Does early treatment of rheumatoid arthritis produce long-lasting benefits?
Summary written by Jon Giles, M.D.
The past 15 to 20 years have witnessed a paradigm shift in the way in which rheumatoid arthritis (RA) is treated. Before this, because of modest efficacy, diminishing effectiveness over time, and toxicity, treatment with disease-modifying antirheumatic drugs (DMARDs) was often postponed until years after diagnosis. The introduction and widespread use of more effective medications with favorable toxicity profiles (i.e. methotrexate and TNF inhibitors) has encouraged use of these agents sooner. Further acceptance of early treatment has been bolstered by clinical trials with a variety of DMARDs showing improved outcomes in terms of joint function, deformity, quality of life, and slowing of radiographic progression.
In the past decade, interest has centered on the “window of opportunity” concept, which purports a long-term advantage to early aggressive treatment of RA with DMARDs that might disappear if a theoretical critical time period in early disease is passed. Though gaining support from several clinical trials (e.g., the COBRA trial), this concept has yet to be solidly proven. A comparison of early vs. delayed DMARD treatment in patients with early RA (Ann Int Med 124(8):699, 1996) showed clear benefit of early DMARD use in terms of disability, pain, joint score, and ESR at one year of followup. At five years of follow-up in this same cohort, Verstappen and colleagues (Arthritis Rheum48(7):1797, 2003) seek to show whether this benefit is sustained.
Methods: 5-year followup data were obtained on 238 patients with RA of less than one year duration who had participated in a 1-year trial comparing DMARDs vs. “pyramid” approach. Patients were recruited from 1990 to 1994 in Utrecht, The Netherlands and randomized to one of four groups:
- Pyramid group – NSAIDs only. DMARDs could be started only if NSAIDs alone not clinically effective
- Aggressive DMARD group – IM gold or D-penicillamine
- Less Aggressive DMARD group – oral methotrexate (7.5-15 mg/week) or sulfasalazine (2-3gm/day)
- Least Aggressive DMARD group – hydroxychloroquine or auranofi
Patients were assessed every three months for the first two years and every 6 months for years 3-5 with measurement of ESR, Visual Analogue Score (VAS) for pain, joint score by Thompson articular index, VAS for general well-being, duration of early morning stiffness, and grip strength. Radiographs of hands and feet were obtained at baseline and every year thereafter and scored according to the Sharp method.
At one year, the DMARD was continued if 50% or greater improvement in 3 of 4 parameters (pain, joint score, duration in morning stiffness, and ESR) had been achieved. If not, a second DMARD was substituted. In patients with a complete response (defined as 3 out of the following 4: morning stiffness less than 15 minutes, VAS for Pain less than 10mm, joint score less than 10 (Thompson articular index), and ESR less than 30) the DMARD (or NSAID) was reduced and eventually stopped.
Results: 5 year data was available on 44 (79%) of the pyramid group and 145 (80%) of the combined early DMARD groups. Baseline characteristics were similar between groups with approximately 70% female subjects and 55-66% RF positivity. Mean ESR was approximately 40 for both groups. Mean total radiographic damage score for both groups was 4, reflecting early disease.
38 (86%) of patients in the pyramid group eventually received DMARDs with an average lag time of 14 months. Only 6 patients in the pyramid group did not receive any DMARDs. DMARDs chosen for NSAID failure in the pyramid group tended to be the more aggressive DMARDs. 98% of patients in the DMARD groups received NSAIDs over the study period, as well. In the DMARD groups, DMARDs were changed an average of 1.8 times. More patients in the pyramid group received oral and intraarticular glucocorticoids than in the early DMARD groups (30% and 41% vs. 19% and 19%, respectively).
64% of patient in the pyramid group and 66% of patients in the early DMARD groups obtained a complete response (see criteria for complete response above). In the last three years of the study, the number of patients not on any DMARDs was approximately 20-30% at any time and equal between the pyramid and early DMARD groups.
Improvements in ESR, Thompson joint score, VAS for general well-being and pain, morning stiffness, functional disability, and grip strength were significantly higher for the combined early DMARD groups in the first 21 months compared to the pyramid group. In the last three years of the study, however, scores for all of these parameters were equivalent in the two treatment groups. Furthermore, annual changes in Sharp score in the DMARD versus pyramid groups were 6.8 and 7.3 units per year respectively and were not statistically significantly different at any point in the five years of followup. A linear increase in Sharp score over time was noted for both groups.
Conclusions: The clinical improvement gained in the first year with use of DMARD therapy in early rheumatoid arthritis compared to the pyramid approach is not sustained if aggressive DMARD therapy is not maintained.
Editorial Comments: At first glance, the convergence in clinical outcomes and lack of a difference in radiographic progression between the pyramid and early DMARD groups would appear to invalidate the importance of early DMARD treatment and support the pyramid approach, even in the first year. However, the design of this study and the relatively low doses of DMARDs used during the period in which this study was initiated introduce some nuances in the interpretation of the data. Most importantly, DMARD treatment could be decreased or stopped in the early DMARD groups while those in the pyramid group were permitted to “catch up” by beginning DMARDs later in the study. It is not surprising therefore, with essentially a cross-over design, that no differences in the groups were observed at the end of the followup period. Also contributing to the “blur” between the two treatment groups were as follows: a) steroids (known to be disease modifying in RA) were used more liberally in the pyramid group; and, b) the doses (e.g., MTX) and types (e.g., IM gold and penicillamine) of DMARDs used in the DMARD group were relatively low and less efficacious than currently accepted standards for “aggressive” therapy.
This study cannot, therefore, either support or refute the concept of a “window of opportunity” for the early and aggressive treatment of RA. Nonetheless, the concept driving this study represented forward thinking and the authors must be commended for their efforts and analytic approach.