Introduction: Reasons for infliximab failure in Crohn's disease and ulcerative colitis are debated. Serum levels of infliximab and anti-infliximab antibodies have been associated with loss of response. We aimed at determining cut-off levels for infliximab and anti-infliximab antibody concentrations associated with clinical response to infliximab maintenance therapy.
Methods: Patients with inflammatory bowel disease (n = 106) were retrospectively classified as having maintained response or loss of response to infliximab maintenance therapy. Trough concentrations were measured by fluid-phase radioimmunoassays.
Results: Infliximab levels were significantly lower, and anti-infliximab antibody levels significantly higher, in Crohn's disease patients with loss of response (median infliximab 0 μg/ml, median anti-infliximab antibodies 35 U/ml) compared to patients with maintained response (median infliximab 2.8 μg/ml, median anti-infliximab antibodies 0 U/ml; p < 0.0001). Receiver operating characteristic (ROC) analysis identified optimal cut-off values: infliximab <0.5 μg/ml, which was associated with loss of response with sensitivity 86% [64-97] and specificity 85% [72-94]; and anti-infliximab antibodies ≥10 U/ml yielding a sensitivity of 81% [61-93] and specificity 90% [79-96]. Combined measurements of infliximab and anti-infliximab antibodies using these cut-off values had higher accuracy yielding a sensitivity of 81% [57-94] and specificity 94% [82-98]. Similar pattern was observed in a smaller cohort of patients with ulcerative colitis.
Conclusions: Combined measurements of infliximab and anti-infliximab antibodies using cut-off levels provided high accuracy for discriminating between clinical response types to infliximab maintenance therapy. Cut-off levels are considered a prerequisite to further investigations of clinical usefulness of measurements of infliximab and anti-infliximab antibodies in patients failing infliximab therapy.