With the growing number of treatment options for psoriatic arthritis (PsA), therapeutic decision-making has shifted to an increasingly tailored and patient-centered approach. A number of factors contribute to the treatment decision-making process, including age, insurance restrictions, route of administration, side effect profile, comorbidities, and extra-articular manifestations of the disease. In this article, we discuss an extra-articular comorbidity, uveitis, which is frequently seen in patients with PsA. We discuss clinical characteristics of uveitis associated with PsA and describe how the presence of uveitis influences our treatment approach to PsA, based on existing data.
Uveitis refers broadly to inflammation of the uvea, the vascularized and pigmented layer of the eye composed of the iris, the ciliary body, and the choroid. While infection is a common cause of uveitis, many cases are noninfectious and are often associated with an underlying autoimmune or systemic inflammatory disorder. Uveitis is frequently reported in diseases in the spondyloarthritis (SpA) family, including axial spondyloarthritis (AxSpA) and reactive arthritis, as well as PsA. Exact estimates of the prevalence of uveitis in PsA vary widely from 7%-25%, depending on the particular cohort studied. 1,2 In all forms of SpA, the anterior chamber of the uvea is the most likely to be affected. 3 However, compared to patients with AxSpA, patients with PsA appear to have a higher rate of posterior involvement. In addition, patients with PsA appear to have higher frequencies of insidious, bilateral uveitis, as compared to the acute, unilateral, anterior uveitis that is most characteristic of AxSpA. 4 Women with PsA may be more likely than men to experience uveitis, although this has not been a consistent finding. 5
Patients with PsA who are human leukocyte antigen B27 (HLA-B27) positive may be at risk for more severe and refractory anterior uveitis compared to those who do not express the allele. 5 Those who are HLA-B27 positive are also known to have higher rates of axial involvement. It has therefore been postulated that 2 phenotypes of uveitis may exist in PsA: patients who are HLA-B27 positive who have axial disease and severe, unilateral anterior uveitis reminiscent of other forms of SpA, and patients who are HLA-B27 negative, often women, with peripheral-predominant arthritis who are prone to the classic anterior uveitis but may also develop atypical bilateral, insidious, and/or posterior involvement. 4 Specific characteristics of PsA may also provide information about the risk for developing uveitis. For example, dactylitis has been linked to a higher risk of developing uveitis in some, but not all, cohorts of patients with PsA, and the risk of uveitis in PsA has been found in many studies to correlate with longer duration of disease. 6-8
The presence of uveitis signals a disruption in the blood-retina barrier and the subsequent entrance of inflammatory cells into the eye. An entire explanation of pathogenesis is beyond the scope of this article; however, it is worth noting that many of the inflammatory mediators of active uveitis mirror those of PsA. For instance, both the mesenchymal cells in enthesitis and the cells of the ciliary body express receptors for interleukin (IL)-23, suggesting a potential role of the signaling pathways involving this cytokine in both diseases. 9 Another study found increased serum levels of IL-17, a known mediator of PsA disease, in patients with active uveitis. 10 Despite these common pathogenesis links, there are limited data on the utility of certain existing PsA treatments on uveitis manifestations.
Our approach is always to manage uveitis associated with PsA in collaboration with a specialized and experienced ophthalmologist. Uncontrolled uveitis can be vision threatening and contribute to long-term morbidity associated with PsA, so timely recognition, evaluation, and appropriate treatment are important. Ocular glucocorticoid (GC) drops may be used as first-line therapy, particularly for anterior uveitis, to quickly quell inflammation. Escalation to systemic GCs for more severe or posteriorly localized disease may be considered carefully, given the known risk of worsening skin psoriasis (PsO) with GC withdrawal after a course of therapy. Use of GC-sparing therapy should be determined on a case-by-case basis. While generalized, noninfectious uveitis often resolves with GC treatment, the risk of uveitis recurrence in patients with PsA and the challenges of systemic GCs with PsO lead us to frequently consider GC-sparing therapy that addresses ocular, musculoskeletal, and cutaneous manifestations. Tumor necrosis factor inhibitors (TNF-I) are our typical first-line considerations for GC-sparing therapy in patients with PsA with inflammatory joint symptoms and uveitis, although nonbiologic therapy can be considered first-line therapy in select populations.
Data establishing the efficacy of TNF-I come largely from randomized controlled trials (RCTs) of adalimumab (ADA) compared to placebo in noninfectious uveitis. 11 While these trials focused on idiopathic posterior or pan-uveitis, these data have been extrapolated to SpA-associated anterior uveitis, and large registry analyses have supported use of TNF-I in this population. 12 When selecting a particular TNF-I in a patient with current or past uveitis, we frequently start with ADA, based on supportive, albeit uncontrolled, data suggesting a reduction in the risk of recurrence with this agent in patients with SpA and uveitis. 12 For patients who are unable to tolerate subcutaneous injections, who fail ADA, or who we suspect will require higher, titratable dosing, we favor infliximab infusions. Other data suggest that golimumab and certolizumab are also reasonable alternatives. 13,14 We do not generally use etanercept, as the limited data that are available suggest that it is less effective at reducing risk of uveitis recurrence. 12 Methotrexate or leflunomide may be an appropriate first line choice for patients with peripheral-predominant PsA and uveitis, but it is important to note that these agents are not effective for axial disease.
Despite the mechanistic data implicating the role of IL-17 in uveitis associated with PsA, the IL-17A inhibitor, secukinumab, failed to show a reduction in uveitis recurrence, compared to placebo, in pooled analysis of RCTs of noninfectious uveitis. 15 However, a phase 2 trial of intravenous secukinumab in noninfectious uveitis showed promise, possibly because this dosing regimen can achieve higher effective concentrations. 16 It is not our current practice to use secukinumab or ixekizumab as a first-line therapy in patients with PsA and concurrent uveitis, owing to a lack of data supporting efficacy. A novel IL-17A/F inhibitor, bimekizumab (BKZ), has recently been used in several successful phase 3 trials in patients with both TNF-naïve and TNF-nonresponder SpA, including AxSpA and PsA. 17 Interestingly, data from the phase 2 and 3 trials of BKZ found low incidence rates of uveitis in patients with SpA treated with BKZ compared to placebo, suggesting that BKZ might be more effective in uveitis than other IL-17 inhibitors, but these data need to be confirmed.
Successful use of Janus kinase (JAK) inhibitors in noninfectious uveitis, including cases associated with inflammatory arthritis, has been described in case reports as well as in current phase 2 trials. 18 The dual IL-12/IL-23 inhibitor, ustekinumab, also showed initial promise in a small, nonrandomized, uncontrolled phase 1/2 study of the treatment of posterior uveitis, as well as success in few case reports of PsA-associated uveitis. 19 However, a post-hoc analysis of extra-intestinal manifestations, including uveitis and iritis, in patients with inflammatory bowel disease treated with ustekinumab found no benefit in preventing or treating ocular disease compared to placebo. 20 Given the paucity of available data, JAK inhibitors and the IL-12/IL-23 inhibitor, ustekinumab, are not part of our typical treatment algorithm for patients with PsA-associated uveitis.
In conclusion, uveitis is a frequent extra-articular comorbidity of PsA, and it may present differently than the typical acute onset, unilateral anterior uveitis seen in SpA. While uveitis may share many different immunologic threads with PsA, the most convincing data support the use of TNF-I as a GC-sparing agent in this setting, particularly ADA, infliximab, golimumab, or certolizumab. Our approach is generally to start with these agents or methotrexate when directed therapy is needed for uveitis in PsA. Further investigation into the use of the IL-17A/F inhibitor BKZ and JAK inhibitors, as well as tyrosine kinase 2 inhibitors, in PsA associated uveitis may yield additional options for our patients. 21