Patients with early inflammatory arthritis not meeting classification criteria for rheumatoid arthritis (RA) may have treatment delayed until typical phenotypic features accumulate. Due to the concern for toxicities of DMARDs, patients are often treated with NSAIDs and simple analgesics during this period. However, delay in DMARD treatment in early RA has been shown to result in poorer outcomes, such as irreversible radiographic damage. Addressing this problem, van Dongen et al (Arthritis Rheum 2007; 56(5):1424) report the results of the “PRObable rheumatoid arthritis: Methotrexate vs. Placebo Treatment” (PROMPT) study.
DMARD naïve patients with undifferentiated inflammatory arthritis (UA) with symptom duration of less than two years were randomized to receive methotrexate (MTX) or placebo. MTX was initiated at a dose of 15 mg per week and titrated to 30 mg per week in 5 mg increments every 3 months if the DAS was greater then 2.4 (low disease activity). Patients continued on background NSAIDs. Enrollment occurred over a five year period (2001- 2006) at four hospitals in the Netherlands. Patients were followed for 30 months with the primary outcome being diagnosis at the end of the study interval (incident RA, maintenance of UA, or remission). Once subjects achieved the primary endpoint, fulfillment of ACR 1987 classification criteria for RA, patients were treated with open-label MTX. Radiographic progression after the first 18 months of treatment was compared according to treatment allocation.
One-hundred ten subjects were randomized equally to receive MTX or placebo. One-hundred subjects completed the 30 months study interval (50 in each treatment group). Two-thirds of subjects were female with a median age of 51 years. The median duration of symptoms was less than one year in both the MTX and placebo groups. Approximately one-third of subjects were seropositive for rheumatoid factor and approximately one-fourth were seropositive for anti-CCP antibodies. Median swollen joints counts at baseline were low, 3 in the MTX group and 2 in the placebo group. Only 5 subjects (5%) had radiographic erosions at enrollment, with no significant difference between treatment groups.
At the 30 month clinical endpoint, the proportion of subjects meeting classification criteria for RA was 53% in the placebo group vs. 40% in the MTX group, yielding a relative hazard of RA of 1.7 (p=0.04). At the 18 month radiographic endpoint, most patients still showed no radiographic progression. However, among those showing radiographic progression, 6 (30%) were treated with MTX vs. 14 (70%) treated with placebo (p-value for comparison = 0.046).
Subgroup analyses demonstrated differential responses based on anti-CCP antibody seropositivity. Anti-CCP antibody positive subjects benefited from MTX therapy, such that progression to clinically defined RA occurred in 93% of the placebo treated anti-CCP positive patients compared to only 67% of the MTX treated anti-CCP positive patients, yielding a relative hazard of RA of 4.9 for this sub-group (p < 0.001). Progression to RA was not as frequent in the anti-CCP antibody negative sub-group and, interestingly, was not different if treated with MTX or placebo (HR 1.3; p=0.51).
Adverse events and serious adverse events were not different between MTX and placebo treated groups.
MTX therapy significantly reduces the time to classification of definite RA and radiographic progression in patients with early, undefined inflammatory arthritis. The protective effect of MTX was greatest in subjects seropositive for anti-CCP antibodies.
Much recent debate has focused on whether a “window of opportunity” exists in early RA whereby early aggressive intervention is beneficial. There remains some skepticism, primarily related to how the “window” is defined. If the definition of “window” is that early treatment, even of vague undifferentiated synovitis, leads to improvements in clinical signs, symptoms, and results in less radiographic damage, then this study confirms previous studies and extends it to patients with very limited disease at onset. However, if the definition of window is that early treatment permanently alters the trajectory of disease such that the momentum of disease is different than if not treated, then these results may not fully convince. This is not an indictment on this study (which is novel and well designed and implemented), as proving the second definition of “window of opportunity” is exceedingly difficult, as it requires some estimate of what a patient’s disease course would have been had they not been treated. Unfortunately, making these projections early in disease is quite difficult, particularly as a small but significant number of patients will actually spontaneously remit, even without treatment.
Putting these issues aside, this study has the potential to contribute to clinical practice in several ways. For one, it confirms the safety and efficacy of aggressive DMARD treatment in those with meager symptomatology. Second, it identifies a susceptible subgroup, those with anti-CCP antibodies, who appear to benefit most from this therapy. Longer term follow-up is required to see if the benefits of early therapy in these patients are sustained and whether other DMARDs are similarly effective as MTX.