Psoriatic arthritis for dermatologists

Abstract

Psoriatic arthritis (PsA) affects up to one-third of patients with psoriasis. It is the major comorbidity of psoriasis because of the likelihood that loss of function and permanent disability will develop if initiation of treatment is delayed. Dermatologists are uniquely positioned to recognize early signs of PsA and be the first-line healthcare practitioners to detect PsA in patients with psoriasis. PsA can affect six clinical domains: peripheral arthritis, dactylitis, enthesitis, psoriasis, psoriatic nail disease, and axial disease. However, not every patient will have involvement of all domains and the domains affected can change over time. Complicating the diagnosis is the condition’s similarity with other arthritic diseases and potential heterogeneity. In this article, we provide practical guidance for dermatologists for detecting PsA in patients with psoriasis. We also review the available treatment options by each clinical domain of PsA and give advice on how to interpret the results of PsA clinical trials. Through early recognition of PsA in patients with psoriasis and initiation of proper treatment, dermatologists can help to prevent PsA disease progression, irreversible joint damage, and resultant permanent disability, and improve quality of life.

Keywords:

• Psoriatic arthritis
• psoriasis
• peripheral arthritis
• dactylitis
• enthesitis
• psoriatic nail disease
• axial disease
• disability
• loss of function

Introduction

Psoriatic arthritis (PsA) is a potentially progressive, erosive, chronic, heterogeneous, and systemic inflammatory disease that develops in up to 30% of patients with psoriasis and can manifest in up to six different clinical domains, including peripheral arthritis, dactylitis, enthesitis, psoriasis, psoriatic nail disease, and axial disease (1–3). Peripheral arthritis can cause pain in a variety of joints in PsA and commonly involves the knee (41%), finger (26%), hip (19%), ankle (19%), and wrist (16%) (3). Dactylitis is colloquially referred to as ‘sausage digit,’ and is a distinguishing feature of PsA, characterized by uniform swelling of an entire digit that occurs in up to 48% of patients (4). Another distinguishing feature of PsA, enthesitis is present in 35% of patients and is defined as inflammation where the tendon, ligament, or joint capsule inserts into bone (5–7). Psoriatic lesions can develop on the skin and may affect the nails (e.g. pitting). Axial symptoms, which may occur in up to 50% of patients with PsA, result in back stiffness and pain that improves with movement (8,9).

Etiologically, the familial aggregation of psoriasis and PsA is indicative of a genetic basis for psoriatic disease. The major histocompatibility complex is a known susceptibility locus for PsA and psoriasis (10–12), as shown by the observation that nearly 25% of patients with PsA are positive for human leukocyte antigen (HLA)-B*27 (13). Specific HLA alleles are associated with different PsA manifestations, including symmetric sacroiliitis (HLA-B*27:05), asymmetric sacroiliitis (HLA-B*08:01 and HLA-C*07:01), enthesitis (HLA-B*27:05 and HLA-C*01:02), dactylitis (HLA-B*27:05 and HLA-B*08:01), and synovitis (HLA-B*08:01) (14). HLA-Cw*0602 is more common in PsA than the general population and is associated with an earlier age of psoriasis onset (15). Additionally, there is a strong familial aggregation of PsA with first- and second-degree relatives of affected individuals having increased risk of PsA (16,17).

The pathophysiology of PsA is similar to that of psoriasis and involves important cytokines in the interleukin (IL)-23-Th17-IL-17 and tumor necrosis factor (TNF) pathway (e.g. IL-12, IL-17, IL-23, and TNF-α), as well as cytokines such as IL-22 (13,18,19). Thus, it is not surprising that many cytokine-targeting biologics and small-molecule inhibitors that provide effective skin clearance in psoriasis also improve joint symptoms and slow radiographic progression in PsA.

Although PsA is a widely recognized comorbidity in psoriasis, nearly 52% of patients with psoriasis have joint pain without a diagnosis of PsA and the average diagnostic delay for PsA is 5 years (3,20). Dermatologists have an essential role in reducing this diagnostic delay by screening patients with psoriasis for PsA, as 85% of patients develop psoriasis before PsA, and in these cases, PsA frequently develops within 10 years following the appearance of psoriasis (21–23). There are a number of comorbidities that are more common in patients with PsA than in patients with psoriasis without signs of PsA, including inflammatory bowel disease, cardiovascular diseases (including obesity, hypertension, and type 2 diabetes mellitus), uveitis, depression, anxiety, and fatty liver disease (24,25). Thus, this indicates a need for screening all patients with psoriasis.

Early detection and treatment of PsA are critically important for improving long-term patient outcomes, as the disease can progress rapidly and cause irreversible joint damage (21,26). Specifically, diagnostic and treatment delays of more than 6 months contribute to peripheral joint erosions and poor functional outcomes (27). When PsA is identified early, initiation of treatment with targeted therapies can significantly improve the signs and symptoms of PsA and prevent radiographic progression (28–30).

The remainder of this review will provide dermatologists with practical insights into how to recognize the varied clinical manifestations of PsA, properly diagnose PsA, and interpret results from clinical trials of PsA.

Diagnosis of PsA

Inflammatory versus non-inflammatory arthritis

Recognizing subtle symptomatic differences between inflammatory and non-inflammatory arthritis is critical for proper diagnosis, as these diseases share similar characteristics such as pain and joint stiffness. In general, clinical characteristics of inflammatory arthritis include prolonged stiffness upon immobility, joint pain, joint tenderness, joint swelling, pain at entheses, swollen digits, and worsened back pain upon immobility (9,31). Patients with non-inflammatory arthritis typically present with normal acute-phase reactants, bony crepitus, asymmetric joint involvement, distal interphalangeal (DIP) joint involvement, a lack of prolonged morning stiffness, and bony outgrowths around the joints (9,31). Of note, non-inflammatory arthritis may be suspected in patients with PsA because inflammatory markers are normal in ∼50% of patients with PsA and DIP involvement in PsA may be confused with osteoarthritis on plain radiography (31,32). Questions to help practitioners identify inflammatory arthritis are presented in Table 1.

Table 1. Questions to ask patients to help identify the presence of inflammatory arthritis.

‘Yes’ supports the presence of inflammatory arthritis
Do you have a family history of psoriasis or psoriatic arthritis?
Have you ever had tender/swollen fingers or toes?
Do you experience morning stiffness lasting longer than a half hour to 1 h?
Does your pain improve with exercise?
Have your symptoms of joint pain persisted for > 3 months?
Does your pain improve with nonsteroidal anti-inflammatory drugs?
How would you describe your pain (gnawing, throbbing, deep)?

Clinical diagnosis of PsA

A simple mnemonic, ‘PSA,’ representing pain (in the joints), stiffness (>30 min after inactivity/sausage digit [dactylitis]), and axial (axial spine involvement/back pain associated with stiffness and pain that improves with activity), has been developed as a convenient method to help practitioners quickly recognize specific characteristics of PsA (9). Enthesitis may present before symptoms of arthritis in individuals with PsA, and early enthesitis is not detectable by radiography. Thus, because enthesitis may be the only presenting musculoskeletal symptom of PsA, it is important for dermatologists to query patients regarding enthesitis symptoms. During clinical examinations, dermatologists can screen for visible signs of enthesitis such as redness and swelling at insertion sites (33). Although enthesitis is typically evaluated by manual palpation, it is particularly challenging to assess because objective signs of inflammation such as redness and swelling are often absent and symptoms of enthesitis are known to mimic those of fibromyalgia, which is also associated with pain on digital palpation at tender points (34–37). Evaluation of enthesitis is a challenge amongst dermatologists and rheumatologists due to these issues and additional education on techniques for evaluating enthesitis would be beneficial. We suggest to first ask about pain at the more commonly affected entheses (lateral epicondyle, medial femoral condyle, and Achilles tendon insertion) and then palpate and look for signs of inflammation (Figure 1). Somatic symptoms and tender point counts can help differentiate PsA from fibromyalgia, in which patients have more tender points and greater enthesitis scores (37). Detecting diffuse tenderness in soft tissue areas away from joints and entheses is also helpful in differentiating fibromyalgia from PsA. Patients with reactive arthritis can also exhibit enthesitis that commonly involves the Achilles tendon and plantar fascia (38). Other factors that can help to differentiate reactive arthritis from PsA include younger age at onset, sacroiliitis, and articular symptoms primarily affecting the lower limbs and the presence of acute anterior uveitis, dysuria, or diarrhea/inflammatory bowel disease (38). Being cognizant of these differences, a diagnosis of reactive arthritis or PsA requires rheumatologic evaluation and the dermatologist will have benefited the patient greatly even if the diagnosis is reactive arthritis rather than PSA.

Figure 1. Enthesitis sites evaluated by clinical assessment instruments. LEI: Leeds Enthesitis Index; MASES: Maastricht Ankylosing Spondylitis Enthesitis Score; SPARCC: Spondyloarthritis Research Consortium of Canada. *For scoring purposes, the inferior patella and tibial tuberosity are considered one site because of their anatomical proximity. Image from Mease 2017 (178).

Although PsA is a type of inflammatory arthritis, it may be challenging to identify, because its symptoms may mimic non-inflammatory arthritis. More specifically, in patients with PsA, the erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) level are within the normal range nearly 50% of the time, joints are affected asymmetrically, DIP joint involvement may be present, and radiographic manifestations include new bone formation and concurrent bone loss as opposed to only bony erosion (13,31,32,39,40). Characteristics that can help dermatologists differentiate PsA from non-inflammatory arthritis (osteoarthritis), inflammatory arthritis (gout and rheumatoid arthritis [RA]), and pain (fibromyalgia) are described in Table 2. These characteristics, however, can be challenging to differentiate when two or more diseases are present simultaneously, and osteoarthritis and/or fibromyalgia can present with PsA (41,42). Practitioners should also be aware that the risk of gout is significantly increased in patients with concurrent PsA and psoriasis (43).

Table 2. Differential diagnosis of PsA.

Disease	Chronic or acute	Inflammation	Pattern of joint involvement	Symmetrical	Axial involvement	Key radiographic findings	Laboratory findings	Extra-articular manifestations	Female-to- male ratio
RA (145)	Chronic	Yes	Small and large joints	Yes	Cervical	Bone erosion, soft tissue swelling, joint space narrowing, marginal erosions	75–80% RF+ 90% ACPA+	Subcutaneous nodules, carpal tunnel syndrome	3:1 to 4:1
Osteoarthritis (146)	Chronic	No	Lower extremity, proximal and distal interphalangeal joints, first carpometacarpal joint	Variably	Cervical and lumbar	Joint space narrowing, sclerosis, osteophytosis	Normal RF and ACPA	None	1:1 to 2:1
Fibromyalgia (147)	Chronic	No	Diffuse	Yes	Yes	None	Normal RF and ACPA	Myalgias, tender points, irritable bowel syndrome	9:1
PsA (64,65)	Chronic	Yes	Large and small	Variably	Yes and no	Bone erosion and new bone growth, enthesitis	13% RF+ 13% ACPA+	Psoriasis, dactylitis, tendonitis, onychodystrophy	1:1
Gout (43,148,149)	Both	Yes	Classically metatarsophalangeal/podagra may involve joints of any size Increased risk in patients with concurrent psoriasis and PSA	No	Atypical	Tophi, bone erosion	Normal RF and ACPA	Rarely: dermatologic manifestation secondary to tophi eruption, eye involvement, and manifestations mimicking a tumor or infection	1:5 to 1:10*

ACPA, anti-citrullinated peptide antibodies; PsA, psoriatic arthritis; RA, rheumatoid arthritis; RF, rheumatoid factor.

*In premenopausal women.

Reproduced with modification from Mies Richie (150).

To address the need for early screening tools to identify PsA in patients with psoriasis, three validated screening tools exist: the psoriasis and epidemiology screening tool (PEST), the psoriatic arthritis screening and evaluation (PASE), and the Toronto psoriatic arthritis screening (ToPAS) (44–48). PEST is a patient self-assessment tool consisting of five questions and a drawing of a mannequin diagram on which patients indicate locations of stiff, swollen, or painful joints (49,50). PASE is a questionnaire of 15 items designed specifically for use in dermatology clinics to identify patients with psoriasis and symptoms of inflammatory musculoskeletal disease (44,51,52). ToPAS is a 12-item questionnaire in which patients use pictures of psoriatic skin, nail disease, inflamed joints, and dactylitis to identify symptoms (49). In a comparison of these tools, PEST, the only nonproprietary screening tool and available on the free GRAPPA mobile phone app, and ToPAS had slightly better sensitivity than PASE in detecting PsA in patients with psoriasis; however, all questionnaires had high false-positive rates, resulting in low specificity (53).

Although screening tools can be useful in some circumstances, they are often inconvenient and challenging to include during routine office visits with patients (9). Additionally, it is important for dermatologists to first recognize the presence of inflammatory arthritis before attempting differential diagnosis for PsA so that other causes of joint pain are not overlooked.

The ClASsification criteria for Psoriatic ARthritis (CASPAR) criteria, which were developed to standardize enrollment in PsA clinical trials, can help dermatologists identify patients with PsA but are not required for diagnosis (54). The diagnosis of PsA is clinical and based on patient history and physical examination and supported by imaging and laboratory evaluation. Patients can fulfill the CASPAR criteria to be diagnosed with PsA and dermatologists may find these criteria useful when evaluating patients for possible signs of PsA (55). These classification criteria consist of two groupings: the stem (or required criteria) and criteria associated with a numerical value (54,55). To fulfill the CASPAR criteria, a patient must present with at least one of the stem components and ≥3 points from the criteria listed in Table 3.

Table 3. CASPAR criteria.

Presence of inflammatory articular disease (joint, spine, or entheseal) with ≥ 3 points from the list below (54)
Characteristics (points)
Psoriasis (2) or Personal history of psoriasis (1) or Family history of psoriasis (1)
Dactylitis or a history of dactylitis recorded by a rheumatologist (1)
Juxta-articular new bone formation (1)
Negative for the rheumatoid factor (1)
Psoriatic nail dystrophy (1)

The stem criteria of CASPAR are defined under the broad term ‘inflammatory articular disease,’ which consists of inflammatory joint (peripheral) disease, enthesitis, and inflammatory axial disease. Inflammatory joint disease is generally recognized by joint swelling, erythema, and pain (even at rest) (55).

As part of the clinical examination for peripheral arthritis (part of the stem criteria of CASPAR), dermatologists should evaluate patients for signs of swelling, which is indicative of active synovitis, and tenderness, which is another sign of inflammation. Practical assessments can be conducted by applying light pressure (e.g. enough to blanch the tip of a fingernail) at the joint line. Key joints to evaluate include DIP joints, proximal interphalangeal joints, metacarpophalangeal and metatarsophalangeal joints, and wrist, elbow, shoulder, sternoclavicular, temporomandibular, hip, knee, ankle, and midtarsal joints (56). Novel educational tools are being developed to help train dermatologists on how to accurately perform a musculoskeletal examination to screen for PsA.

Axial inflammatory articular disease (part of the stem criteria of CASPAR) is typically characterized by slow-developing back pain persisting for >3 months before age 40 years, alternating buttock pain (caused by sacroiliitis), pain causing waking from sleep during the second half of the night and prolonged morning stiffness or stiffness upon immobility (57,58). Additionally, treatments that are effective for other PsA domains may not improve axial disease.

The numerically valued criteria of CASPAR include psoriasis or a personal/family history of psoriasis, psoriatic nail dystrophy, a negative test result for the presence of rheumatoid factor (RF), current dactylitis or history of dactylitis (recorded by a rheumatologist), and juxta-articular new-bone formation (54).

Among the CASPAR criteria, ongoing psoriasis (which is weighted more than all other criteria) or a personal or family history of psoriasis is encompassed within one category, whereas nail dystrophy is counted as a separate category. Of the remaining criteria, dactylitis requires the most careful clinical attention, because it may occur in any digit, but most frequently manifests in the toes (59). A dactylitic digit, which is very specific to PsA, may be red, hot, and tender (acute dactylitis) or swollen without acute inflammatory changes (chronic dactylitis) and may occur in isolation or as one of several digits affected (6,59,60). During clinical examinations, dermatologists can identify dactylitis by screening for ‘sausage-shaped’ digits, which present with uniform inflammation such that soft tissue between the metacarpophalangeal and proximal interphalangeal, proximal and DIP, and/or DIP and digital tuft are diffusely swollen (61). The entire digit is affected and swelling at each joint cannot be independently recognized, and flexion is difficult or impossible (62). The presence of RF, a type of autoantibody first found to be associated with RA and not typically present in PsA, is determined through a blood test (63). Although negative RF is one of the numerically valued CASPAR criteria, RF is present in approximately 13% of patients with PsA, and the incidence of positive RF increases with age (21,64). Anti-cyclic citrullinated peptide antibodies are an additional negative marker of PsA (not included in the CASPAR criteria); however, roughly 13% of patients with PsA present with these antibodies (65). When positive in patients with PsA, both RF and anti-cyclic citrullinated peptide antibodies typically have a low titer, especially in comparison to patients with RA (66,67). Finally, the last numerically valued CASPAR criteria, juxta-articular new bone formation, is not typically a feature of early PsA.

Psoriasis subtypes and PsA

Although psoriasis can occur in a variety of locations on the body, specific manifestations are clinical predictors of PsA (68). Several studies have found an increased risk of PsA has correlations with psoriatic scalp lesions, intertriginous/inverse psoriasis, and nail dystrophy (68–71). The hazard ratios for PsA in patients with scalp, intergluteal/perianal, and nail psoriasis are 3.89, 2.35, and 2.93, respectively (68), and in mild psoriasis, 83% of patients with scalp and nail psoriasis, 40% of patients with intergluteal/perianal lesions, and 37% of patients with isolated scalp psoriasis met CASPAR criteria for PsA (69). Similarly, the incidences of scalp, intergluteal, and nail psoriasis were higher in patients with PsA (100, 83, and 64%, respectively) than in patients with psoriasis alone (67, 25, and 40%, respectively) (70). Recognition of inverse psoriasis is an important clue for informing a diagnosis of PsA that is commonly missed during physical examination. Inverse psoriasis has typically been considered uncommon but contemporary findings report a prevalence of 21–30% for inverse psoriasis in patients with psoriasis (72).

Although nail psoriasis is common in moderate-to-severe psoriasis without PsA, it is even more prevalent in PsA (over 80% of patients affected) and is hypothesized to indicate involvement of the distal phalanx in PsA (55,68,73). This unique link between nail dystrophy and PsA can be explained by the physiological relationship between nails and entheses. Histological studies have demonstrated that the extensor tendon, which is attached to the terminal phalanx, extends distally and connects with the nail root, revealing that the fascia of the nail root is an extension of the enthesis and providing evidence of how inflammation of these entheses often gives rise to nail pitting (Figure 2) (74). The link between nail disease and adjacent enthesitis has further been demonstrated through magnetic resonance imaging (MRI) and ultrasound studies (75,76). Therefore, although nails are developmentally related to the skin, they are functionally linked with entheses, underscoring the correlation between nail disease/osteolysis and PsA – particularly in DIP joints (77,78).

Figure 2. The fascia of the nail root is an extension of the entheses. Image from McGonagle 2009 (179).

Overall, these observations indicate that dermatologists should pay particular attention for PsA manifestations in patients with psoriasis affecting the scalp, intergluteal/perianal areas, and nails (79).

Treatment of PsA

The European League Against Rheumatism (EULAR) and the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) have developed recommendations for the treatment of PsA, with guidelines for appropriate use of nonsteroidal anti-inflammatory drugs (NSAIDs), traditional disease-modifying anti-rheumatic drugs (DMARDs; e.g. methotrexate, sulfasalazine, leflunomide, cyclosporine), biologics, and small-molecule inhibitors (80,81). These recommendations provide specific guidance for treatment by PsA clinical domain. Choice of therapy should be optimized to target the PsA symptoms that are the most burdensome to each patient, with the overarching goals of maximizing long-term health-related quality of life and preventing irreversible joint structural damage (80,81). Briefly, NSAIDs are recommended to control pain and inflammation in patients with peripheral arthritis, axial disease, and enthesitis, and traditional DMARDs are recommended as treatments for peripheral arthritis, dactylitis, and skin/nail disease. Guidelines clarify that the term ‘DMARD’ to describe these agents was chosen for historical purposes and that these agents do not provide any disease-modifying effects on radiographic damage. Biologics that inhibit TNF-α, IL-17A, or IL-12/23 are recommended across all PsA domains, and the small-molecule phosphodiesterase-4 inhibitor, apremilast, is recommended for all domains except axial disease (80,81). These guidelines were developed in 2015, when data were available for the TNF-α inhibitors, etanercept, adalimumab, infliximab, certolizumab pegol, and golimumab, and for the IL-12/23 inhibitor, ustekinumab (80,81). Since then, additional data have been published for apremilast, the IL-17A inhibitors secukinumab and ixekizumab, the Janus kinase inhibitor tofacitinib, and the T-cell inhibitor abatacept, all of which are now approved for adults with active PsA. The IL-23 inhibitor guselkumab is approved for moderate-to-severe plaque psoriasis and is in late-stage development for PsA. Additionally, NSAIDs, adalimumab, certolizumab pegol, etanercept, golimumab, infliximab, and secukinumab are approved for the treatment of ankylosing spondylitis (AS), a condition related to PsA.

Domain-specific evaluation of PsA in clinical trials

Dermatologists should be particularly aware of measures used to assess PsA, as the International Dermatology Outcome Measures (IDEOM) group has recently encouraged screening and subsequent measurement of PsA symptoms in all psoriasis clinical trials (82). The IDEOM group evaluated the patient global (PG)-arthritis, routine assessment patient index data (RAPID)-3, and Psoriatic Arthritis Impact of Disease (PsAID)-9 tools for PsA symptom evaluation and recommended that PsAID9 and RAPID3 are potentially more appropriate measures of PsA symptoms than PG-arthritis (82). In PsA clinical trials, various measures (Table 4) are used to assess peripheral arthritis, dactylitis, enthesitis, skin/nail disease, and radiographic outcomes. Additionally, several different instruments are used to evaluate patient health status (patient-reported outcomes [PROs]), and several individual clinical and PRO measures have been integrated into composite indices to create more comprehensive measures.

Table 4. Indices used to measure PsA activity by clinical domain (56,89,96–99,121,151–158).

PsA domain	Measure	Description/guidance for practical use	Advantages/disadvantages
Peripheral arthritis	66 SJC, 68 TJC	Specific joints are palpated for signs of tenderness and swelling, which are signs of inflammation and active synovitis. To perform, apply ∼4 kg/cm^2 of pressure at the joint line	Easy to perform; simple scoring Can be completed in ∼2 min Good responsiveness in PsA clinical trials Severity of tenderness and swelling is not graded
	DAS28	A composite joint count measure that includes joints of both shoulders, elbows, knees, and wrists, as well as all MCP joints, and all PIP joints of the hands	Good performance characteristics in PsA Not recommended for use in PsA inclusion criteria or as a primary endpoint in clinical trials because it may underestimate disease in lower extremities and DIP joints Severity of tenderness and swelling is not graded Does not evaluate foot involvement, which is common in PsA
	ACR	The ACR joint count documents the number of joints with tenderness and swelling	Reliable measure of peripheral disease activity in PsA clinical trials Can be used for patient follow-up in clinical practice
Dactylitis	0–3 score assigned to each digit of the hands and feet	Overall scores range from 0 (none) to 60 (severe)	Simple counting and scoring method Demonstrated responsiveness in PsA trials of biologics Not as quantitative as LDI
	LDI	Uses an instrument called a dactylometer to measure digit circumference and evaluate pain and tenderness on a 0–3 scale; size comparisons are made vs. contralateral digits	Minimal burden for patients Objective quantitative measure Requires a specialized tool (dactylometer) More complicated scoring system that involves comparison of the ratio of circumferences between affected and contralateral digits
Enthesitis	4-point enthesitis measure	Enthesitis at Achilles tendon and plantar fascia graded as present or absent or scored on a 0–3 scale	Demonstrated responsiveness in PsA clinical trials Does not perform as well as indices that evaluate more sites
	LEI	Bilateral evaluation of six entheseal sites (lateral epicondyles, medial femoral condyles, and Achilles tendon insertions) for the presence (1) or absence (0) of tenderness	Developed specifically for PsA Can be completed in ∼30 s Correlates well with other PsA disease activity measures Least floor effect of available enthesitis indices
	MASES	Evaluation of 13 entheseal sites (bilateral first costochondral joints, seventh costochondral joints, posterior superior iliac spines, anterior superior iliac spines, iliac crests, proximal insertion of Achilles tendons, and fifth lumbar spinous process) for the presence (1) or absence (0) of tenderness	Recommended by ASAS for use in SpA randomized controlled trials Can be completed in ∼2–5 min Demonstrated responsiveness in PsA clinical trials Not validated in PsA Lower ICC in patients with PsA vs. AS
	SPARCC	Bilateral evaluation of 16 entheseal sites (Achilles tendons, plantar fascia insertion at the calcaneus, patellar tendon insertion at base of the patella, quadriceps insertion into the superior border of the patella, supraspinatus insertion into the greater tuberosity of the humerus, and medial and lateral epicondyles) for the presence (1) or absence (0) of tenderness. Modified versions are available evaluating 6 and 8 sites	Can be completed in ∼2–5 min Correlates well with AS disease activity measures Not validated in PsA
	Berlin	Evaluation of 12 entheseal sites scored as absent (0) or present (1)	Has not been used in PsA clinical trials Lower ICC in patients with PsA vs. AS
	San Francisco	Evaluation of 17 tendon-insertion sites scored based on pain and tenderness on a scale of 0–3	More sensitive than Berlin index and MASES in AS clinical trial Has not been used in PsA clinical trials Lower ICC in patients with PsA vs. AS
Skin	PASI	Quantitative assessment of skin lesions based on BSA affected and severity of erythema, induration, and scale, weighted by body area	Widely used in psoriasis and PsA clinical trials Can be administered in ∼5 min Not used in clinical practice, in part because of complexity
	PGA	Assessment of skin lesions scored on a scale from 0 (clear) to 6 (very severe)	Simpler to use than PASI Widely recognized and accepted by dermatologists Less quantitative than PASI
	BSA	Assessed by considering the surface area of skin affected by psoriasis lesions; patient’s handprint (palm and fingers) is estimated to be 1% BSA	Basic, easy-to-perform assessment that can be used in daily practice Does not include assessment of lesion severity or body location affected
Nails	NAPSI	Nails are scored based on the presence or absence of nail bed and nail matrix psoriasis, with each nail divided into four quadrants (scores for each nail range from 0 to 8)	Has been used extensively in psoriasis clinical trials Relatively straightforward physical examination Can be completed in ∼5–10 min Not validated in psoriasis or PsA Not used in clinical practice
	Modified NAPSI	Each nail is evaluated for seven features: pitting, onycholysis and oil-dropping dyschromia, nail plate crumbling, leukonychia, splinter hemorrhages, hyperkeratosis, and red spots in the lunula	Shorter and easier to use than NAPSI Good inter-rater reliability Used in PsA clinical trials Well correlated with other clinical measures of PsA Relatively straightforward physical examination Can be completed in <5 min
Radiographic progression	Modified Steinbrocker global scoring method	40 joints in the hands and feet are classified on a 0–4 radiographic scale in which 0 = normal, 1 = juxta-articular osteoporosis or soft-tissue swelling, 2 = erosions, 3 = erosion and joint-space narrowing or subluxation, and 4 = total joint destruction (lysis or ankylosis)	Simple scoring method used since 1949 Can be performed quickly in clinical practice (∼6 min) Does not include use of any radiographic standards Less sensitive than other methods
	Sharp method for PsA	Evaluates joints in the hands (40–44 joints) and feet (8–12 joints) for erosions and joint-space narrowing on a scale of 0–5; separate grading is used for pencil-in-cup phenomenon, gross osteolysis, periostitis, and tuft resorption	Various Sharp score modifications have been used in different PsA clinical trials More sensitive to detecting change in RA than PsA Time-consuming and not practical for use in a clinical setting
	Sharp/van der Heijde method for PsA	Erosions are scored from 0 to 5 and joint-space narrowing from 0 to 4, with pencil-in-cup and osteolysis scored separately	Has been widely used in PsA clinical trials Can be performed in ∼15 min Changes > MCID have not been observed in PsA clinical trials because disease progression is slow relative to study durations
	PARS	Assessment of 40 joints in the hands and feet scored separately for destruction (0–5 scale) and bony proliferation (0–4 scale) for a total score range of 0–360	Validated in PsA Only instrument to evaluate proliferation, which is a PsA-specific feature Can be performed in ∼10 min Destruction and proliferation scales are only weakly correlated Has not been used in any clinical trials
Spine assessment	BASDAI	Set of 6 questionnaires scored on a 0–10 cm VAS about symptoms of fatigue, pain, and stiffness	Most commonly used measure in AS clinical trials Can be completed in ≤2 min Does not discriminate well between peripheral and axial PsA disease activity Fails to discriminate between high and low PsA disease activity
	BASFI	Set of 10 items related to functional anatomy (bending, reaching, changing position, standing, turning, and climbing steps) and ability to cope with everyday life rated on a 0–10 cm VAS	Commonly used in AS disease impact studies Can be completed in <3 min Does not discriminate well between peripheral and axial disease activity

ACR: American College of Rheumatology; AS: ankylosing spondylitis; ASAS: Assessment of SpondyloArthritis international Society; BASDAI: Bath Ankylosing Spondylitis Disease Activity Index; BASFI: Bath Ankylosing Spondylitis Function Index; BSA: body surface area; DAS28: Disease Activity Score in 28 joints; DIP: distal interphalangeal; ICC: intraclass correlation coefficient; LDI: Leeds Dactylitis Index; LEI: Leeds Enthesitis Index; MASES: Maastricht Ankylosing Spondylitis Enthesitis Score; MCID: minimal clinically important difference; MCP: metacarpophalangeal; NA: not available; NAPSI: nail psoriasis severity index; PARS: Psoriatic Arthritis Ratingen score; PASI: psoriasis area and severity index; PGA: physician global assessment; PIP: proximal interphalangeal; PsA: psoriatic arthritis; RA: rheumatoid arthritis; SDD: smallest detectable difference; SJC: swollen joint count; SpA: spondyloarthritis; SPARCC: spondyloarthritis research consortium of Canada; TJC: tender joint count; VAS: visual analog scale.

Peripheral arthritis measures

Peripheral arthritis is predominantly assessed by indices, including the 68 tender and 66 swollen joint count, the 28-joint Disease Activity Score (DAS28), a composite measure that assesses 28 peripheral joints along with a measure of ESR or CRP (in the DAS28-CRP measure), and the American College of Rheumatology (ACR) criteria, another composite measure that specifies a percentage improvement (i.e. 20, 50, or 70%) in tender or swollen joint counts, as well as 3 of 5 additional measures (Table 4) (56,83). These measures include the patient’s assessment of pain, the patient’s global assessment of disease activity, the physician’s assessment of physical function, the patient’s assessment of physical function, and acute-phase reactant value. Furthermore, when trials last longer than 1 year and agents are being tested as DMARDs, ACR includes radiography or another imaging technique (83,84). Across PsA clinical trials, 20% improvement in ACR criteria (ACR20) has been used as a primary endpoint to assess the efficacy of biologics and small-molecule inhibitors for the treatment of articular symptoms. Results from these trials are summarized in Figure 3. However, because DAS28 was originally developed for RA, it does not evaluate some of the most frequently involved joints in PsA (particularly in the lower extremities) and all PsA domains. Although ACR response is used as the primary endpoint in PsA clinical trials, it is not employed in clinical practice and does not evaluate all PsA domains.

Figure 3. ACR20 response rates reported at week 24 in PsA trials of biologics and small-molecule inhibitors (86,102,103,132,134–136,138,139,159–161,180). For agents without a dose specified, results are presented for the approved dosage. Figure is for visualization purposes only and should not be used to make direct comparisons of efficacy between therapies. Data values have not been statistically corrected. Unless otherwise indicated, the proportion of patients with ACR20 at week 24 was the primary endpoint of the study. ACR: American College of Rheumatology; BID: two times daily; BIW: twice weekly; PsA: psoriatic arthritis; Q2W: every 2 weeks; Q4W; every 4 weeks; QW: once weekly. *Primary endpoint: ACR20 at week 12; ^†Primary endpoint: ACR20 at week 16; ^‡ACR20 at week 24 is a secondary endpoint; ^§Primary endpoint: ACR20 at week 14; ^‖‖Results reported at week 16 (primary endpoint).

Dactylitis measures

Dactylitis is also measured using multiple approaches (Table 4). The simplest assessment is counting dactylitic digits (tender and non-tender or just tender) (85). Alternative clinical indices include a 0–3 scale of physician-rated severity to assess all 20 digits (86), the Leeds Dactylitis Index (LDI), and LDI-Basic (87,88). The LDI and LDI-Basic multiply the tenderness score of each affected digit (based on the Ritchie index – graded 0–3 for LDI or binary score for LDI-Basic) by the ratio of the circumference of the affected digit to the circumference of the digit on the opposite hand or foot. Results from clinical trials using these indices to evaluate the effects of treatment on dactylitis are summarized in Table 5. Drugs that have shown statistically significant improvements in dactylitis indices compared with placebo include infliximab, certolizumab pegol, intravenous golimumab, subcutaneous golimumab 100 mg, ustekinumab, guselkumab, secukinumab, and ixekizumab. Statistical significance was not reached or was not tested for in trials of adalimumab, etanercept, tofacitinib, apremilast, and abatacept. It is not possible to indirectly compare results across trials because different clinical indices and different reporting methods were used to assess dactylitis across studies. At present, there is no consensus on the best method for assessing dactylitis in PsA clinical trials, and minimal clinically important difference (MCID) thresholds have not been established for available dactylitis indices.

Table 5. Results of dactylitis assessments by clinical index from trials of biologics and small-molecule inhibitors in PsA.

Dactylitis index	Trial name/description	Key dactylitis outcomes
Scale of 0–3 for each digit of the hands and feet	ADEPT (Adalimumab Effectiveness in Psoriatic Arthritis Trial) (159)	Mean improvement in dactylitis was greater with adalimumab vs. placebo but statistical significance was not reached (values not reported)
	PRESTA (Psoriasis Randomized Etanercept STudy in Subjects with Psoriatic Arthritis) (160)	Mean percent change from baseline to week 24 was 85% with etanercept 50 mg BIW/QW and 85% with etanercept 50 mg QW/QW
	IMPACT (Infliximab Multinational Psoriatic Arthritis Controlled Trial) (86)	At week 16, mean percent improvement was 85% in the infliximab group vs. 29% in the placebo group (p < .001)
	GO-REVEAL (A Randomized Evaluation of Safety and Efficacy in Subjects with Psoriatic Arthritis Using a Human Anti-TNF Monoclonal Antibody) (161)	Median percent change from baseline to week 24 was 100% with golimumab 50 mg, 100% with golimumab 100 mg, and 42% with placebo (both p ≤ .09 vs. placebo)
	GO-VIBRANT (A study of golimumab in participants with active PsA) (102)	Mean change from baseline to week 24 in dactylitis score was –8.2 with IV golimumab 2 mg/kg and –5.0 with placebo (p < .001)
	PSUMMIT 1 (A study of the safety and effectiveness of ustekinumab in patients with PsA) (138)	Among patients with dactylitis at baseline, dactylitis was present at week 24 in 56% of patients treated with ustekinumab 45 or 90 mg vs. 76% of the placebo group (p=.0013)
	OPAL Beyond (Oral Psoriatic Arthritis Trial; Tofacitinib in PsA subjects with inadequate response to TNF inhibitors) (134)	Mean dactylitis score change from baseline to month 3 was –5.2 with tofacitinib 5 mg, –5.4 with tofacitinib 10 mg, and –1.9 with placebo (statistical significance not tested)
	Efficacy and safety of guselkumab in patients with active PsA: a phase 2a, randomized, double-blind, placebo-controlled study (132)	Median percent improvement from baseline to week 24 was 100% with guselkumab 100 mg and 33% with placebo (p < .001)
Presence (1) or absence (0) for each digit of the hands and feet	FUTURE 1 (Efficacy at 24 weeks and long-term safety, tolerability and efficacy up to 2 years of secukinumab (AIN457) in patients with active PsA (133)	At week 24, complete resolution of dactylitis was achieved by 52% of patients receiving secukinumab 75 or 150 mg compared with 16% of the placebo group (p < .05)
	PALACE 1 (Psoriatic Arthritis Long-term Assessment of Clinical Efficacy 1) (135)	The percentage of patients with complete resolution of dactylitis at week 24 was 51% with apremilast 20 mg BID, 48% with apremilast 30 mg BID, and 41% with placebo (statistical significance not reached)
LDI or LDI-Basic	RAPID-PsA (Certolizumab pegol in subjects with adult onset active and progressive PsA) (139)	Mean LDI change at week 24 was –40.7 with certolizumab pegol 200 mg Q2W, –53.5 with certolizumab pegol 400 mg Q4W, and –22.0 with placebo (both p ≤ .002)
	SPIRIT-P1 (A study of ixekizumab in participants with active PsA) (136)	Mean change from baseline in LDI-Basic was –75.4 with ixekizumab Q4W, –66.1 with ixekizumab Q2W, and –33.7 with placebo (both p < .05 vs. placebo). Complete resolution of dactylitis (LDI-Basic score = 0) was reported by 80% of the ixekizumab Q4W group, 77% in the ixekizumab Q2W group, and 25% of the placebo group (both p ≤ .001 vs. placebo)
	ASTRAEA (Active PSoriaTic Arthritis RandomizEd TriAl) (103)	Using LDI-Basic, complete resolution of dactylitis at week 24 was achieved by 44% of patients treated with abatacept vs. 34% with placebo; statistical significance was not reached

BID: two times daily; IV: intravenous; LDI: Leeds Dactylitis Index; PsA: psoriatic arthritis; Q2W: every 2 weeks; Q4W: every 4 weeks; TNF: tumor necrosis factor.

Enthesitis measures

Enthesitis can be measured with imaging techniques, such as ultrasound and MRI; however, in dermatology clinics, use of domain-specific clinical indices is more practical. Instruments available for the assessment of enthesitis include the Leeds Enthesitis Index (LEI), the Maastricht Ankylosing Spondylitis Enthesitis Score (MASES), the 4-point enthesitis measure, the Spondyloarthritis Research Consortium of Canada (SPARCC) enthesitis measure, the Berlin index, and the San Francisco index (Table 4) (56).

As with dactylitis, there is currently no consensus on which enthesitis index is best; therefore, PsA trials of biologics and small-molecule inhibitors have used different instruments. As shown in Table 6, even when the same instrument was used in different studies, differences in reporting methods and study designs restrict the ability to indirectly compare results across studies. Furthermore, the clinical significance of improvements on different indices is not well established, and MCID is not available for any enthesitis measure (89,90). However, it is noteworthy that statistically significant improvements in enthesitis have been reported compared with placebo in phase 3 studies of infliximab, certolizumab pegol, golimumab (subcutaneous and intravenous formulations), secukinumab, ustekinumab, and apremilast, and in a phase 2 study of guselkumab (Table 6).

Table 6. Results of enthesitis assessments by clinical index from trials of biologics and small-molecule inhibitors in PsA.

Enthesitis index	Trial name/description	Key enthesitis outcomes
4-point scale	ADEPT (Adalimumab Effectiveness in Psoriatic Arthritis Trial) (159)	Mean improvement in enthesitis was greater with adalimumab vs placebo, but statistical significance was not reached (values not reported)
	PRESTA (Psoriasis Randomized Etanercept STudy in Subjects with Psoriatic Arthritis) (160)	81% of patients treated with etanercept 50 mg BIW/QW, and 81% of patients treated with etanercept 50 mg QW/QW had a reduction in enthesitis sites at week 24
	IMPACT (Infliximab Multinational Psoriatic Arthritis Controlled Trial) (86)	At week 16, 14% of patients treated with infliximab compared with 31% of the placebo group had enthesitis (p < .001)
	FUTURE 1 (Efficacy at 24 weeks and long term safety, tolerability and efficacy up to 2 years of secukinumab (AIN457) in patients with active PsA (133)	At week 24, complete resolution of enthesitis was achieved by 48% of patients treated with secukinumab 75 or 150 mg compared with 13% of the placebo group (p < .05)
LEI	RAPID-PsA (Certolizumab pegol in subjects with adult onset active and progressive PsA) (139)	In patients with LEI >1 at baseline, mean change at week 24 was –2.0 with certolizumab pegol 200 mg Q2W, –1.8 with certolizumab pegol 400 mg Q4W, and –1.1 with placebo (both p ≤ .003)
	GO-VIBRANT (A study of golimumab in participants with active PsA) (102)	Mean change from baseline to week 14 in LEI was –2.1 with IV golimumab 2 mg/kg and –1.1 with placebo (p < .001)
	SPIRIT-P1 (A study of ixekizumab in participants with active PsA) (136)	Mean LEI score change from baseline to week 24 was −1.3 for IXEQ4W, −1.4 for IXEQ2W, and −0.8 for placebo (statistical significance not reached)
	OPAL Beyond (Oral Psoriatic Arthritis Trial; Tofacitinib in PsA subjects with inadequate response to TNF inhibitors) (134)	Mean LEI score change from baseline to month 3 was –1.3 with tofacitinib 5 mg, –1.3 with tofacitinib 10 mg, –0.5 with placebo (statistical significance not tested)
	ASTRAEA (Active PSoriaTic Arthritis RandomizEd TriAl) (103)	Complete resolution of enthesitis at week 24 was achieved by 33% of patients treated with abatacept vs. 21% with placebo (statistical significance not reached)
	Efficacy and safety of guselkumab in patients with active PsA: a phase 2a, randomized, double-blind, placebo-controlled study (132)	Median percent improvement from baseline to week 24 was 100% with guselkumab 100 mg and 33% with placebo (p = .009)
MASES or modified MASES	GO-REVEAL (A Randomized Evaluation of Safety and Efficacy in Subjects with Psoriatic Arthritis Using a Human Anti-TNF Monoclonal Antibody) (161)	Median percent change on modified MASES from baseline to week 24 was 60% with golimumab 50 mg, 67% with golimumab 100 mg, and 12% with placebo (both p < .001 vs. placebo)
	PSUMMIT 1 (A study of the safety and effectiveness of ustekinumab in patients with PsA) (138)	Among patients with modified MASES ≥1 at baseline, enthesitis was present at week 24 in 65% of patients treated with ustekinumab 45 or 90 mg vs. 81% of the placebo group (p = .0006)
	PALACE 1 (Psoriatic Arthritis Long-term Assessment of Clinical Efficacy 1) (135)	The percentage of patients with complete resolution of enthesitis at week 24 was 32% with apremilast 20 mg BID, 34% with apremilast 30 mg BID, and 14% with placebo; both p ≤ .0037 vs. placebo

BID: two times daily; BIW: twice weekly; IV: intravenous; IXEQ2W: ixekizumab 80 mg once every 2 weeks; IXEQ4W: ixekizumab 80 mg once every 4 weeks; LEI: Leeds Enthesitis Index; MASES: Maastricht Ankylosing Spondylitis Enthesitis Score; PsA: psoriatic arthritis; Q2W: every 2 weeks; Q4W: every 4 weeks; QW: once weekly; TNF: tumor necrosis factor.

Skin and nail psoriasis measures

Psoriasis is measured by tools frequently used in dermatology trials, including the psoriasis area and severity index (PASI), body surface area (BSA), physicians global assessment (PGA), Investigator’s Global Assessment (IGA), or IGA modified 2011 (Table 4) (56,91). The primary endpoint in psoriasis trials and the key skin endpoint in PsA trials are PASI responses (usually percentage of patients achieving at least 75% improvement in PASI scores [PASI75]). Nail psoriasis is assessed using the nail psoriasis severity index (NAPSI) or modified NAPSI (Table 4) (56).

The National Psoriasis Foundation recently defined acceptable response as either BSA ≤3% or ≥75% BSA improvement from baseline and target response as BSA ≤1% for patients with psoriasis (92). Additionally, the product of PGA and BSA (PGAxBSA) is a simple tool for assessing psoriasis severity and response to treatment with concordance to PASI response that can evaluate treatment targets for psoriasis (93–95). As use of newer biologics and small-molecule inhibitors will be needed to meet these goals, it is critical that dermatologists understand the impact that these drugs have on PsA and other domains of psoriatic disease so that patients are appropriately treated.

Treatment of skin disease by dermatologists is an important component of comprehensive PsA disease management. While PsA clinical trials have included skin assessments, results from psoriasis trials can provide the most specific and focused information on the effects of different therapies on clearance of skin and nail lesions. Table 7 summarizes PASI and NAPSI response rates for biologic and small-molecule inhibitor therapies approved for the treatment of moderate-to-severe plaque psoriasis, including etanercept, infliximab, adalimumab, ustekinumab, guselkumab, secukinumab, ixekizumab, and apremilast.

Table 7. Results of psoriasis assessments from trials of biologics and small-molecule inhibitors in chronic plaque psoriasis.

	Skin outcomes		Nail outcomes
Drug	Trial	PASI result	Trial	NAPSI result
Etanercept	Etanercept as monotherapy in patients with psoriasis (162)	Week 12 PASI 75 49% with 50 mg BIW (p < .001 vs. placebo) 34% with 25 mg BIW (p < .001 vs. placebo) 14% with 25 mg QW (p < .001 vs. placebo) 4% with placebo Week 12 PASI 90 22% with 50 mg BIW (p < .001 vs. placebo) 12% with 25 mg BIW (p < .001 vs. placebo) 3% with 25 mg QW 1% with placebo	Open-label trial of the effects of etanercept 50 mg BIW/QW or QW/QW on nail psoriasis in patients with moderate-to-severe plaque psoriasis (163)	Mean target fingernail NAPSI score change from baseline to week 24: –4.3 with etanercept 50 mg BIW/QW –4.4 with etanercept 100 mg QW/QW
Infliximab	EXPRESS (Infliximab induction and maintenance therapy for moderate-to-severe psoriasis) (164)	Week 10 PASI 75 achieved by 80% with infliximab vs. 3% with placebo (p < .0001) PASI 90 achieved by 57% with infliximab vs. 1% with placebo (p < .0001)	EXPRESS (164)	NAPSI % improvement from baseline to week 24: 56% with infliximab vs. –3% with placebo (p < .0001)
Adalimumab	REVEAL (Adalimumab therapy for moderate to severe psoriasis) (165)	Week 12 PASI 75 achieved by 68% with adalimumab vs. 5% with placebo (p < .001 vs. placebo) PASI 90 achieved by 37% with adalimumab vs. 2% with placebo (p < .001 vs. placebo)	Post hoc nail psoriasis subanalysis of data from the BELIEVE study in patients with moderate-to-severe psoriasis (166)	Median NAPSI response at week 16, 40%
Ustekinumab	PHOENIX 1 (Efficacy and safety of ustekinumab in patients with psoriasis) (167)	Week 12 PASI 75 66% with ustekinumab 90 mg (p < .0001 vs. placebo) 67% with ustekinumab 45 mg (p < .0001 vs. placebo) Placebo 3% Week 12 PASI 90 37% with ustekinumab 90 mg (p < .0001 vs. placebo) 42% with ustekinumab 45 mg (p < .0001 vs. placebo) Placebo 2%	PHOENIX 1 (168)	NAPSI % improvement from baseline to week 24: 47% with ustekinumab 45 mg 49% with ustekinumab 90 mg
Secukinumab	ERASURE (Efficacy or response and safety of two fixed secukinumab regimens in psoriasis) (169)	Week 12 PASI 75 82% with secukinumab 300 mg (p < .001 vs. placebo) 72% with secukinumab 150 mg (p < .001 vs. placebo) 5% with placebo Week 12 PASI 90 59% with secukinumab 300 mg (p < .001 vs. placebo) 39% with secukinumab 150 mg (p < .001 vs. placebo) 1% with placebo	TRANSFIGURE (Phase 3b study of secukinumab vs. placebo in patients with moderate-to-severe psoriasis with nail involvement (170)	Mean NAPSI % change from baseline to week 16: –45% with secukinumab 300 mg (p < .0001 vs. placebo) –38% with secukinumab 150 mg (p < .0001 vs. placebo) –11% with placebo
Ixekizumab	UNCOVER-3 (Comparison of ixekizumab with etanercept or placebo in moderate-to-severe psoriasis) (171)	Week 12 PASI 75 84% with ixekizumab Q4W (p < .0001 vs. placebo) 87% with ixekizumab Q2W (p < .0001 vs. placebo) 7% with placebo Week 12 PASI 90 65% with ixekizumab Q4W (p < .0001 vs. placebo) 68% with ixekizumab Q2W (p < .0001 vs. placebo) 3% with placebo	UNCOVER-3 (172)	Mean NAPSI % change from baseline to week 12: –39% with ixekizumab Q2W –40% with ixekizumab Q4W +5% with placebo
Apremilast	ESTEEM 1 (Efficacy and safety trial evaluating the effects of apremilast in psoriasis) (173)	Week 16 PASI 75 33% with apremilast 30 mg BID (p < .0001 vs. placebo) 5% with placebo	Subanalysis of ESTEEM 1 and ESTEEM 2 data for apremilast 30 mg BID vs. placebo in patients with moderate-to-severe psoriasis (174)	Mean NAPSI % change vs placebo at week 16: –23% vs. +6.5% in ESTEEM 1; p < .0001 –29% vs. –7.1% in ESTEEM 2; p < .0001
Guselkumab	VOYAGE 1 (Efficacy and safety of guselkumab compared with adalimumab for the continuous treatment of patients with moderate-to-severe psoriasis (175)	Week 16 PASI 75 91% with guselkumab (p < .001 vs. placebo) 6% with placebo Week 16 PASI 90 73% with guselkumab (p < .001 vs. placebo) 3% with placebo	VOYAGE 1 (175)	Mean NAPSI % change from baseline to week 16: 34% with guselkumab –1% with placebo

BIW: twice weekly; NAPSI: Nail Psoriasis Severity Index; PASI: Psoriasis Area and Severity Index; Q2W: every 2 weeks; Q4W: every 4 weeks; QW: once weekly.

Radiographic measures

As shown in Table 4, the most common radiographic scoring methods for PsA clinical trials are modified from RA scoring methods and include the modified Steinbrocker, the modified Sharp method for PsA, the modified Sharp/van der Heijde method for PsA (also reported as the van der Heijde-modified total Sharp score [mTSS]), and the Psoriatic Arthritis Ratingen Score (PARS) (96). Although these tools all analyze joints in the hands and feet, scoring criteria vary between methods. For example, both modified Sharp and mTSS separately assess erosions and joint space narrowing (a product of cartilage destruction), whereas the PARS method separately assesses erosions and bony proliferations. Further, the modified Steinbrocker uses only one scale that considers erosions, joint space narrowing, and ankylosis. GRAPPA consensus is that the Sharp/van der Heijde method for PsA is the optimal tool for use in randomized controlled trials (97).

Radiographic progression results from PsA trials are summarized in Table 8. All drugs, except abatacept, were associated with statistically significant differences compared with placebo in mTSS/modified Sharp/van der Heijde score (not reported for guselkumab, apremilast, or tofacitinib). However, differences between active treatment and placebo were below the MCID based on the smallest detectable difference observed in RA studies (98,99). In many cases, most (>80%) patients in both placebo and active-treatment groups had a change in Sharp/van der Heijde score that was below the smallest detectable change, indicating that no measurable progression was observed in the majority of patients during the study period (29,100–102). This lack of progression has made it difficult to detect meaningful treatment differences in radiographic progression (103). Of the drugs available for PsA, approval for inhibition of radiographic progression is given for adalimumab, etanercept, certolizumab pegol, infliximab, ixekizumab, secukinumab, and abatacept by both the FDA and EMA (104–117). For golimumab, ustekinumab, and tofacitinib, only the EMA gives approval for inhibition of radiographic progression (118–120). Apremilast does not have approval for inhibition of radiographic progression from either agency.

Table 8. Results of radiographic progression assessments from trials of biologics and small-molecule inhibitors in PsA.

Trial name/description	Key radiographic outcomes
Etanercept treatment of PsA: safety, efficacy, and effect on disease progression (137)	Annualized rate of change in modified TSS at 12 months was –0.03 units with etanercept vs. +1.00 unit for placebo (p = .0001)
ADEPT (Adalimumab Effectiveness in Psoriatic Arthritis Trial (131)	At week 24, mean change from baseline in modified TSS was –0.1 with adalimumab and +0.9 with placebo; changes at week 48 were +0.1 and +1.0, respectively (SDC = 1.88)
FUTURE 5 (Study to Demonstrate the Efficacy [Including Inhibition of Structural Damage], Safety and Tolerability up to 2 Years of Secukinumab in Active Psoriatic Arthritis) (176)	At week 24, mean change from baseline in modified TSS was +0.08 with secukinumab 300 mg with loading dose (p < .01 vs. placebo), +0.17 with secukinumab 150 mg with loading dose (p < .05 vs. placebo), –0.09 with secukinumab 150 mg without loading dose (p < .05 vs. placebo), and +0.50 with placebo
SPIRIT-P1 (A study of ixekizumab in participants with active PsA) (136)	At week 24, mean change from baseline in modified TSS was +0.17 with ixekizumab Q4W, +0.08 with ixekizumab Q2W, and +0.49 with placebo (both p ≤ .01)
IMPACT 2 (Infliximab Multinational Psoriatic Arthritis Controlled Trial 2) (100)	On a 0–528 scale (SDC = 2.7), mean PsA-modified SHS change from baseline to week 24 was –0.70 with infliximab 5 mg and +0.82 with placebo (p = .001); 97% of the infliximab group and 88% of the placebo group had scores ≤ SDD
GO-REVEAL (A Randomized Evaluation of Safety and Efficacy in Subjects with Psoriatic Arthritis Using a Human Anti-TNF Monoclonal Antibody) (101)	On a 0–528 scale (SDC = 1.56), mean changes from baseline to week 24 on the PsA-modified SHS were –0.16 with golimumab 50 mg (p = .011 vs. placebo), –0.02 with golimumab 100 mg (p = .086 vs. placebo), and +0.27 with placebo. The percent of patients with a PsA-modified SHS change from baseline ≤ SDD was 96% with golimumab 50 mg, 94% with golimumab 100 mg, and 89% with placebo
GO-VIBRANT (A study of golimumab in participants with active PsA) (102)	On a 0–528 scale (SDC = 2.49), mean change in total PsA-modified SHS was –0.4 with intravenous golimumab 2 mg/kg vs. 2.0 in the placebo group (p < .001)
RAPID-PsA (Certolizumab pegol in subjects with adult onset active and progressive PsA) (177)	Imputation method significantly affected assessment and reporting of radiographic progression With no imputation, mTSS change from baseline to week 24 was +0.06 with certolizumab pegol (pooled doses) and + 0.29 with placebo (p = .008)
PSUMMIT-1 and PSUMMIT-2 (Study of the safety and effectiveness of ustekinumab in patients with PsA) (29)	On a 0–528 scale (SDC = 2.01), mean changes from baseline to week 24 was 0.4 with ustekinumab 45 or 90 mg and 1.0 with placebo (p < .001). In both groups, change in SHS was ≤ SDD for most patients (92% with ustekinumab and 84% with placebo; p < .001)
ASTRAEA (Active PSoriaTic Arthritis RandomizEd TriAl) (103)	Mean change from baseline in PsA-modified total SHS score was 0.30 with abatacept vs. 0.35 with placebo at week 24; statistical significance and SDC not reported

mTSS: modified total Sharp score; PsA: psoriatic arthritis; Q2W: every 2 weeks; Q4W: every 4 weeks; QW: once weekly; SDC: smallest detectable change; SHS: Sharp/van der Heijde score; TSS: total Sharp score.

Spine measures

The most commonly used indices for assessment of spondylitis include the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and the Bath Ankylosing Spondylitis Functional Index (BASFI) (Table 4). Components of the BASDAI and BASFI, along with patient assessments of spinal pain and global impression of disease activity rated on a 0–10 cm visual analog scale, are included in the composite Assessment of SpondyloArthritis international Society (ASAS) improvement criteria (121), which is the most common primary endpoint in recent AS clinical trials.

While axial disease is an important clinical domain of PsA, fewer than 5% of patients have isolated spinal joint involvement (26). Thus, because not all patients with PsA present with spinal involvement and use of MRI in clinical trials is expensive, formal assessment of spondylitis has not been performed in randomized controlled PsA trials. However, the effects of biologic and small-molecule inhibitor therapies on axial disease in PsA can be inferred from results of controlled clinical trials of these agents in patients with AS. All TNF-α inhibitors approved for the treatment of active PsA (etanercept, adalimumab, infliximab, golimumab, and certolizumab pegol) are also approved for the treatment of AS. In AS clinical trials, ASAS20 response rates were similar to all TNF-α inhibitors, ranging from 58 to 61% (122). In the MEASURE 1 and MEASURE 2 studies of the IL-17A inhibitor secukinumab in patients with AS, ASAS20 response rates at week 16 were 61 and 61%, respectively, with the 150-mg dose and 60 and 41%, respectively, with the 75-mg dose (123). ASAS20 response rates were significantly higher with tofacitinib 5 mg (80.8%) compared with placebo (41.2%; p < .001) in a phase 2 dose-ranging study in AS (124). In a small open-label pilot study of abatacept in AS, 27% of TNF inhibitor-naïve and 20% of TNF inhibitor-experienced patients achieved ASAS20 responses, suggesting limited efficacy of abatacept for this indication (125). Methotrexate has not been demonstrated to show benefit in patients with AS (126). A clinical trial of ixekizumab (NCT02696798) is ongoing in AS.

PRO measures

PRO measures were developed to assess patients’ perspectives of various quality-of-life domains. These measures include the pain visual analog scale (VAS), the health assessment questionnaire disability index (HAQ-DI), the EuroQoL 5 dimension questionnaire (EQ-5D), and the 36-item short form survey (SF-36), which is broken into the physical component summary (PCS) and mental component summary (MCS) (56). Changes from baseline of ≥10 mm for pain VAS and ≥0.30–0.35 for HAQ-DI have been used as MCID in PsA trials, and an MCID for change from baseline in SF-36 has been defined as 5–10 (2.5–5 for PCS and MCS individually) in PsA (101,127–130). Clinically important improvements in these PROs have been reported in PsA clinical trials of adalimumab, etanercept, infliximab, golimumab (subcutaneous and intravenous), certolizumab pegol, ustekinumab, guselkumab, secukinumab, ixekizumab, apremilast, tofacitinib, and abatacept (100–103,131–139). Additionally, the PsAID12, which is available on the GRAPPA mobile-phone app, and the RAPID3 questionnaires are useful for dermatologists to use in clinical practice to evaluate the impact of PsA symptoms on patients (140,141).

Additional composite measures

Composite indices, such as the Psoriatic Arthritis Disease Activity Score (PASDAS), have been developed as a means of integrating information from a variety of individual outcome indices, including physician’s global assessment scores, disease activity indices, and PROs into one meaningful measure of PsA (142). Minimal disease activity (MDA), a measure that encompasses all aspects of PsA, is different from PASDAS and similar composite indices because it was created to be an objective target for treatment. When MDA criteria are met, PsA is considered to be in a satisfactory state of disease activity for both patients and physicians. Patients are classified as achieving MDA when they meet 5 of 7 criteria, including tender joint count ≤1; swollen joint count ≤1; PASI ≤1 or body surface area ≤3%; patient pain VAS ≤15; patient global disease activity VAS ≤20; health assessment questionnaire ≤0.5; and tender entheseal points ≤1 (143). The value of using MDA as a treatment target was demonstrated in the TIght COntrol of Psoriatic Arthritis (TICOPA) trial (144). In this study, MDA was successfully used as an objective goal in a treat-to-target approach as patients with early PsA treated with this tight control goal had significantly improved joint outcomes compared with patients receiving standard care.

Summary

PsA is a heterogeneous, rapidly developing disease with potentially deleterious effects and frequently occurs within 10 years following the appearance of psoriasis (21). We consider PsA to be the major comorbidity of psoriasis due to the likelihood of PsA to result in permanent disability if left untreated. Early PsA diagnosis and initiation of treatment can prevent possibly adverse consequences, whereas a diagnostic delay of more than 6 months contributes to poor radiographic and functional outcomes (27). Therefore, while providing important care to manage the skin symptoms of psoriasis, dermatologists have the unique ability to recognize early signs of PsA in patients with psoriasis and be the first healthcare practitioners to detect PsA. Dermatologists can then initiate treatment with agents that are effective for both psoriasis and PsA. There are currently available PsA drugs that block TNF-α and IL-17A. These medications can inhibit radiographic progression, control signs and symptoms, improve physical function, prevent disability, and improve quality of life. Thus, dermatologists, by detecting PsA early, are key to preventing disability in these patients. Treatments should be selected that target the most burdensome or progressive PsA clinical domain(s) for each patient based on the efficacy profiles of available treatments for these domains.

Acknowledgments

Technical assistance with editing and styling of the manuscript for submission was provided by Oxford PharmaGenesis Inc. and was funded by Novartis Pharmaceuticals Corporation. The authors were fully responsible for all content and editorial decisions and received no financial support or other form of compensation related to the development of this manuscript.

Disclosure statement

Alice Gottlieb: Consultant/advisory board agreements for Janssen, Celgene, Bristol-Myers Squibb, AbbVie, UCB, Novartis, Incyte, Lilly, Reddy Labs, Valeant, Dermira, Allergan, Sun, XBiotech, and Leo. Pharmaceutical industry research/educational grants from Janssen, Incyte, UCB, Novartis, and Lilly, and XBiotech.

Joseph F. Merola: Consultant and/or investigator for Merck Research Laboratories, Abbvie, Eli Lilly and Company, Novartis, Janssen, UCB, Samumed, Celgene, Sanofi Regeneron, GSK, Almirall, Sun Pharma, Biogen, Pfizer, Incyte, Aclaris, and Leo Pharma.

Additional information

Funding

This work was supported by Novartis Pharmaceuticals Corporation.