The last 3 decades have seen a wholesale change in the approach to the management of rheumatoid arthritis (RA). The emphasis has changed from symptom control, moderation of disease progression after structural damage had occurred, and accommodation of disability, to modern outcome-driven strategies. Rheumatologists now have treat to target1, remission induction2,3, damage prevention strategies4, and increasingly stretch targets of maintaining productivity5 and drug-free remission6. These management approaches and outcome targets demonstrate how much the treatment and our expectations have changed with advancing knowledge and new therapies in RA.
It was not until cohort studies of the early 1980s demonstrated the true consequence of uncontrolled inflammation that attitudes to RA management began to change7. It was in 1987 that the American College of Rheumatology (ACR) revised the criteria for the classification of RA8. The criteria were developed to improve the homogeneity of patients entered into clinical trials with established disease. However, the pathognomonic features of RA such as classical clinical deformity, periarticular joint erosion, and nodules occur late in the disease process. The fact that the 1987 ACR criteria included the latter diminished their value in early disease, and several studies demonstrated diminished sensitivity for diagnosis of RA in the early months of symptoms9,10. In recognition of this and also of the specialty’s change in approach to the management of RA, it was timely for the classification criteria for RA to be revised jointly by the ACR and the European League Against Rheumatism (EULAR) in 201011.
The working group of the ACR and EULAR determined that the 1987 ACR classification for RA, for the reasons outlined above, hampered recruitment of patients with early disease to intervention studies in RA11. The remit was to identify patients newly presenting with undifferentiated inflammatory synovitis, who were likely to have persistent and/or erosive disease. Thus the new criteria would go hand in hand with the accepted benefit of early diagnosis and intervention. Additionally, the new criteria were to act as a basis for commencing disease-modifying therapy and also to remain applicable in patients presenting with late disease. The final criteria omit radiographic changes and the presence of nodules and concentrate on joint involvement, serology, including anti-citrullinated protein antibody (ACPA) testing, markers of acute-phase response, and symptom duration. A numerical value determines the likelihood of definite RA. As a result of this reclassification, some patients that have previously been recruited to studies in early undifferentiated inflammatory arthritis may well have been classified as RA according to the revised criteria12.
In this issue of The Journal the Canadian Early Arthritis cohort (CATCH) investigators present findings from a large, early synovitis cohort and examine disease activity in patients meeting the old (1987) versus the new (2010) criteria13. CATCH is a nationwide, multicenter, prospective, real-world cohort of adults with early inflammatory arthritis, which started recruiting in 2007. The participants should be congratulated on establishing a national early RA cohort providing important data on the outcome and management of early RA in Canada. Caution may be required regarding the national applicability of the findings, given that 2 provinces, Ontario and Quebec, provided over 80% of the patients.
The results are clearly presented in the CATCH report, but there are some interesting observations. Of note, there is a trend to reduced symptom duration at time of presentation between 2007 and 2011. It is established that the best predictor of entering remission after intervention with conventional disease-modifying drugs in RA is short disease duration14 and as such these data show promise. Although significance is not reached, it may be an early sign of the success of the many initiatives undertaken by Canadian rheumatologists to promote awareness of RA among both the general public and health professionals. Many centers worldwide have worked tirelessly to establish where the delays lie in the assessment pathway for inflammatory arthritis and reduce symptom duration at time of presentation to the rheumatologist, but there is little objective evidence of reliable effective strategies15. The group have also demonstrated that patients presenting with disease satisfying the 2010 criteria are less severe than patients presenting with disease satisfying the 1987 criteria, but are also less likely to be treated with methotrexate. Recognition of such findings may be paramount in improving patient outcomes.
Hence, the new criteria herald a new era for researchers in RA, creating opportunities for rheumatologists to enroll real-life patients with early RA into therapeutic clinical trials and potentially provide data for best practice at the earliest stages of RA. For clinicians, an understanding of the lack of comparability of new and historically classified RA cohorts is essential as new studies using the new ACR classification criteria emerge. The CATCH study group have demonstrated that patients with less severe disease are likely to be identified using the new criteria, a fact that alone may offer a greater opportunity for intervention and optimal outcomes. We may hypothesize that the development of the new criteria for the classification of RA may in themselves improve patient outcomes. Further caution may also be required. A recent review has suggested that the new RA classification criteria improve sensitivity for the diagnosis of RA, but specificity is decreased16. As such, patients fulfilling the criteria have a high probability of disease, but not fulfilling the criteria does not rule out RA. Equally, the CATCH cohort has demonstrated that patients with less severe disease may be identified and their natural history and prognosis may differ from what we currently understand from RA cohort studies. For example: a higher proportion of patients may enter remission and be able to stop therapy. Although therapy guidelines suggest rheumatologists should slowly withdraw therapy once a remission state is achieved, there is a dearth of data allowing clinicians to accurately predict which patients will successfully remain therapy-free. Thus the research agenda is set. In essence the 2010 ACR/EULAR classification criteria for RA may just open the “window of opportunity” for intervention even wider, providing rheumatologists and patients a chance of optimizing therapy and best possible outcomes.