Efficacy assessment in trials of combination therapy for rheumatoid arthritis

The comparison of a combination disease modifying antirheumatic drug (DMARD) regimen with a single DMARD, or one combination with another, raises the same issues encountered when studying rheumatoid arthritis: how to design, conduct, and analyze randomized controlled trials. However, these claims of comparison are unique in that the setting to show primary efficacy is the same as that showing the primary therapeutic placement relative to standard therapy. Ordinarily standard therapy is not determined prior to approval. Accordingly, any combination DMARD trial presupposes agreement on what constitutes standard therapy and, if the intent is to show equivalence, on what (small) difference can be condoned to grant the claim. I address some important considerations regarding the rigor and credibility of designs for these claims, problems far less significant with simple drug vs placebo strategies. These can be grouped into 3 categories: (1) choice of controls/designs/patients to ensure a fair comparison; (2) sufficiently broad design formulations to yield controlled assessments of safety that are as thorough as those for efficacy in the past, and (3) maintaining blinding despite new design features that threaten it.