Background The efficacy of Golimumab (GLM) treatment in rheumatoid arthritis (RA), psoriatic arthritis (PsA) and ankylosing spondylitis (AS) has been widely documented. In everyday clinical practice, however, many patients discontinue therapy due to adverse events (AE), lack of efficacy or other reasons. 
Objectives To compare first year retention rates of GLM treatment among RA, PsA and AS, using the first one as reference standard, in a multicentric observational cohort (the LORHEN Registry).
Methods All patients in the LORHEN database who started GLM were included. All patients were treated according to current EULAR recommendations for the management of RA and spondyloarthritis. Drug survival during the first year of treatment was measured, along with specific reasons for discontinuation (inefficacy or adverse events). We compared drug retention rates using the Kaplan-Meier method. Cox regression analyses, using RA as reference category, were used to adjust for age, sex, disease duration, number of previous csDMARDs and treatment with low dose prednisone.
Results One hundred and thirty-four RA patients, 84 PsA patients and 108 AS patients were included. 49 (37%) RA patients, 29 (35%) PsA patients and 22 (20%) AS patients discontinued treatment during the first year. Thirty-six (27%) and 12 (9%) RA patients, 17 (20%) and 11 (13%) PsA patients, 14 (13%) and 7 (6%) AS patients discontinued GLM due to lack of efficacy or AE, respectively.Patients with a diagnosis of AS showed a lower, but not significant, risk of discontinuation, with an adjHR (95%CI) of 0.35 (0.10 – 1.17).Patients treated with low dose prednisone showed a reduced risk of discontinuation with an adjHR (95%CI) of 0.35 (0.14 – 0.86). Age, sex, disease duration and number of previous csDMARDs did not significantly influence the risk of discontinuation.
Conclusions AS patients seem to have better GLM retentions rates with respect to RA and PsA patients; however this difference is significantly reduced after adjusting for confounders.
The significantly lower adjHR observed for low dose prednisone therapy might reflect a beneficial effect even in patients treated with bDMARDs.
Scirè CA et al. Clin Exp Rheumatol. 2013;31:857-63.
Disclosure of Interest None declared