Abstract
Psoriatic arthritis (PsA) is an inflammatory musculoskeletal disease with both autoimmune and autoinflammatory features. Evidence supports the distinct nature of PsA regarding its clinical, genetic, immunohistochemical, and imaging features. Such features can help to distinguish PsA from other common rheumatic diseases. Apart from peripheral joint involvement, the musculoskeletal lesions in PsA include enthesitis and involvement of the distal interphalangeal joint (frequently associated with nail involvement, dactylitis, and axial involvement). The traditional model of pathogenesis in PsA has identified it as an autoimmune disease; however, an alternative model classifies it as having autoinflammatory features. Similarly, there are important new genetic observations focusing on the HLA region, and genome-wide association that confirms the genetic heterogeneity of patients with psoriasis and patients with PsA. Newer imaging techniques have also provided a much more detailed characterization of tissue abnormalities, in particular highlighting the extent of new bone formation, which is quite distinct from rheumatoid arthritis.
Psoriatic arthritis (PsA) is an inflammatory musculoskeletal (MSK) disease with both autoimmune and autoinflammatory features. While there are no agreed diagnostic criteria, PsA is characterized by distinct clinical, genetic, immunohistochemical, and imaging features. We outline here each of these distinct features.
PsA IS CLINICALLY DISTINCT
PsA may be defined as a disorder of the MSK system that occurs in association with the skin condition, psoriasis. Apart from peripheral joint involvement, the MSK lesions in PsA include involvement of the enthesis (enthesitis is inflammation of the entheses, the sites where tendons or ligaments insert into the bone) and of the distal interphalangeal joint, which is frequently associated with nail involvement, dactylitis, or “sausage”-shaped digital swelling and axial involvement. Patients frequently present with more than 1 of these MSK lesions. The lesions may present before the development of psoriasis in 10% of cases; in other words, PsA may be diagnosed in patients without the classic rash by meeting other criteria. It can sometimes prove difficult to assess which MSK or skin lesion predominates. Both dermatologists and rheumatologists must try to assess and quantify all the ways in which PsA may affect an individual. Only then can appropriate therapeutic decisions be made that are tailored to the individual patient’s needs. A number of composite disease activity tools1,2 have been proposed that may well fulfill this clinical need, and these are currently being tested and compared in randomized clinical trials.
TRADITIONAL MODEL OF PATHOGENESIS OF PsA
The traditional model of pathogenesis is that PsA is predominately an autoimmune disease driven by T cells reacting to as yet unknown skin or synovial antigens presented by macrophages or dendritic cells. This antigen-specific T cell activation in turn leads to an upregulation of proinflammatory cytokines and to the influx of nonspecific T cells and neutrophils into involved tissues. Evidence supports the important role of T cells in PsA3 (Table 1)4,5,6,7,8,9,10,11,12.
ALTERNATIVE MODEL OF PATHOGENESIS OF PsA
McGonagle and McDermott have proposed a new classification for immunological diseases with PsA that demonstrates both autoimmune and autoinflammatory features13. In autoinflammatory conditions, local factors such as microtrauma at sites predisposed to disease may lead to activation of innate immune cells including macrophages and neutrophils with resultant target tissue damage. It was suggested that this site-specific inflammation might be independent of adaptive immune responses. In further elaboration of this hypothesis, the authors propose that the enthesis is an area of high mechanical stress where, together with the associated bone and adjacent soft tissue, inflammation might be initiated14. While an attractive hypothesis, because some patients with PsA show prominent entheseal or capsular involvement, this falls short of explaining all PsA lesions.
On the basis of these pathogenetic models, perhaps a unifying explanation can be put forward in which microtrauma at entheseal sites exposes entheseal antigens that in a genetically predisposed individual (i.e., HLA genes) trigger antigen-specific T cell responses.
PsA IS GENETICALLY DISTINCT
Multiple studies have implicated several HLA Class 1 alleles in susceptibility to psoriasis. The principal susceptibility HLA allele is HLA-C*0602 (psors1), which is present in 60% of most series15. A group of HLA-B alleles including HLA-B57, HLA-B37, and HLA-B13 are also associated, as they are in linkage disequilibrium with HLA-C*060216,17,18. Other Class 1 alleles implicated in PsA, such as HLA-B27, HLA-B38, or HLA-B39, are not strongly associated with psoriasis.
In relation to PsA, there is little agreement on the actual frequency of HLA alleles. For example, the frequency of HLA-B27 ranges from 17% in Northern Spain to 39% in Taiwan, and the frequency of HLA C*0602 ranges from 29% in Israel to 60% in Northern Spain. There are many factors that can confound identifying susceptibility genes in PsA. Examples are the accuracy of PsA diagnosis, the relationship between psoriasis and PsA, heterogeneity in clinical findings and response to therapy, the difference in the genetic features of PsA, whether the studied cohort is ascertained in a rheumatology clinic or a dermatology clinic, and heterogeneity in the distribution in HLA alleles in different populations.
A recent study has examined the relationship of psoriasis to PsA in the same genetically homogeneous population by comparing ascertainment of a PsA case series with a psoriasis case series, where all individuals with arthritis were specifically excluded following rheumatological assessment19. If psoriasis and PsA are indeed homogeneous, one would expect that the frequencies of HLA-C*0602 and HLA-B locus alleles in linkage disequilibrium would be similar in both cohorts. The results of this particular study indicated that in PsA the frequency of HLA-C*0602 was significantly lower at 28.7%, compared to those with psoriasis only at 57.5% (p = 9.9 × 10(–12)). Further, haplotypes containing B-2705 or B-3901 were significantly increased in frequency in PsA but not in the psoriasis cohort. Being HLA-B27-positive was associated with an interval of 0.98 years between onset of skin and MSK disease, compared to 10.14 years if HLA C*0602-positive (p = 2.5 × 10(–6)). These data suggest that the psoriasis phenotype includes those with the classic psoriasis susceptibility gene HLA-C*06 and is characterized by more penetrant skin disease with less penetrant and more time-dependent MSK phenotype development. PsA appears to be mediated by HLA-B alleles, notably –B27 or –B39, which includes temporally more coincident MSK involvement that is nearly equivalent in penetrance to skin disease.
The results of the first genome-wide association study (GWAS) in PsA have been published20. Apart from confirming previously described associations with the HLA region and with interleukin 12ß (IL-12ß), the results suggested a significant association with the other arm of the sixth chromosome, where there were 4 mutations in TRAF3IP2. This is a region that codes for Act1, a signaling adapter involved in the regulation of adaptive immunity. Given the lack of an autoantibody signature in PsA, it is of interest that Act1 is a negative regulator of humoral immunity, but on the other hand Act1 concomitantly operates as a positive signaling adapter in IL-17-mediated cellular immune responses21,22,23. Early functional studies suggest that the introduction of the risk allele into Act1 results in loss of Act1’s ability to interact with TRAF6, a change that may have important effects on IL-17 signaling downstream. An association with TRAF3IP2 was also found in a recent psoriasis GWAS (Wellcome Psoriasis GWAS), but the association was not as strong, and it is possible that some of the association might relate to the inclusion of PsA cases24. To date, similar associations have not been described in other autoimmune diseases such as rheumatoid arthritis (RA).
PsA SYNOVIAL TISSUE IS DISTINCT
In comparison to the synovial immunohistological features of RA, a significant reduction in the number of macrophages, a reduction in lining layer depth, and an increase in synovial vascularity has been observed in PsA; however, the number of T cells, T cell subsets, and B cells were similar in both cohorts25. Neovascularization is a hallmark of synovitis, and arthroscopic studies have shown distinct vascular patterns of synovium. For example, patients with PsA have predominantly tortuous, bushy vessels, whereas predominantly straight, branching vessels are seen in patients with RA26. Kruithof and colleagues showed an increase in vascularity in PsA as compared to RA in a large series, where they also highlighted an increase in neutrophil infiltration in PsA27. While these changes are potentially of pathogenetic significance, none are specific enough yet to be of diagnostic utility.
PsA IS DISTINCT ON IMAGING
There are a number of plain radiographic features that help distinguish PsA from RA. While there are fewer erosions present in PsA compared to RA, PsA is associated with additional features such as new bone formation, bony resorption, pencil-in-cup type deformity, and asymmetrical spinal involvement. Bony changes develop more slowly in PsA, where 47% of patients have demonstrated erosive change and 29% will demonstrate features of new bone formation within 2 years of disease onset28. Bone edema on magnetic resonance imaging (MRI) scans has been described as the strongest predictor of subsequent radiographic progression, and one study found this present in 50% of patients with PsA at baseline; the situation improved or resolved in all cases following infliximab therapy29. In a study of swollen metacarpophalangeal (MCP) joints in patients with PsA or RA, MRI was not able to differentiate between the 2 diseases on the basis of the entheseal-related disease. However, a subgroup of patients with PsA had diffuse extracapsular enhancement (30%) or diffuse bone marrow edema (20%), suggesting dominant entheseal-related disease30. The RA group had a greater degree of MCP joint synovitis and tenosynovitis than the PsA group. There was no significant difference in either the total number of erosions or the presence of periarticular bone edema between the groups. The authors concluded that enthesitis-associated pathology is not sufficiently common to be of diagnostic utility in PsA.
A recent study using microcomputed tomography has also compared MCP joints in patients with PsA or RA. This study found that both cohorts had the same number of bone erosions, but the erosions in PsA were smaller in size and in depth, and more Ω-shaped and tubule-shaped, whereas U-shaped lesions were mostly typical for RA31. Interestingly, new bone formation was increased in PsA, often affecting the entire circumference of the bone, and forming a “bony corona” or crown. That study strongly suggests that the mechanisms of bone repair may be more active in PsA. Consistent with this observation, a study has shown a consistent increase in levels of the bone isoenzyme of alkaline phosphatase, reflecting new bone formation, in PsA as compared with RA32.
CONCLUSION
Based on the above clinical, genetic, immunohistochemical, and imaging evidence, PsA appears to be a distinct, inflammatory MSK disease with both autoimmune and autoinflammatory features. Further research focusing on genotype:phenotype correlations, the role of the IL-17 pathway, and mechanisms of new bone formation are likely to further inform our emerging concepts of disease pathogenesis.