Background/Purpose: Treatment
with biologics can elicit unwanted immune responses such as antidrug antibodies
(ADA), which may decrease their clinical efficacy and increase adverse events. However,
use of different assay methods and inconsistent sample collection times has
made direct comparison difficult.

To assess incidence of ADA for etanercept (ETN), adalimumab (ADL),
and infliximab (IFX), and to correlate it with trough drug concentration, efficacy,
and patient health outcomes in a real world setting.

Methods: This
non-interventional, cross-sectional study was conducted in 57 centers across Argentina,
Australia, Bulgaria, Turkey and the United States. Adult patients with
rheumatoid arthritis, treated with ETN, ADL, or IFX continuously for 6–24
months were eligible. ADA, using RIA, and trough drug concentrations, using
ELISA, were measured in samples collected on one occasion ≤2 days prior to
the next scheduled dose. All assays were performed at the same laboratory. Efficacy
was measured as Disease Activity Score 28-joint count (DAS28), Clinical Disease
Activity Index (CDAI), Simplified Disease Activity Index (SDAI), C-reactive
protein (CRP) levels, erythrocyte sedimentation rate (ESR) and health outcomes
measures, eg, HAQ-DI, EQ-5D, SF-36. History of injection site or infusion reactions,
serum sickness, and thromboembolic events while on the current biologic were
also assessed.

Results: A total of 595 eligible
patients (ETN: n=200; ADL: n=199; IFX: n=196) were enrolled in the study.
Baseline demographics were similar across all groups. The mean duration of
treatment was 14.6, 13.5, and 13.1 months for ETN, ADL, and IFX, respectively. None
of the patients receiving ETN tested positive for ADA, whereas 62 (31.2%) and
34 (17.4%) of patients treated with ADL and IFX, respectively, tested positive
for ADA (p<0.0001). In ADL- or IFX-treated patients, those with ADA had
significantly lower trough drug levels. Efficacy and health outcomes are
summarized in the table. DSR28-ESR LDA and remission rates are higher in
patients without ADA. There was a correlation noted in objective measurements
of inflammation in patients with ADA. All 3 treatments were well tolerated with
regard to reporting targeted medical events.

Conclusion: ADA were not detected in ETN-treated patients. Statistically
significant higher proportion of patients in the ADL/IFX treatment groups was
positive for ADA. Patients with ADA had lower serum concentration of the drug
and showed a higher value for CRP and ESR. Patients without ADA showed statistical
differences in clinical and health outcomes between groups compared with
patients with ADA. This is a cross-sectional study recruiting patients
currently on continuous treatment of tumor necrosis factor α inhibitors. A
prospective longitudinal study may be needed to further evaluate the impact of
ADA on efficacy.

Table