Epidemiology

Psoriatic arthritis (PsA) is defined as a unique inflammatory arthritis affecting the joints and connective tissue, and is associated with psoriasis of the skin or nails.¹ There are difficulties in estimating its prevalence due to the lack of a precise definition and diagnostic criteria for PsA.² The prevalence of psoriasis in the general population has been estimated at between 2% and 3%,¹ and the prevalence of inflammatory arthritis in patients with psoriasis has been estimated to be up to 30%.³ PsA affects males and females equally, with a worldwide distribution. Figures for the UK have estimated the adjusted prevalence of PsA in the primary care setting to be 0.3%, based on data from north-east England involving six general practices, covering a population of 26,348.⁴ Another study reported PsA prevalence rates per 100,000 of 3.5 for males and 3.4 for females, based on data from 77 general practices in the Norwich Health Authority, with a population of 413,421.⁵ Severe PsA with progressive joint lesions can be found in at least 20% of patients with psoriasis.⁶

Aetiology, pathology and prognosis

Psoriatic arthritis is a hyperproliferative and inflammatory arthritis that is distinct from rheumatoid arthritis (RA).^7,8 The aetiology of PsA is not fully known; genetic susceptibility and exogenous influences might play roles in the cause of disease.⁹ The expression of major histocompatibility complex antigens [e.g. human leucocyte antigen (HLA)-B27] might also predispose certain patients to develop PsA, as well as a number of environmental factors, such as trauma, repetitive motion, human immunodeficiency virus infection, and bacterial infection.⁹ PsA is diagnosed when a patient with psoriasis has a distinctive pattern of peripheral and/or spinal arthropathy.¹⁰ The rheumatic characteristics of PsA include stiffness, pain, swelling, and tenderness of the joints and surrounding ligaments and tendons.¹¹

Several clinical features distinguish PsA from RA. In PsA, the absolute number of affected joints is less and the pattern of joint lesion involvement tends to be asymmetric.¹² The joint distribution tends to occur in a ray pattern in PsA, with the common involvement of distal interphalangeal (DIP) joint and nail lesions. All joints of a single digit are thus more likely to be affected in PsA, whereas in RA the same joints on both sides tend to be affected.¹ Dactylitis, spondylitis and sacroiliitis are common in PsA, whereas they are not in RA.¹² In PsA the affected joints are tighter, contain less fluid, and are less tender than those in RA, with a propensity for inflammation of the entheseal sites. PsA and RA also show differences in the inflammatory reaction that accompanies each form of arthritis.¹² Extra-articular manifestations of PsA are also different from those of RA; rheumatoid nodules are particularly absent in patients with PsA.¹ Most patients with PsA develop psoriasis first, while joint involvement appears first only in 19% of patients, and concurrently with psoriasis in 16% of cases.¹⁰ For those who develop psoriasis first, the onset time of PsA is around 10 years after the first signs of psoriasis.¹ In addition, rheumatoid factor (RF) (an antibody produced by plasma cells) may be detected in about 13% of patients with PsA, whereas it can be detected in more than 80% of patients with RA.¹

Psoriatic arthritis is a progressive disorder, ranging from mild synovitis to severe progressive erosive arthropathy.^11,13 Research has found that patients PsA presenting with oligoarticular disease progress to polyarticular disease; a large percentage of patients develop joint lesions and deformities, which progress over time.⁹ Despite clinical improvement with current disease-modifying antirheumatic drug (DMARD) treatment, radiological joint damage has been shown in up to 47% of patients with PsA at a median interval of 2 years.¹⁴ Untreated patients with PsA may have persistent inflammation and progressive joint damage.¹¹ The deformities resulting from PsA can lead to shortening of digits due to severe joints or bone lysis.¹ Remission can occur in PsA, especially in patients with Health Assessment Questionnaire (HAQ) levels of < 1 score.¹⁵ Of those who can sustain clinical remission, only a small fraction of patients can discontinue medication with no evidence of damage.¹⁶ Research has reported that the frequency of remission was 17.6% in patients with PsA, and the average duration of remission was 2.6 years, from data of 391 patients with peripheral arthritis.¹⁶ Joint damage can occur early in the disease often prior to functional limitation.^9,17 This appears to be associated with the development of inflamed entheses close to peripheral joints, although the link still remains largely unclear.¹³ It has been shown that there is an association between polyarthritis and functional disability, with higher mean HAQ scores than those in oligoarthritic patients.^18,19

A number of risk factors have been found to be predictive of the progression of PsA. A polyarticular onset (five or more swollen joints) of PsA is an important risk factor in predicting the progressive joint deformity.²⁰ Each actively inflamed joint in PsA is associated with a 4% risk of increased damage within 6 months.¹ HLA antigens have also been found to be predictive of the progression of joint damage. It has been shown that HLA-B27, HLA-B39 and HLA-DQW3 were associated with disease progression.²¹ Other risk factors for a more progressive course of PsA include the presence of an elevated erythrocyte sedimentation rate (ESR) and being female.^1,22

A classification scheme for PsA on the basis of joint manifestations describes five patterns of disease:^9,23

1. Distal interphalangeal arthritis This condition is considered as the classic form of PsA. It can occur as the sole presentation or in combination with other symptoms. It can be symmetrical or asymmetrical and can involve a few or many joints. Adjacent nails may demonstrate psoriatic changes and progressive joint erosions are common.
2. Arthritis mutilans It is a severe presentation of the disease with osteolysis of the phalanges, metatarsals and metacarpals.
3. Symmetric polyarthritis The clinical feature of symmetric polyarthritis is similar to RA, with inflammation of the metacarpals and the proximal interphalangeal joints being prominent. However, it is usually milder than RA and patients are often RF-negative.
4. Oligoarthritis This is the most common condition of PsA, which is characterised by asymmetric involvement of a small number of joints (fewer than four). Arthritis in a single knee might be the first symptom of oligoarthritis.
5. Spondylitis and/or sacroiliitis It resembles ankylosing spondylitis, but is generally less severe and less disabling. The axial skeleton tends to be involved in an atypical fashion, with the lumbar spine as the most common site of involvement.

Despite this classification, these patterns of PsA often overlap and evolve from one pattern to another as the disease progresses and diagnostic investigations become more thorough.¹³ A common feature of PsA is dactylitis (or ‘sausage digit’) in which the whole digit appears swollen due to inflammation of the tendons and periosteum as well as the joints.^9,11 Radiographic features of PsA involve the distinctive asymmetric pattern of joint involvement, sacroiliitis and spondylitis, bone erosions, new bone formation, bony ankylosis, bony outgrowths in the axial skeleton, osteolysis and enthesopathy.

Significance in terms of ill health

The health burden of PsA can be considerable. PsA is a lifelong disorder and its impact on patients' functional status and quality of life (QoL) fluctuates over time.²⁴ As it involves both skin and joints, PsA can result in significant impairment of QoL and psychosocial disability^7,10 compared with a healthy population. Patients with PsA score significantly worse in health-related quality of life (HRQoL) assessment on physical mobility, pain, energy, sleep, social isolation and emotional reaction.²⁵ A comparison of HRQoL between patients with PsA and patients with RA found that both patient populations had lower physical health than healthy control patients.²⁶ Patients with PsA reported more role limitations due to emotional problems and more bodily pain after the adjustment of the difference in vitality and other covariates. These findings were also reflected in another comparison of disability and QoL between patients with RA and patients with PsA; this study reported that despite greater peripheral joint damage in patients with RA the function and QoL scores were similar for both groups.^27,28 These reveal that there might be unique psychological disabilities associated with the psoriasis dimension (i.e. skin lesion) of PsA. Due to the skin involvement, patients with PsA may also suffer from other psychological consequences, such as embarrassment, self-consciousness and even depression. Because of a significant reduction in a patient's HRQoL, ideally PsA should be diagnosed early and treated aggressively in order to minimise joint damage and skin disease.¹⁷

The severity of PsA is also reflected in increased mortality. Patients with PsA have a 60% higher risk of mortality relative to the general population.^24,29,30 The causes of premature death are similar to those noted in the general population, with cardiovascular causes being the most common.¹ The estimated reduction in life expectancy for patients with PsA is approximately 3 years.³¹

The economic costs of PsA have not been well quantified. In the USA, the mean annual direct (health and social care) cost per patient with PsA is estimated as US$3638 according to data from Medstat MarketScan in 1999–2000.³² In Germany, the mean annual direct cost per patient with PsA is estimated as €3162, with a mean indirect cost (time lost from work and normal activities) per patient of €11,075.³³ Studies of RA^34–36 and psoriasis³⁷ have shown that costs increase with the severity of both diseases, and productivity losses are significant,^38,39 largely as a consequence of extensive work disability.³⁵ These findings are likely to be generalisable to PsA.

Studies of the economic impact of RA in the UK before the introduction of biologic therapies found that direct health-care costs represented about one-quarter of all costs, and these were dominated by inpatient and community day care,⁴⁰ with DMARDs representing a minor proportion: 3%–4% of total costs and 13%–15% of direct costs.⁴¹ Evidence from the USA suggests that expenditure on biologic therapies might represent 35% of direct cost,⁴² but similar data are not yet available for the UK. Increasing expenditure on biologic therapies might be at least partly offset by cost savings elsewhere,⁴³ although, as yet, the evidence for this is only suggestive.

American College of Rheumatology response criteria

The ACR response criteria were developed after the identification of a set of core disease activity measures. ACR 20 requires a 20% reduction in the tender joint count (TJC), a 20% reduction in the swollen joint count (SJC), and a 20% reduction in three out of five additional measures, including patient and physician global assessment, pain, disability and an acute-phase reactant. In patients with RA, ACR 20 has been confirmed as being able to discriminate between a clinically significant improvement and a clinically insignificant one.^46,47 It is unclear whether the ACR 20 has the same discriminatory validity in PsA.⁴⁸ The ACR 20 is generally accepted to be the minimal clinically important difference that indicates some response to a particular intervention. The ACR 50 reflects significant and important changes in the patient's disease status that may be acceptable to both clinician and patient in long-term management. The ACR 70 represents a major change and approximates in most minds to a near remission. Because of the differences between PsA and RA, it is imperative that, when the ACR response criteria are used in the trials of treatment for PsA, the DIP joints are included. Rather than changes from bad to moderate synovitis in any individual joint, these criteria detect improvement from swollen to not swollen or from tender to not tender joints. Therefore, patients with oligoarthritis in a few large joints may not appear to respond as well on this outcome as patients with polyarthritis involving many smaller joints.

Psoriatic Arthritis Response Criteria

The Psoriatic Arthritis Response Criteria were developed for a trial of sulfasalazine in PsA,⁴⁹ and incorporate four assessment measures (patient self-assessment, physician assessment, joint pain/tenderness score and joint swelling score). Treatment response was defined as an improvement in at least two of these four measures, one of which had to be joint pain/tenderness score or joint swelling score, with no worsening in any of these four measures. PsARC has not been validated, but responses assessed by it do parallel those identified with ACR 20. A limitation of PsARC is that although it was developed for assessment of PsA, it does not incorporate an assessment of psoriasis. The working group producing the British Society for Rheumatology (BSR) guidelines for the use of biologics in PsA⁵⁰ elected to use PsARC as the primary joint response to biologic treatment, although it advocates some extra data collection, such as a patient self-assessed disability (HAQ), and a biochemical marker of disease activity, such as ESR or C-reactive protein (CRP).

Radiological assessments

In all arthropathies progression of the disease can only be truly measured by assessment of the joint damage. The radiological assessments include the Steinbrocker, Sharp and Larsen methods. A modification of the Steinbrocker method, which assigns a score for each joint has been validated for PsA. The Sharp method grades all the joints of the hand separately for erosions and joint space narrowing, each erosion being assigned a score of 0–5 and each joint space narrowing a score of 0–4. A total score (maximum 149) is calculated. The Total Sharp Score (TSS), modified to include the DIP and metatarsophalangeal joints of the feet and interphalangeal joint of the first toe, has been used in the trials of etanercept and adalimumab.^51,52 None of these methods that were developed for RA score additional radiographic changes that are specific to PsA. A new score has been tested by Wassenberg et al.,⁵³ but this scoring method has not yet been validated in clinical trials. Whichever method is selected it is important that trials should be stratified by baseline radiographic findings.

Health Assessment Questionnaire

The HAQ score is a well-validated tool in the assessment of patients with RA.⁴⁸ It focuses on two dimensions of health status: physical disability (eight scales) and pain, generating a score of 0 (least disability) to 3 (most severe disability). A modification of the HAQ for spondylarthropathies (HAQ-S) and for psoriasis (HAQ-SK) have been developed, but when tested against HAQ their scores were almost identical,⁵⁴ suggesting either can be used in PsA.⁴⁸ The HAQ is one component of the ACR 20 (50 or 70) response criteria.

The HAQ has been tested in patients with PsA, showing a moderate-to-close correlation with disease activity as measured by the actively inflamed joint count and some measures of clinical function (including the ACR functional class).⁵⁵ Although the HAQ has been used as a disability measure and is a common outcome measure in PsA trials, it may not sufficiently incorporate all aspects of disease activity (i.e. deformity or damage resulting from disease process, especially in late PsA), therefore, clinical assessment of disease activity and both clinical and radiological assessments of joint damage remain important outcome measures in PsA.⁵⁶

Overall, the advantage of the HAQ as an instrument is that it can measure the functional and psychological impact of the disease. HAQ is conventionally used as a driver of QoL scores and costs in main economic evaluations on the use of biologics and DMARDs in RA.^57–59

Psoriasis Area and Severity Index

When evaluating the efficacy of interventions in the treatment of PsA, the outcome measures used must assess disease activity in both the joint and the skin.⁴⁸ In clinical trials of patients with psoriasis, assessment of the response to treatment is usually based on the Psoriasis Area and Severity Index (PASI). The PASI is also used in trials of PsA; given the various degrees of severity of psoriasis in these patients, not all patients are evaluable for the assessment of response – at least 3% of the body surface area (BSA) has to be affected by the skin disease in order for the PASI measure to be used.⁴⁸ Although it is widely used, the PASI measure also has a number of deficiencies: its constituent parameters have never been properly defined; it is insensitive to change in mild-to-moderate psoriasis; estimation of disease extent is notoriously inaccurate; and the complexity of the formula required to calculate the final score further increases the risk of errors. It combines an extent and a severity score for each of the four body areas (head, trunk, upper extremities and lower extremities). The extent score of 0–6 is allocated according to the percentage of skin involvement (e.g. 0 and 6 represent no psoriasis and 90%–100% involvement, respectively). The severity score of 0–12 is derived by adding scores of 0–4 for each of the qualities of erythema (redness), induration and desquamation representative of the psoriasis within the affected area. It is probable, but usually not specified in trial reports, that most investigators take induration to mean plaque thickness without adherent scale and desquamation to mean thickness of scale rather than severity of scale shedding. The severity score for each area is multiplied by the extent score and the resultant body area scores, weighted according to the percentage of total BSA that the body area represents (10% for head, 30% for trunk, 20% for upper extremities and 40% for lower extremities), are added together to give the PASI score. Although PASI can theoretically reach 72, scores in the upper half of the range (> 36) are not common even in severe psoriasis. Furthermore, it fails to capture the disability that commonly arises from involvement of functionally or psychosocially important areas (hands, feet, face, scalp and genitalia), which together represent only a small proportion of total BSA.

Although the optimum assessment outcomes for PsA trials are yet to be defined, those selected as the primary measures of efficacy in this review, namely PsARC-, ACR 20/50/70-, HAQ- and PASI-based measures, all have discriminatory capability and are generally accepted for the assessment of treatment effect. HAQ has been chosen as our primary outcome variable of arthritis in the economic evaluation because it makes it technically feasible to evaluate the impact of retarding and/or halting the progression of the disease, both in an economic sense and in terms of QoL. PASI has been chosen as the primary outcome variable of psoriasis in the economic evaluation because it is recommended to assess severity and response in the British Association of Dermatologists (BAD) guidelines and used in the majority of randomised controlled trials (RCTs).

Costs of current service

Based on prices from the British National Formulary (BNF),⁶⁵ weekly treatment costs with the most commonly used DMARDs in PsA, sulfasalazine and MTX are approximately £2 and < £0.50, respectively. The cost of ciclosporin is approximately £40–80 per week.

Prescriptions for DMARDs for all indications have been rising rapidly in general practice in England from 300,000 per quarter year in December 2003 to over 500,000 in December 2008, with expenditure increasing from £2M per quarter year to nearly £4.5M during this period. In addition to the cost of DMARDs, the cost of NSAIDs was almost £4M per quarter year in December 2008, although the number of prescriptions and expenditure on NSAIDS has fallen sharply in recent years.⁶⁶

Variation in service

No surveys of UK service models for PsA have been conducted. Although PsA is a disease of joints and skin it is treated mainly by rheumatologists. A study of patients with confirmed PsA in the Netherlands found considerable variations among rheumatologists in the delivery of care; 29% failed to diagnose PsA, mainly due to their failure to enquire about skin lesions.⁶⁷ Of those who did correctly diagnose PsA, only 43% referred patients to a dermatologist and 66% ordered laboratory tests. The median costs for imaging and laboratory investigations were higher for those patients who were correctly diagnosed with PsA than for the remaining patients who were incorrectly diagnosed.

Anticipated costs of biologic interventions

Based on the recommended dose regimen (25-mg injections administered twice weekly as a subcutaneous injection), the initial 3-month acquisition cost of etanercept is £2324, and the annual cost thereafter is £9296. The acquisition costs of adalimumab are the same, based on the recommended dose regimen (40-mg subcutaneous injections administered every other week). The recommended dose for infliximab is 5 mg/kg is given as an intravenous (i.v.) infusion over a 2-hour period followed by additional 5-mg/kg infusion doses at 2 and 6 weeks after the first infusion then every 8 weeks thereafter, each dose corresponding to three, four or five vials of infliximab depending upon the patient's body weight. The initial 3-month acquisition cost of infliximab is estimated to be £5035, assuming four vials, and the annual cost thereafter is £10,912.

Current expenditure on biologic therapies in England is considerable. For all indications, the cost of prescribing in 2008 was £152.2M for etanercept, £102.7M for adalimumab and £77.1M for infliximab, with > 95% of these prescriptions dispensed by hospitals.⁷² Expenditure for biologic drugs increased during 2008 by 15% for etanercept, 55% for adalimumab and 25% for infliximab. Among the drugs appraised by the National Institute for Health and Clinical Excellence (NICE), etanercept and adalimumab are now ranked in the top five by estimated cost of prescribing in England.

This report contains reference to confidential information provided as part of the NICE appraisal process. This information has been removed from the report and the results, discussions and conclusions of the report do not include the confidential information. These sections are clearly marked in the report.