In contrast to many previous trials of biologic agents, which, with unbalanced designs, have compared the combination of a biologic agent plus methotrexate (MTX) with placebo plus MTX, the Treatment of Early Aggressive Rheumatoid Arthritis (TEAR) study correctly compared the combination of a biologic agent plus MTX with a combination of disease-modifying antirheumatic drugs (DMARDs) in a double-blind trial (1). Apart from a small difference in total Sharp score (2), no differences in outcomes between the 2 treatments were reported.
Similar results have been found in previous related studies. In 2005 the open BeSt (Dutch acronym for Behandel-Strategieën, “treatment studies”) study showed that a combination of 2 DMARDs plus an initial temporary course of glucocorticoid is as effective as infliximab plus MTX in the treatment of rheumatoid arthritis (3). Furthermore, in a recent meta-analysis we showed by means of indirect comparisons that inhibition of the radiographic progression rate was similar in 3 studies comparing 3 DMARDs with a single DMARD and 12 studies comparing the combination of a biologic agent plus MTX with a single DMARD (79% versus 84%) (4). In general, our results were in accordance with those of another contemporary meta-analysis (5). Recently, results of a 2-year followup assessment in an open, randomized trial showed that effects of triple DMARD therapy and of infliximab plus MTX were similar (6).
According to the Results section in the report by Moreland et al, the etanercept plus MTX group had a borderline significantly smaller increase in total Sharp score compared with the triple therapy group (0.64 versus 1.69; P = 0.047). On the basis of the results shown in Table 2, we recalculated the total Sharp score and had a similar, though statistically insignificant, result (0.67 versus 1.57; P = 0.47). The difference is probably because we did not have data on individual patient differences in the radiographic scores at our disposal, which therefore may have exaggerated the standard deviations in our calculations. Our calculations also show that there was no difference in erosion score between the 2 groups (0.56 versus 0.70; P = 0.8). We suggest that the authors provide a table showing the mean change in total scores and erosion scores and standard deviations for the 4 groups (0–2 years). As the number of patients without progression exceeded 50% in all 4 groups, it could be deduced that the median progression is 0 in all 4 groups. Probably the radiographic data are not normally distributed, and our guess would be that a nonparametric statistical test would show no difference in the radiographic data between the etanercept plus MTX group and the triple therapy group. Statistical significance or not, it seems fair to conclude that the numerical differences concerning radiographic progression are unimportant and that the radiographic data are in accordance with the remaining results of the TEAR study.
Consequently, the TEAR study, with its very strong design, is very important as it confirms the results of previous studies and adds credence to the conclusion that a combination of DMARDs is as efficient as the combination of a biologic agent plus a single DMARD, and expensive biologic agents should therefore be reserved for patients who have an inadequate response to or cannot tolerate a combination of DMARDs.