Dose escalation of infliximab in clinical practice: improvements seen may be explained by a regression-like effect.

OBJECTIVE To determine whether increased infliximab doses result in better clinical outcome in rheumatic diseases. METHODS Subjects were 124 patients with rheumatoid arthritis treated with biological agents at a single institute. Index cases were 44 patients whose infliximab doses had been increased. Controls were patients treated with infliximab without dose increase (n = 44), and patients treated with etanercept (n = 36). Disease activity score (DAS28), ACR28 swollen joint counts, and numerical ACR responses were compared before and after dose increases. For the controls, the point at which the DAS28 value showed any increase (despite infliximab/etanercept treatment) was used as the reference time point. Comparisons were made between three sets of outcomes: best outcome achieved before the dose increase (cases) or before the reference time point (controls); outcomes at this point; and best outcomes after this point. RESULTS Following dose increase, disease activity showed modest but statistically significant improvements. The improvement achieved after dosage escalation was equal to, but not better than, the best values before dose escalation. While this finding could be interpreted as "recapturing" the previous response, similar improvements were seen in both control groups. Thus the same pattern of worsening and subsequent improvement was seen with or without the infliximab dose increase. CONCLUSIONS Clinical improvement with increased infliximab dose, and the impression that a previous response can be "recaptured" with higher doses, cannot be taken at face value, as similar improvements occurred in two control groups. The use of infliximab at doses higher than 3 mg/kg needs to be evaluated further.