Preventing the Progression From Undifferentiated Arthritis to Rheumatoid Arthritis: The Clinical and Economic Implications
Published Online: November 19, 2010
Michael H. Schiff, MD
Over the past few years, the management of rheumatoid arthritis (RA) has emphasized early, aggressive therapy with traditional diseasemodifying antirheumatic drugs (DMARDs) and/or antirheumatic biologic agents. This has been the result of a greater understanding of the pathophysiology of undifferentiated arthritis (UA) and RA, along with the availability of several biologic agents that down-regulate aspects of the autoimmune response that contribute to RA.
Approximately 40% to 50% of patients with UA experience spontaneous remission while roughly 30% progress to RA, and the remainder develop other conditions.1,2 Researchers have investigated criteria that can help identify those patients with UA who have a high probability of progressing to persistent and/or erosive RA. Once reliable criteria have been defined, the next question is whether active therapy, with either traditional DMARDs or antirheumatic biologic agents, can be effective in slowing or preventing disease progression. Clinical trials demonstrate that early treatment can delay or stop disease progression.3-6 From a managed care perspective, limiting treatment to only those patients with UA who would most benefit from active intervention has significant implications.
This article reviews the latest research supporting efforts to identify patients with UA who are at high risk of progressing to persistent and/or erosive RA, and offers insights on therapeutic approaches for these patients.
Defining Undifferentiated Arthritis
The term "undifferentiated arthritis" was incorporated into studies performed by the Leiden Early Arthritis Clinic in the Netherlands to describe cases of inflammatory arthritis where a specific diagnosis could not be made (Table 1).1,7,8 The question that follows is, why is UA important?
UA to RA Progression
Disease progression in patients who initially present with UA is comparable to that in patients who are initially diagnosed with RA. A study by van Aken and colleagues compared disease progression in patients who initially presented with UA and progressed to RA within 1 year (UA-RA group; n = 92) with disease progression in those who initially presented with RA (RA-RA group; n = 62) based on the 1987 American College of Rheumatology (ACR) criteria.1 Outcome measures were followed for 4 years following initial presentation.
No significant differences between the 2 groups were observed during any follow-up period (years 1, 2, 3, and 4) for radiographic progression of joint damage (as measured by Sharp/van der Heijde score) or disease activity and functional capacity (as measured by Health Assessment Questionnaire [HAQ] or Disease Activity Score [DAS]). Similar disease progression was observed in both groups despite more patients in the RA-RA group initiated on a DMARD by 3 months after presentation (100% vs 46%). This study suggests that patients with UA who eventually progress to RA have the same outcomes as those initially diagnosed with RA.
Predicting UA to RA Progression
Prediction rules are important tools in determining which patients with UA will progress to RA. The first validated prediction rule was developed in Europe and published in 2008. It used patient information that is regularly gathered during the initial office visit9,10-age, gender, number and types of joints involved, duration of morning stiffness, C-reactive protein level, and serologic assay results (presence of rheumatoid factor [RF] and anti-citrullinated protein antibody [ACPA]). When applied to 3 independent cohorts of patients with early-onset UA, this prediction rule accurately estimated the risk of developing RA in more than 75% of the patients with UA.10
ACR/EULAR RA Classification Criteria, 2010
In September 2010, the ACR, in collaboration with the European League Against Rheumatism (EULAR), released the updated RA classification criteria with the goal of distinguishing newly presenting patients with undifferentiated inflammatory arthritis who had a high probability of developing persistent or erosive RA (Table 2).11 These represent a significant change from the 1987 ACR criteria, which were not designed to identify patients with early RA or those at high risk of progressing to RA.7,12 The authors emphasize that the 2010 scoring system requires testing in various patient cohorts to demonstrate its validity. It will be important to determine if these criteria perform better for some patient cohorts than others and allow accurate individualization of treatment decision-making.
The 2010 ACR/EULAR RA classification criteria can have a significant impact on the management of patients at the very early stages of disease. Although the authors clearly state that this classification system is not designed as a referral tool to help diagnose patients with RA, it is likely that the criteria will be incorporated into clinical practice as part of the overall diagnostic procedure. This can aid clinicians in identifying patients at early stages of RA who would benefit from therapeutic intervention and/or referral to a specialist. It is also possible that these criteria will be adopted by managed care organizations to determine the appropriateness of particular interventions for patients with early signs of RA. There are also economic benefits to limiting therapy to those who have a high probability of developing persistent RA while minimizing therapy for low-risk patients.
Screening Patients With UA at Risk for Progressing to RA
Screening patients who initially present with UA to identify those who are at a high probability of progressing to RA can be beneficial for the patient and the provider.
Serologic factors are a convenient and accurate measure to predict progression to RA. RF has been used for several decades and is the only serologic factor mentioned in the 1987 ACR criteria.7,13 More recently, several other biologic markers have been identified, including various ACPAs (tested as anti-cyclic citrullinated peptides [anti-CCPs]).14,15
To test the predictive value of ACPAs for progression to RA, van Gaalen and colleagues compared serologic assay results and disease progression in patients with UA.16 Of the 318 patients with UA, 69 were anti-CCP positive and 249 were anti-CCP negative at study entry. After 3 years, 93% of anti-CCP positive patients met the ACR criteria for RA, compared with 25% of anti-CCP negative patients (odds ratio, 37.8; 95% confidence interval [CI], 13.8-111.9) (Figure 1). The authors concluded that the presence of anti-CCP antibodies helps differentiate those patients with UA who are at risk for developing RA. The predictive value of RF and ACPA was further supported by a 4-year follow-up study of patients with undifferentiated polyarthritis.17
Although several different serologic markers can be used to screen patients with UA, studies have not definitively determined whether there are differences in predictive values associated with them. One study compared the predictive value of several of these markers, including RF and anti-CCPs.18 The results suggested that there were no significant differences in the predictive value of these serologic markers in distinguishing patients with UA who would later progress to RA. Furthermore, using a combination of serologic tests did not improve the predictive value, indicating that 1 assay is sufficient to make a determination.
In addition to serologic markers, researchers have attempted to identify other measures to predict progression from UA to RA. These include radiologic evaluation of baseline erosions in joints of the hands and feet, magnetic resonance imaging (MRI) of joints, or a combination of radiologic evaluation, serologic assays, and symptoms.19-22 Although these methods may prove effective in predicting which patients with UA may progress to RA, the inclusion of radiographic criteria, particularly MRI scans, can be prohibitive in terms of cost and time as a screening tool, and because an experienced musculoskeletal radiologist would be needed.
2010 ACR/EULAR RA Classification Criteria
The 2010 ACR/EULAR RA classification criteria provide a standardized algorithm for assessing patients with UA. These criteria will be important when developing inclusion criteria for clinical trials with patients in the very early stages of disease. They will also allow a comparison of patient cohorts from different clinical trials that adopt these criteria. Because these criteria use information regularly gathered at initial presentation in the clinic, assessment will not add to the time and cost of evaluation. Although these criteria are not meant to be used as a reference tool for diagnosis, they can be incorporated as part of the screening and diagnostic process for patients presenting with UA, and aid in the treatment decision-making process.
PDF is available on the last page.