been found to be considerably more effective than synthetic DMARDs in blocking radiological damage by direct inhibition of osteoclasts [3]. However, it has yet to be established whether PD positivity also predicts ongoing radiological damage in RA patients receiving TNFa blockers who are in clinical remission. The objective of this study was to assess ultrasound alterations detectable in patients receiving TNFa blockers who are considered to be in clinical remission, and to determine the usefulness of PD in predicting structural damage.

S131

Methods A prospective observational study was performed on 109 consecutive patients with RA who were receiving treatment with TNFa blockers. Four Italian rheumatologic centres participated. Informed consent of patients and approval from ethical committee were obtained. Patients in DAS28 remission for 6 months were eligible for inclusion. At baseline after clinical and laboratory evaluation confirming clinical remission, ultrasound examination was performed on all patients on the metacarpophalangeal (MCP), proximal interphalangeal (PIP), wrist and metatarsophalangeal (MTP) joints with MyLab70 (7–18 MHz linear probe; Esaote SpA, Genoa, Italy). The presence or absence of synovial hypertrophy was assessed for each patient. Active synovitis was identified by PD positivity. Number of joints presenting PD positivity was recorded and categorized by semi-quantitative scores (0–3), as defined for PD signal. At baseline and after one year, X-rays were taken of the hands and feet. Radiological progression was expressed as the difference in van der Heijde modified Total Sharp Score (DTSS) [0. Correlations between PD positivity and radiological progression were analyzed by the Spearman test. Multivariate logistic regression was performed to identify predictive factors for radiological progression among clinical, treatment and PD data.

Results A total of 3706 joints were evaluated by ultrasound and PD, 34 joints per patient. Synovial proliferation was found in all patients analyzed in at least one joint side. Of 109 patients, 54 (49.5%) showed no PD signal at the time of remission, whereas 55 (50.5%) had PD signals: 31 (28.4%) grade 1, 15 (13.7%) grade 2 and 9 (8.2%) grade 3. Thirty-four patients (31.2%) showed PD positivity in one joint, 21 (19.2%) in two or more joints. PD positivity was found most frequently in the wrist joints (68.8%), followed by MCP (22.2%), PIP (6.6%) and MTP (2.4%) joints. All patients without PD signals showed no radiological progression. Radiological progression occurred in 27.2% of patients with PD positivity. This corresponded to 13.7% of all patients in clinical remission who were analyzed. The number of PD-positive joints and PD score correlated with radiological progression (p \ 0.001). Logistic regression identified PD grade as a predictor of radiological progression (odds ratio 3.07; 95% confidence interval 1.7–5.4; p \ 0.0001). Clinical, serological, therapeutic and

other sonographic parameters such as synovial hypertrophy were not found to be predictive.

Discussion Remission is the aim of treatment in patients with RA [4]. The state of remission should represent an absence of inflammation with no clinical symptoms or signs and no radiographic progression, and should result in optimal structural, functional, and quality of life outcomes. Modern RA management strategies and therapies enable a more complete suppression of inflammation than in the past, as reflected by the increasing rates of remission that have been reported [5–7]. However, disease remission may not necessarily be associated with a consistently good outcome. A number of reports have suggested a disparity between clinical state and outcome, providing evidence of progression of joint damage despite apparent remission [8, 9]. This observation has been interpreted by some investigators as evidence of dissociation between synovitis and subsequent erosive joint damage [10]. Alternatively, it may reflect the inadequate sensitivity of the traditional clinical approaches to accurately detect synovitis and therefore assess true remission status. Objective imaging techniques definitively identify inflammation and the potential for further joint damage in patients satisfying current clinical remission criteria. This suggests that a definition of remission based only on clinical and laboratory assessments is not an accurate instrument because it does not guarantee an inflammation-free, non-damaging disease state. Also patients are seen only at certain time points and this is not necessarily reflective of sustained remission. Musculoskeletal ultrasound and MRI allow the valid and reliable assessment of both ongoing inflammatory disease activity and structural outcome in RA at the same time. Ultrasound in particular has become part of daily clinical practice for many rheumatologists. For someone well trained and practiced, the method is fast, does not require radioactive or parenteral supplies and can be repeated at each visit. PD can accurately show the presence and grade of synovial inflammation with the potential of joint destruction. In this study as recently reported by our group, PD was demonstrated to be very useful for the evaluation of patients considered to be in remission [11]. Absence of PD signal guarantees the arrest of radiological progression, whereas patients with PD signals are at risk for radiological progression despite treatment with TNFa blockers. This risk increases with higher PD grades. Patients in clinical remission to prevent structural damage and functional loss should carefully be checked by PD for subclinical activity

S132

and considered for treatment changes in case of higher grade positivity.

Conclusion PD assessment of patients with RA who have achieved clinical remission with TNFa blocker therapy can identify subclinical synovitis and predict the occurrence of radiological progression. Acknowledgments Native English editing was provided by Raewyn Poole of inScience Communications, Springer Healthcare. This editorial support was funded by Pfizer.

References 1. Ometto C, Botsios C, Sfriso P, et al. Comparison between adalimumab, etanercept and infliximab in achieving and maintaining clinical remission in rheumatoid arthritis in daily clinical practice. Ann Rheum Dis. 2009; 68(Suppl 3):120. 2. Brown AK, Conaghan PG, Karim Z, et al. An explanation for the apparent dissociation between clinical remission and continued structural deterioration in rheumatoid arthritis. Arthritis Rheum. 2008;58(10):2958–67.

3. Taylor PC. Anti-TNF therapy for rheumatoid arthritis and other inflammatory diseases. Mol Biotechnol. 2001;19(2):153–68. 4. Emery P, Salmon M. Early rheumatoid arthritis: time to aim for remission? Ann Rheum Dis. 1995;54:944–7. 5. Boers M, Verhoeven AC, Markusse HM, et al. Randomised comparison of combined step-down prednisolone, methotrexate and sulphasalazine with sulphasalazine alone in early rheumatoid arthritis. Lancet. 1997;350:309–18. 6. Klareskog L, van der Heijde D, de Jager JP, et al. Therapeutic effect of the combination of etanercept and methotrexate compared with each treatment alone in patients with rheumatoid arthritis: double-blind randomized controlled trial. Lancet. 2004;363:675–81. 7. Korpela M, Laasonen L, Hannonen P, for the FIN-RACo Trial Group, et al. Retardation of joint damage in patients with early rheumatoid arthritis by initial aggressive treatment with diseasemodifying antirheumatic drugs: five-year experience from the FIN-RACo study. Arthritis Rheum. 2004;50:2072–81. 8. Mulherin D, Fitzgerald O, Bresnihan B. Clinical improvement and radiological deterioration in rheumatoid arthritis: evidence that the pathogenesis of synovial inflammation and articular erosion may differ. Br J Rheumatol. 1996;35:1263–8. 9. Molenaar ET, Voskuyl AE, Dinant HJ, et al. Progression of radiologic damage in patients with rheumatoid arthritis in clinical remission. Arthritis Rheum. 2004;50:36–42. 10. Kirwan JR. The synovium in rheumatoid arthritis: evidence for (at least) two pathologies [editorial]. Arthritis Rheum. 2004;50:1–4. 11. Raffeiner B, Sfriso P, Gutierrez M, et al. Usefulness of Power Doppler to predict radiological damage in rheumatoid arthritis patients in clinical remission by TNF blockers. Ann Rheum Dis. 2011;70(Suppl3):76.