Abstract
Although arthritis among patients with psoriasis was described in the 19th century, the question of whether there was a specific form of arthritis associated with psoriasis was not raised until the middle of the 20th century. It was only after the seminal work of Moll and Wright that psoriatic arthritis (PsA) was recognized as a distinct entity. There was little interest in studying the disease until the second half of the 20th century. Initially, it was thought to be a mild disease, but subsequent studies over the past 50 years demonstrated that it was a severe disease, occurring much more frequently than first described and leading to progressive joint damage, disability, reduced quality of life and function, and an increased mortality risk. Comorbidities were found to be more common in this patient population, possibly contributing to the poor outcomes. Advances of new therapies and better assessment tools have led to improvement in the outcomes of patients with PsA. However, there are still unmet needs that will require addressing in the next few years to improve the lives of patients with this disease.
Psoriatic arthritis (PsA) is an inflammatory immune-mediated musculoskeletal (MSK) disease associated with psoriasis. Psoriasis is an inflammatory skin disease that affects 1% to 3% of the global population. The prevalence of PsA was thought to be quite low, with only 7% of patients hospitalized for psoriasis having arthritis,1 but with the development of classification criteria for PsA2 and better recognition of the disease, the prevalence of PsA among patients with psoriasis is between 20% and 30%.3 Therefore, the prevalence of PsA in the general population may vary between 0.3% to 1%.
This article aims to describe the recognition of PsA over the past century, recognize milestones in the management of PsA, and identify unmet needs and future directions in the management of the disease.
The recognition of PsA over the past century
The first report on a relationship between psoriasis and arthritis is attributed to Luis Alibert in an 1818 publication, where he described joint pains in patients with psoriasis.4 This was followed by a publication by Pierre Bazin, who in 1860 coined the term “psoriasis arthritica” to describe patients with psoriasis who had arthritis.5 Charles Bourdillon provided the first description of PsA in 1888.6 Jeghers and Robinson presented a case of a patient with PsA and reviewed the literature.7 They pointed out that dermatologists tended to ignore joint symptoms. They describe the polyarticular nature of the disease, involving both small and large joints, specifically the terminal joints of the fingers and toes. Although these authors were all in favor of the uniqueness of PsA, Walter Bauer in 1939 was of the opinion that the arthritis that occurred in patients with psoriasis was the coincidental occurrence of rheumatoid arthritis (RA) with psoriasis.8 However, several studies describing series of patients demonstrated that a unique form of arthritis was occurring in patients with psoriasis. Vilanova and Pinol described PsA as a unique entity in 1951.9
Although most people recognize the important contribution of Verna Wright to the recognition of PsA as a distinct entity, there was an earlier publication in 1952 by Mary Sherman, an orthopedic surgeon at the University of Chicago.10 She noted the predilection for the distal joints of the hands and feet, as well as the presence of arthritis mutilans and axial involvement. She also noted that the clinical and radiographic features of PsA were distinct from those of RA. Wright recognized some of the features described by Sherman in his 1956 publication.11 Wright described 42 patients with psoriasis and arthritis, of whom 34 had erosive arthritis, 6 degenerative joint disease, 1 gout, and 1 rheumatic fever. Information was available for 38 patients, which was compared to that of 55 patients with rheumatoid factor (RF)-positive RA and a group of 310 patients with psoriasis without arthritis. There were more male individuals among the patients with psoriasis and arthritis than among patients with RA and patients with psoriasis without arthritis. Family history of psoriasis was more common in psoriasis with arthritis than in RA. Patients with psoriasis and arthritis were more likely to have nonpolyarticular disease onset than those with RA. Psoriasis preceded the arthritis anywhere from 2 months to 37 years. In 23% of the patients, the skin lesions came after joint disease. There was a higher prevalence of nail lesions among patients with psoriasis and arthritis compared to those with psoriasis alone. Two patients who had a positive RF were thought to be examples of RA with coincidental psoriasis, but the rest were considered a distinct entity. A further comparison between RA and psoriasis-associated arthritis was published by Wright in 1959. Again, he noted that 16% of the patients presented with arthritis before the diagnosis of psoriasis.12 Wright then went on to describe the radiographic differences between PsA and RA.13 He compared 91 patients with seronegative erosive arthritis and psoriasis matched on sex and age with patients with RF-positive RA. He compared radiographs of the hands and feet as well as sacroiliac joints (SIJs). The distal joints were more commonly affected in PsA than RA, and the distal joints of the toes were also involved more often. He also described tuft resorption, which is a feature of PsA. The proximal interphalangeal joints were more commonly affected in patients with RA. Sacroiliac changes occurred more commonly in PsA than RA. Arthritis mutilans was noted in 4 patients with PsA but not in RA. All 4 had sacroiliac changes.
These studies convinced the American Rheumatism Association (now the American College of Rheumatology [ACR]) to recognize PsA as a unique entity. Moll and Wright published their seminal paper on PsA in 1973, where they defined PsA as a psoriasis-associated inflammatory arthritis usually seronegative for RF.14 They also described the clinical characteristics of PsA as the following: equal sex distribution; involvement of small joints in an asymmetric distribution; the presence of sausage digits, arthritis mutilans, and spondylitis; gout-like onset; and a high incidence of nail involvement. In 1976, Wright and Moll published their book introducing the concept of the spondyloarthritis (SpA) group of diseases. In the chapter on PsA, they describe the clinical patterns of the disease, including the distal arthritis, oligoarthritic often asymmetric, polyarticular that may be indistinguishable from RA, predominantly axial, and arthritis mutilans. They also highlight the shared features within the SpA group of diseases.15
Although the clinical patterns of PsA are well recognized and there has been a tendency to consider these as classification criteria, these features may be present at presentation but during follow-up, they do not necessarily persist. Patients may begin with distal involvement and develop polyarthritis or spondylitis; others may present with spondylitis but acquire peripheral arthritis during follow-up. Some patients begin with polyarthritis but, with treatment, the joint count is reduced, and they may remain with oligoarthritis. For that reason, the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) developed the concept of PsA domains. These include peripheral arthritis, which may be distal, oligoarticular, or polyarticular; axial disease, which usually occurs with peripheral involvement but may be isolated in 2% to 4% of patients with PsA; dactylitis, characterized by swelling of the whole digit that may result from inflammation in all 3 joints of the digit, with soft tissue inflammation as well; enthesitis, manifested as inflammation at the insertion of ligaments and tendons into bone; skin psoriasis; and nail lesions.16 It is recommended that all domains be assessed in each patient with PsA.
Milestones in managing PsA
Incidence and prevalence. Based on the work of previous investigators, PsA was thought to be infrequent and mild compared to RA. We initiated a prospective study of patients with psoriasis, whose psoriasis was confirmed by a dermatologist and who were confirmed not to have PsA at presentation to the program.17 Patients with psoriasis were followed at yearly intervals according to the same standard protocol used in our PsA program.18,19 Over the first 8 years of the program, 51 patients were diagnosed with PsA based on the assessment by a PsA clinic rheumatologist. The annual incidence of PsA was estimated at 2.7%. If we include 9 additional patients who were suspected of having PsA according to the Toronto Psoriatic Arthritis Screen (ToPAS),20 then the annual incidence increases to 3.2%. Out study also showed that the risk for developing PsA is linear. Thus, the concept that PsA is unlikely to develop in patients who have had psoriasis for more than 10 years is incorrect. A similar observation was made in Germany.21 Interestingly, an ultrasound study, which used the Glasgow Ultrasound Enthesitis Scoring System (GUESS), showed a high frequency of subclinical enthesitis in patients with psoriasis without arthritis. After 3.5 years of follow-up, 7 of 28 patients fulfilled the Classification Criteria for Psoriatic Arthritis (CASPAR) for PsA, suggesting an annual incidence of 5%.22
Axial PsA. The prevalence of axial disease among patients with PsA varies with disease duration. In early disease, the reported prevalence ranges from 5% to 28%, whereas in longstanding disease it ranges from 25% to 70%. These frequencies depend on what definition was used for axial disease.23 A study that compared patients with ankylosing spondylitis (now termed radiographic axial SpA [r-axSpA]) with psoriasis to patients with axial PsA (axPsA) demonstrated that the frequency of HLA-B*27 is higher in patients with r-axSpA, whereas arthritis is significantly more frequent in axPsA. The mobility assessment using the Bath Ankylosing Spondylitis Metrology Index (BASMI) is higher in patients with r-axSpA, and they are more likely to have grades 3 and 4 sacroiliitis.24 Moreover, patients who present with axPsA without peripheral disease, who would be expected to resemble patients with r-axSpA and psoriasis, show differences. Patients with axPsA are older at diagnosis, they have a higher Psoriasis Area and Severity Index (PASI), and more have nail lesions than patients with r-axSpA and psoriasis; on the other hand, the latter are more likely to have inflammatory back pain, higher BASMI scores, higher Health Assessment Questionnaire (HAQ) scores, lower 36-item Short Form Health Survey physical component and mental component scores, and a higher frequency of HLA-B*27.25 Thus, axPsA seems to be a distinct entity.
To address the issue of whether axPsA is a separate entity and develop criteria for its classification, the Assessment of SpondyloArthritis international Society (ASAS) and GRAPPA performed a study of Axial Involvement in Psoriatic Arthritis cohort (AXIS). AXIS was designed as a cohort study, where over 400 patients with PsA with < 10 years of symptoms and who had not been exposed to biologic or targeted therapy were recruited worldwide through members of AXIS and GRAPPA. Complete assessment, including a detailed history, physical examination, laboratory evaluation, and imaging of the spine and SIJs (both through radiographs and magnetic resonance imaging [MRI]) were performed on each participant. The images were read locally as well as by a central imaging review committee, and the records were all reviewed by a central clinical committee. Preliminary results demonstrated that axial involvement after the initial evaluation by the local rheumatologist was 37%. After the local rheumatologists were provided with the central imaging reading, the frequency of axPsA dropped to 27%. Back pain (including inflammatory back pain), HLA-B*27 positivity, higher levels of C-reactive protein (CRP), and the presence of active inflammatory and structural changes in the SIJs and spine on MRI were associated with the final conclusion on the presence of axial involvement.26
Prognosis. Earlier studies in PsA, including those of Wright, suggested that the disease was milder than RA. However, over the past 5 decades it has become clear that the disease is more severe than initially thought. Twenty percent of patients with PsA develop clinical deformities and damage resulting in functional disability.27 After 10 years of follow-up, 55% of patients have 5 or more deformed joints.28 In a study of early PsA, where patients were recruited if they had symptoms for < 2 years and, on average, presented within 10 months of symptom onset, Kane et al found that 27% already had at least 1 erosion at presentation to the clinic. At 2-year follow-up, 47% of the patients had at least 1 erosion despite the fact that over half of the patients had been treated with disease-modifying antirheumatic drugs (DMARDs).29
An important aspect of investigations in PsA has been identifying the predictors for clinical and radiological damage. We initially demonstrated that the number of tender and swollen joints at presentation to the clinic as well as the initial erythrocyte sedimentation rate (ESR) were predictive of progression of clinical damage, as measured by the increase in the number of clinically damaged joints.30 We then looked for genetic predictors and identified the following HLA antigens: HLA-B27 in the presence of HLA-DR7, HLA-B39, and HLA-DQw3 in the absence of HLA-DR7; conversely, HLA-B22 was protective. HLA-B39 was associated with the first development of damage, whereas the other antigens were associated with progression from any stage of damage. The addition of the clinical variables did not improve the model.31 We found that radiological damage often occurs before clinical damage.32 The predictors for progression of both clinical and radiological damage include evidence of inflammation—both tender and swollen joints at the previous visit were predictive of progression of damage in the current visit. In addition, age, time in clinic, and initial ESR were predictive.33 Previous damage is also associated with progression of damage. Further confirmation that inflammation leads to damage came from a study that investigated the occurrence of damage in individual joints.34 The analysis of the hand and foot joints demonstrated that the activity (tenderness and/or swelling) history of the specific joint is associated with subsequent damage in that joint. Digits with dactylitis also demonstrated more radiographic damage than digits without dactylitis.35 Whether this is due to the number of joints inflamed within the same digit or the soft tissue inflammation is currently unclear. Thus, if inflammation leads to damage, and damage leads to progression of damage, it behooves us to treat the inflammation well so that damage can be prevented in the first place. Further investigations into the genetic predictors for progression of joint damage showed that, in addition to the HLA antigens and alleles, killer immunoglobulin receptor gene KIR2DS1 is associated with damage progression.31,36 The HLA-DRB1 RA shared epitope has also been associated with erosive disease in PsA, as has the interleukin (IL)-4I50V allele.37,38
Initial investigations demonstrated that patients with PsA are at increased risk of mortality compared to the general population.39 More recently, we have shown in our PsA clinic that the survival among patients with PsA has improved.40,41 There was still an increase in mortality in patients aged 20-39 years, but older patients did not demonstrate an increased risk. In time-dependent Cox regression models, factors associated with mortality included presence of nail lesions and comorbidities including heart disease and cancer, whereas a higher level of education was protective.41 Interestingly, a population-based study using the Ontario population identified a standardized mortality ratio (SMR) for patients with PsA at 1.59 in 2008; even in 2016, it was higher than the SMR we found in our clinic population (0.92).42 One wonders whether patients followed in a specialized clinic fare better because of the close follow-up and introduction of advanced therapy.
Importance of early diagnosis. Another important observation is the fact that late diagnosis is detrimental to patients with PsA. In an observational study of 1077 patients, 641 of whom presented after 2 years from diagnosis and 436 of whom who presented to the clinic within 2 years from diagnosis, patients with late referral were significantly more likely to develop damage progression than those who presented early. Even in comparing those who presented in the second year from diagnosis to those who presented within the first year, there was already an increased risk of progression, and the risk increased with longer disease duration at presentation.43 Likewise, the study of early PsA from Dublin showed that even a 6-month delay in diagnosis was associated with more severe disease.44 Thus, patients with PsA must be identified and treated early so that damage will be prevented.
How do we identify patients with PsA early? A number of clinical and laboratory features might help identify PsA early. The extent of psoriasis, measured by the body surface area (BSA), and the site of psoriasis, including the scalp and the intergluteal areas, have been suggested as being associated with the development of PsA.45,46 Nail lesions have been consistently associated with the development of PsA in patients with psoriasis.17 Several laboratory tests are also predictive of developing PsA, including the presence of HLA-B*27,47 increased levels of matrix metalloprotein 3 (MMP3), and an elevated CRP.48 A multivariable PsA risk estimation tool (Psoriatic Arthritis Risk Estimation Tool [PRESTO]) has been developed based on clinical data from a prospective study of 635 patients with psoriasis (confirmed by a dermatologist) who were assessed by a rheumatologist at baseline to rule out the presence of PsA, and who were followed prospectively at yearly intervals. Fifty-one of the patients developed PsA within 1 year, and 71 developed PsA by year 5. The risk for developing PsA at year 1 was associated with younger age, male sex, family history of psoriasis, back stiffness, nail pitting, joint stiffness, use of biologic medications, and patient global health and pain severity, whereas the risk of developing PsA within 5 years was associated with morning stiffness, nail lesions, psoriasis severity, fatigue, and use of systemic nonbiologic medications or phototherapy.49 The same prospective study was used to predict PsA in patients with psoriasis using DNA methylation profiles.50 Thirty-six highly relevant methylation markers across 15 genes and several intergenic regions were identified. A classification model based on these markers identified converters and nonconverters, with an area under the receiver-operating characteristic curve of 0.96. The model was not improved with the addition of clinical information. Thus, there may be biomarkers that will identify patients with psoriasis at risk to develop PsA.
The development of CASPAR was helpful to rheumatologists in identifying PsA early.2 According to these criteria, a patient can be classified as having PsA if they have an inflammatory MSK disease, including peripheral arthritis, spondylitis, or enthesitis. Patients can be classified with high sensitivity and specificity as having PsA if they score 3 points based on the following criteria: evidence of psoriasis based on either the current assessment (2 points), or a history of psoriasis (1 point), or a family history of psoriasis (1 point); presence of nail lesions (1 point); presence of dactylitis or a history of dactylitis recorded by a rheumatologist (1 point); negative RF by nephelometry (1 point); radiological evidence of juxta-articular ill-defined new bone formation (1 point).
However, only an expert in rheumatology can identify inflammatory MSK disease reliably. Therefore, several groups have developed screening questionnaires that can be administered to patients in order to identify those who should be referred to rheumatology for early identification of PsA.51,52 Although all the screening tools developed to date functioned well in the development and early validation stage, when compared, the results have been conflicting. It is probably fine to use one of the screening tools as a first step and then have the rheumatologist assess the patient. It should be noted, however, that even rheumatologists do not always make the diagnosis of PsA in a timely manner.53
Comorbidities. Comorbidities are common among patients with PsA and add to the disease burden for these patients. These include cardiovascular disease,54 diabetes mellitus,55 metabolic syndrome,56 gout,57 depression and anxiety,58 and malignancy.59 It is important to appreciate that not only do these comorbidities occur more frequently among patients with PsA than the general population, but for the most part, they are related to disease activity.60 Therefore, another reason to treat patients early and appropriately is to prevent the development of comorbidities. Moreover, both obesity and anxiety and depression have been associated with lower frequency of achieving a state of minimal disease activity (MDA).61,62
What should patients with PsA be treated with? Since it is clear that patients with PsA would benefit from early diagnosis and prompt treatment, the question is what is the best way to manage them. Conventional DMARDs, including hydroxychloroquine, gold, penicillamine, azathioprine, cyclosporine, and methotrexate, have become available for the treatment of PsA in the past 100 years. However, none have been demonstrated to modify the course of the disease.63
Studies into the pathogenesis of PsA have identified targets for therapeutic interventions, which led to the development of biologic and targeted therapies for the treatment of PsA.64,65 These advanced therapies seem to work much better to control the inflammatory features of the disease as well as prevent progression of damage. The Figure provides the timeline for the introduction of various therapies in Canada. There is no need to present the results for each of the medications in this article (see Kharouf et al63 for details on PsA medications). It is sufficient to say that all these advanced therapies work similarly well for the joints, but there are differences in the effect on the skin. For example, etanercept does not work as well for the skin as some anticytokine agents. There are also differences in the effect on extra-MSK manifestations, including uveitis and inflammatory bowel disease (IBD). In particular, etanercept does not work for IBD. There is concern about IL-17 inhibitors aggravating IBD. Tumor necrosis factor inhibitors and IL-17 inhibitors work well for axPsA. In general, the drugs are relatively safe, although there is an increased risk for infection as well as concerns about reactivation of tuberculosis and injection site or infusion reactions.
Timeline of treatment advances in psoriatic arthritis. TNFi: tumor necrosis factor inhibitors; IL-12/23i: interleukin 12/23 inhibitor; JAKi: Janus kinase inhibitors; PDE4i: phosphodiesterase 4 inhibitor; Ps: psoriasis.
Monitoring treatment response. Initial response measures included the ACR 20%, which was borrowed from RA. More recently, a specific measure designed for PsA was developed: the Disease Activity Index for Psoriatic Arthritis (DAPSA).66 Although these 2 response measures assess the joints only, one needs to assess the other domains of PsA. For the skin assessment there is PASI.67 Initially, a 50% reduction in the PASI score was considered adequate, but over the past 20 years, the measure of response has become increasingly more stringent to 75%, 90%, and, more recently, 100% reduction in the PASI. In order to consider some of the other domains of PsA, composite measures have been developed. One such measure is the MDA criteria.68 These were developed through a GRAPPA survey. The criteria include tender joint count ≤ 1, swollen joint count ≤ 1, PASI ≤ 1 or BSA ≤ 3, patient pain on a visual analog scale (VAS; out of 100) ≤ 15, patient global disease activity on a VAS ≤ 20, HAQ score ≤ 0.5, and tender entheseal points ≤ 1. If a patient meets 5 of the 7 criteria, they are considered in MDA. If they meet all 7 criteria, they are considered in very low disease activity. This tool is recommended for use as the target for treatment in PsA.69
The Tight Control of PsA (TICOPA) study compared responses in patients treated according to standard of care, followed quarterly, with patients in a tight control, followed monthly; both were treated first with methotrexate. If they did not achieve low disease activity, leflunomide was added, and if that was not sufficient, a biologic agent was introduced. The study demonstrated superiority for the tight control in reducing disease activity, although there was very little progression of damage in both groups.70
Over the past 20 years, the concept of guidelines for the management of many diseases has evolved, including for PsA. There are European guidelines, an American guideline, and individual country guidelines. The GRAPPA guidelines are international and have evolved to include all the extra-MSK manifestations as well as comorbidities. The most recent guidelines were published in 2022.71 The guidelines are set according to the domains of PsA and address peripheral arthritis, axial disease, enthesitis, dactylitis, psoriasis, and nail lesions, as well as inflammatory bowel disease and uveitis. They recommend conventional DMARDs first, followed by biologic agents, but there is otherwise no preference for first-line therapy. The guidelines recommend evaluation of all domains, identifying the most bothersome domain for each patient, and using the therapy that would address not only that domain but all the domains that are present. Prognostic indicators should be taken into account, patients should be followed and monitored regularly, and it should be determined whether the patient achieved their target. Shared decision making between the physician and patient is advised. If the target agreed upon by the physician and patient is not achieved, switching therapy is recommended.
Unmet needs and future directions in the management of PsA
Difficult-to-treat PsA. The concept of “difficult to treat” has been recognized in many chronic diseases, including PsA. A recent online survey was distributed to healthcare professionals, undertaken through GRAPPA.72 The survey was completed by 223 physicians, of whom 80% were rheumatologists and 20% were dermatologists. Eighty-two percent of the participants favored 2 terms to be considered: “difficult to treat” and “complex to manage.” The first term describes patients who have persistent disease activity despite multiple appropriate treatments, whereas “complex to manage” includes other reasons that might prevent the patient from improving. However, there was less than 50% agreement on specific treatment failure criteria. This concept requires additional refinement before it can be further studied.
Unmet needs. Although the available advanced therapies for patients with PsA have made a significant difference in patients’ lives, they do not work for more than 50% of patients. Therefore, additional medications are necessary. Better understanding of the pathophysiology of PsA might provide additional targets for treatment. We need to address compliance issues. We also need to manage the psychological and physical concerns of patients with PsA. A multidisciplinary team approach will likely help manage the disease better. We need to figure out the best way to develop this approach.
To develop personalized medicine for PsA, we need to identify patients with psoriasis at risk to develop PsA. Although some progress has been made, we do not have replication studies to confirm the findings in other cohorts. We need to reliably identify patients who are likely to progress so that they can be treated more aggressively. It would be useful to have biomarkers for disease activity, as the assessments we currently use are not ideal. We need to identify biomarkers for drug response or lack thereof so that we provide the right drug for the right person at the right time.
In summary, over the past century, the concept of PsA as a distinct entity has evolved, with the recognition that it is not just skin and joint inflammation but also comorbidities. Inflammation leads to damage and thus needs to be treated before damage has occurred. The drugs currently available are helpful, but not for all patients. What is needed is better assessment of disease activity, early diagnosis and treatment, biomarkers for disease recognition, and biomarkers for response and toxicity.
Footnotes
FUNDING
The work described has been supported by grants from the Krembil Foundation, the Canadian Institutes of Health Research, the National Psoriasis Foundation, and the Arthritis Society.
COMPETING INTERESTS
DDG has received grant support and/or consulting fees from AbbVie, Amgen, BMS, Eli Lilly, J&J, Novartis, Pfizer, and UCB.
- Copyright © 2025 by the Journal of Rheumatology
This is an Open Access article, which permits use, distribution, and reproduction, without modification, provided the original article is correctly cited and is not used for commercial purposes.
