Between 1989 and 1995, a series of clinical trials of anti-CD4 monoclonal antibody (MAb) in the treatment of patients with rheumatoid arthritis (RA) purportedly resulted in positive clinical benefits in 8 of the studies, which were open trials involving a total of 129 patients (1-8), and no clinical benefit in 3 of the studies, which were blinded, placebo-controlled trials involving a total of 133 patients (9-11).

In response to the studies that resulted in no clinical benefit, those investigators who have remained convinced that some form of biologic anti-T helper cell therapy is a promising road to effective treatment of RA have focused on the origin and specificity of the MAb used in these clinical trials (12). Some have commented on the dose per infusion and spacing of infusions as a possible explanation for the divergent results (6, 13), while others have shifted their focus from CD4 to tumor necrosis factor (TNF)(14) and other cellular targets. Analysis of the various anti-CD4 MAb trial reports seems to suggest that an equally plausible explanation for the different results may be the occurrence of expectation bias on the part of both investigators and subjects in the open trials; ie, awareness that a new drug is being administered imparts an expectation of