RESUMO
OBJECTIVE: The immunogenicity of abatacept, a selective costimulation modulator, administered intravenously, was assessed across Phase II and III trials in patients with rheumatoid arthritis (RA). METHODS: Two direct-format enzyme-linked immunosorbent assays evaluated antibody responses [whole abatacept molecule (CTLA-4 and Ig portion) and CTLA-4 portion only (Assay A)] in the Phase II trials. During the Phase III trials and 2-year open-label periods, a similar, but more sensitive, Assay B was employed. Serum samples collected prestudy, during treatment, and 56 and/or 85 days following the last dose were evaluated. Seropositive samples with anti-CTLA-4 reactivity and sufficiently low drug levels were further characterized for neutralizing activity (cell-based bioassay). RESULTS: A total of 2237 patients with both pre- and post-baseline serum samples were eligible for assessment. Of these, 62 (2.8%) patients demonstrated an anti-abatacept or anti-CTLA-4 response, determined using either Assay A or B. Using the more sensitive Assay B, 60 of 1990 patients (3.0%) demonstrated an antibody response to the whole abatacept molecule (n = 41, 2.1%) or the CTLA-4 portion (n = 19, 1.0%). Of the 1764 RA patients evaluated in the Phase III studies, 203 discontinued therapy and had sera collected 56 and/or 85 days after discontinuation. Patients who discontinued had a higher incidence of immunogenicity versus patients who did not discontinue (7.4% vs 2.6%, respectively). Of 20 patients positive for anti-CTLA-4 reactivity, 13 were eligible for assessment with the neutralization bioassay. Of these, 8 patients exhibited neutralizing activity. Seroconversion occurred with no adverse safety outcomes or effect on pharmacokinetic parameters. No consistent pattern was observed between antibody response and loss of efficacy (American College of Rheumatology 20 and Health Assessment Questionnaire responses). CONCLUSION: Abatacept was associated with a low incidence of immunogenicity in patients with RA and lacked any adverse sequelae.
