Psoriatic arthritis (PsA) is a systemic inflammatory disease that affects 20%–30% of patients with psoriasis. It is characterized by inflammation involving the musculoskeletal system and the skin, including its clinical manifestations, the axial spine, peripheral joints, enthesitis, dactylitis and nail and skin lesions.1 Certain patients have a mild course, whereas others can develop radiological joint damage, peripheral joint destruction and functional disability. Systematic reviews (SR) have been carried out, some with a meta-analysis,2–4 on the efficacy of treatment in PsA,2–5 and have demonstrated a low level of evidence (LE) regarding the effectiveness of synthetic disease-modifying antirheumatic drugs (DMARDs). However, in the attempt to offer guidance to specialists, given the heterogeneity of the clinical presentation and the different degrees of severity in joint involvement,1 2 international groups, on the basis of those SR, have formulated recommendations concerning treatment. These groups are the European League Against Rheumatism (EULAR),6 with an algorithm focused mainly on musculoskeletal symptoms, and the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA),7 that evaluated 5 domains (peripheral arthritis, skin and nail involvement, enthesitis, dactylitis and axial arthritis).8 As a result, for active peripheral arthritis, both recommended treatment with DMARDs, such as methotrexate (MTX), sulfasalazine (SSZ) and leflunomide (LEF).

Methotrexate has frequently been utilized as the first DMARD to be administered in PsA because of the efficacy it shows in the treatment of joint and skin disorders.9 Despite the low LE, MTX is still one of the most widely used drugs in PsA.1 Thirty percent of the visits to the rheumatology department of patients referred from dermatology involves the addition of a new DMARD, with MTX being that most frequently employed.10 The scarcity of high-quality clinical trials supporting its efficacy in PsA compelled us to ask the reason for the generalized use of MTX in PsA,1 its effectiveness and for which clinical phenotype of PsA. Moreover, although clinical trials have not shown that the efficacy of MTX versus placebo is sufficiently significant, it is considered that the design of those studies had methodological limitations that make it unviable to draw definitive conclusions.11 Therefore, the objective of this report was to carry out a systematic reassessment of the available evidence on the utility of DMARDs in the management of PsA. This forms part of the process of updating the clinical practice guidelines for the treatment of axial spondyloarthritis and PsA of the Spanish Society of Rheumatology (SER) (ESPOGUIA),12 the previous version of which was issued in 2009.

We performed a SR in 3 databases: Medline, EMBASE and the Cochrane Library. They included studies in English, Spanish and French, dating from January 2008 to November 2014, to retrieve articles on the efficacy of DMARDs in PsA. The terms employed for the search strategy involving high-sensitivity descriptors (MG) are shown in the Supplementary appendix.

Inclusion criteria: using the population, intervention, comparison and outcome (PICO) format, we selected those randomized controlled trials (RCT) that met the following requirements: (1) adults with PsA with a patient population of 50 or more individuals; (2) intervention with traditional DMARDs (MTX, SSZ, LEF) versus placebo; (3) efficacy outcome measures in terms of changes in: (a) peripheral symptoms (American College of Rheumatology [ACR] 20/50/70, EULAR response based on Disease Activity Score in 28 joints, Psoriatic Arthritis Response Criteria [PsARC]; peripheral radiological structural damage [Sharp/van der Heijde score—hands, wrists and feet—modified for PsA, which includes distal interphalangeal joints); (b) axial involvement (Bath Ankylosing Spondylitis Disease Activity Index, Bath Ankylosing Spondylitis Functional Index, Assessment of SpondyloArthritis international Society [ASAS] 20/40 5/6); (c) enthesopathy according to ultrasound or magnetic resonance imaging with number of entheses (Maastricht Ankylosing Spondylitis Enthesitis Score, Leeds Enthesitis Index) on inclusion and at the end of the study (percentage improvement); (d) dactylitis, with number of affected digits at baseline and at the end of the study (percentage improvement); and (e) uveitis, with number of episodes prior to and after treatment. We included SR of RCT (with preference for phase III or IV). We also selected studies that helped to partially respond to the question posed (efficacy and safety of a traditional DMARD versus a traditional DMARD or versus a combination of several traditional DMARDs). When there was no existing evidence on a question because of the design of a SR of RCT, we considered observational studies included in the SR. We excluded reports that did not adjust to the components of the PICO question, as well as abstracts, posters, narrative reviews, letters, editorials and any text that had not been published. The selection of the studies was performed by a reviewer (JM) in successive phases: selection of titles and abstracts from the selected titles, review of the complete text of the selected studies and their evaluation, to eliminate articles that did not meet the inclusion criteria. We also performed a hand-search in the bibliography of the studies included. Doubts that arose during the selection process were weighed by 2 methodologists from the SER (PD, DS) and a consensus was reached in every case. To manage the literature references we utilized EndNote X7. In the critical reading of the articles and the evaluation of the quality, we employed the checklists of the Scottish Intercollegiate Guidelines Network (SIGN).13 To assign the LE, we used the Oxford system.14

We retrieved 1662 documents to review on the basis of their title and abstract (Medline, n=433; EMBASE, n=1132; Cochrane, n=97). In all, 48 articles were selected for full-text reading; of these, 40 were excluded (and can be made available on request). Ultimately, 8 studies met the inclusion criteria (Fig. 1) (Tables 1 and 2 providing evidence synthesis). In all, there were 2 SR,2,5 2 RCT included in 1 of these SR,15,16 an open-label RCT not included in the SR,17 3 observational studies (2 prospective studies18,19 and 1 retrospective20); 2 of the latter18,19 were reviewed in the SR of Acosta Felquer et al.,5 with a LE that is summarized in Table 3. Methotrexate was the drug considered in the intervention in 2 RCT15,16 and in an observational study,20 whereas LEF was the subject in 3 articles17–19: an open-label RCT17 and 2 observational studies.18,19

Fig. 1.

Flow chart indicating the inclusion and exclusion of the studies retrieved.

(0.11MB).

The SR published by Ash et al.2 was carried out in 2012, with studies conducted between 1962 and January 2010. Their objective was to review the available evidence on the efficacy and safety of all the non-topical drug therapies for PsA, from nonsteroidal anti-inflammatory drugs (NSAID) to DMARDs and biological agents, to provide data on the development of EULAR recommendations for the management of PsA (LE 1a). Likewise, in the SR by Acosta Felquer et al.,5 in 2014, the objective was to evaluate the efficacy of different treatments and therapeutic strategies to provide information for the development of new GRAPPA recommendations for the treatment of peripheral PsA, and summarize the evidence on the available therapies. They searched for documentation in Medline, EMBASE and the Cochrane Library, from 2006 to the present. They also reviewed abstracts from the EULAR and the ACR, from 2010 to 2013. In the SR by Ash et al., the authors concluded that there was some evidence to support treatment with NSAIDs and DMARDs, with good evidence of the efficacy of anti-tumor necrosis factor agents in PsA,2 whereas, with respect to DMARDs, the conclusions of the SR of Acosta Felquer et al. were that the basis for the evidence of their effectiveness was not well established.5

In the analysis of the studies retrieved for the SR by Ash et al.,2 with regard to the use of MTX, there were 3 RCT (n=93 patients) that compared MTX in monotherapy vs placebo and 7 open-label or retrospective studies on MTX in PsA. These studies did not meet the inclusion criteria for the present SR. Likewise, when the intervention involved the administration of LEF, several of the studies included did not comply with the inclusion criteria for the present SR, among them a RCT and 2 open-label trials. When the drug employed was SSZ, the reports cited to evaluate its efficacy (7 RCT that compared SSZ in monotherapy vs placebo or symptomatic treatment) corresponded to evidence prior to 2008 and included in the 2009 ESPOGUIA.

The results of the present SR, which is part of the Clinical Practice Guidelines for the Treatment of Axial Spondyloarthritis and Psoriatic Arthritis,12 are in accordance with the findings of the 2 previous SR included2,5; the first determined that there is little evidence of the efficacy of DMARDs in PsA and the second concluded that traditional DMARDs are utilized for the treatment of PsA, although the basis for the evidence of their efficacy is not well established.5 These results are also supported by the fact that similar to the findings in the 2009 Espoguia,21 it could be said that the present SR reveals diverse and asymmetrical evidence from the studies: specifically, with respect to the evaluated outcomes; we should mention that only 1 study took into account radiological damage20 and none of them commented on functional status or quality of life.

The studies involving LEF had similar results when compared. Thus, in the open-label RCT by Asaduzzaman et al.,17 LEF had some degree of efficacy and safety in PsA, as did MTX, since the primary objective was achieved in both groups; in the secondary outcome, there was a significant difference in ACR70 in favor of LEF. Likewise, in the observational study of Asiri et al.,18 LEF resulted in improvement in at least 50% of the patients, whereas concomitant treatment with MTX is shown to be a predictor of achieving the PASI 50 response. Likewise, the observational study published by Behrens et al.19 indicates that treatment with LEF is an effective alternative in the management of peripheral PsA, with a beneficial effect on pain, fatigue and dactylitis. On the other hand, although MTX is the DMARD most frequently utilized in PsA (39%),22 this could be because of the clinical benefit of MTX in cutaneous psoriasis.1 For example, in the study of Asiri et al.,18 a higher percentage of patients from the group taking a combination of MTX and LEF obtained a PASI 50 response, in comparison to the group receiving LEF as monotherapy.

However, there is a lack of evidence on treatment with MTX with respect to other DMARDs in the management of PsA, as occurred in previous reviews.23 The attempt has been made to explain this situation by different authors with a number of hypotheses. Thus, according to some authors, it could be due to the fact that the evidence of the treatment is supported, like that of the outcome measures,24 by its previous utilization in rheumatoid arthritis. Moreover, it can be considered that the number of well-designed clinical trials with relevant outcomes, as well as that of observational studies, is insufficient for the evaluation of the efficacy and safety of DMARDs in PsA.1,18,25,26

This same argument is stressed by Pincus et al.,27 as well as by other authors,28 the former group in a recent analysis on treatment with MTX, in which they evaluate the design of the study, among other causes, in the attempt to explain the influence of a number of factors on the result and on the quality of the evidence; they explore several hypotheses based on the results of the MIPA study, with different theories in terms of the high proportion of drop outs, a relatively low MTX dose and less severe inclusion criteria for PsA and fewer active joints.15,29

As an outcome of the proposed scenario, although studies may be included that result in the failure to achieve the best quality, as the absence of evidence does not imply evidence of absence.30 In agreement with a SR that evaluates the quality of 6 recent clinical guidelines on the treatment of PsA (with the exclusion of biological agents), the consequence would indicate that clinical practice guidelines have sufficient quality. The process of their development intrinsically involves objectivity and they are assessed in clinical practice, although it is considered that there could be some improvement in certain domains or items.26 One important limitation in this SR is the failure to perform a meta-analysis, as the number of studies retrieved was inadequate and their design was highly diverse, a drawback pointed out by Ceponis and Kavanaugh,1 among others. In accordance with the conclusions of the studies and SR evaluated,2,20 although, as suggested by Sevrain et al., it is unlikely that additional RCT be carried out because of the lack of interest in studying old drugs.26 We could conclude by proposing the need to conduct RCT or high-quality observational studies, with outcomes appropriate for the “core sets” proposed by OMERACT.31 They should adequately evaluate the efficacy of DMARDs according to the profile of clinical heterogeneity of PsA and support this indication, to improve the consistency of the level of evidence and, thus, reduce the possible uncertainty involving the therapeutic decision regarding early treatment.

The authors declare they have no conflicts of interest.