Predictors of response to rituximab in patients with active rheumatoid arthritis and inadequate response to anti-TNF agents or traditional DMARDs.

Objectives: Identifying early predictors of response to biological agents is important for both the individual patient and health economics. The aim here was to identify clinical variables that are easily assessed in clinical practice which are associated with a major response to rituximab (moderate to good EULAR response, according to DAS28 values) in patients with active rheumatoid arthritis and inadequate response to anti-TNF agents or traditional DMARDs.

Methods: Rituximab (2x1g, two weeks apart) was administered to 108 patients in four different Spanish hospitals. The primary efficacy endpoint was the percentage of patients who achieved a major response at six months. Potential predictors of a major response were identified using multivariate binary logistic regression models.

Results: At six months of treatment 75.9% of patients achieved a major response (24% good and 52% moderate). Comparing the clinical features at baseline between patients who did or did not achieve a major response, significant differences were found in rheumatoid factor (RF) and anti-CCP positivity, as well as in the number of failed anti-TNF agents prior to rituximab. While rituximab delivers clinical benefit in seronegative patients, the presence of RF and/or anti-CCP consistently enriches clinical responses. The multivariate analysis showed that the best model for predicting a major EULAR response to rituximab was comprised of the following two variables: the anti-CCP antibody positivity (p=0.045) and the number of previous anti-TNF agents used (p=0.028). Using a cut-off level for CCP of 300 U/ml we found that patients with an anti-CCP titre >300 U/ml were 3-4 times more likely to achieve a major EULAR response [odds ratio (OR): 3.38; 95% CI: 1.025-11.17]. By contrast, those patients who had failed to respond to 2 or more anti-TNF agents had a 72.5% lower probability of achieving a moderate to good EULAR response (OR: 0.275; 95% CI: 0.087-0.871) than did patients who had only failed to respond to one such agent.

Conclusions: A lower number of previously-failed TNF blockers and high anti-CCP titre can help select the best candidates for RTX therapy in patients with RA.