Objectives To study the retention rate of the anti-TNF biologics in the treatment of rheumatic diseases and the associated factors for drug withdrawal
Methods The Hong Kong Biologics Registry was established in 2005. Data on the use of biological agents for the treatment of rheumatological disorders in 14 public hospitals were collected prospectively. Efficacy data, serious adverse events (SAEs), withdrawal of biologics and reasons for withdrawal were collected at regular time intervals. This study focused on the retention rate of the anti-TNF biologics, which was studied by the Kaplan-Meier method. Factors associated with drug withdrawal were studied by the Cox proportional hazard models.
Results From December 2005 to September 2011, 864 courses of biological agents in 671 patients were reported to our Registry. 59 courses of treatment that involved the non-TNF biologics were excluded. There were 348 women and 323 men. The underlying rheumatic diseases were: rheumatoid arthritis (RA) (N=305, 46%), spondyloarthropathy (SpA) (N=274,41%), psoriatic arthritis (PSA) (N=83,12%) and others (N=9,1%). The choice of the anti-TNF biologics was: infliximab (IFX) (N=439,55%), etanercept (ETN) (N=274,34%) and adalimumab (ADA) (N=92,11%). The dosages of the biologics were: IFX (intravenous;3-5mg/kg for RA, 5mg/kg for SpA/PSA), ETN (subcutaneous; 50mg/week) and ADA (subcutaneous; 40mg/2 weeks). The mean duration of administration was 25±18 months. 358 courses of anti-TNF agents were abbreviated because of the following reasons: 106 (13%) lack/loss of efficacy (primary or secondary failure); 99 (12%) SAEs; 74 (9%) financial reasons; and 79 (10%) other/unspecified reasons. The cumulative drug withdrawal rate (due to either lack/loss of efficacy/AEs) at 12, 24 and 36 months was 24%, 33% and 38% for IFX, and 16%, 19% and 22% for ETN, respectively (log rank test;p<0.001). When drug withdrawal was broken down into that due to loss of efficacy and AEs, data remained in favor of ETN to IFX. The commonest reasons for drug withdrawal because of adverse events were: infusion/injection site reactions (3.4%) and tuberculous infection (3.1%). Among the 21 patients who developed tuberculosis, 19 patients were IFX users whereas only 2 were ETN users. Regarding the underlying rheumatic diseases, RA patients had lower drug retention rate than SpA or PSA patients but the difference did not reach statistical significance (log rank test;p=0.07 between RA and AS). Cox regression analysis revealed that the anti-TNF agent used (IFX versus ETN/ADA; HR1.91[1.38-1.63];p<0.001) and the female sex (HR 1.78[1.07-2.95];p=0.03) was independent predictors for drug withdrawal due to inefficacy after adjustment for age, disease duration and underlying diagnosis (RA vs non-RA). In another Cox regression model with serious adverse events being the reason for drug withdrawal, only the choice of drug itself (IFX versus ETN/ADA; HR1.94[1.23-3.06]; p=0.005) was significantly associated.
Conclusions There exists a difference among the anti-TNF biologics for drug retention in the long run. Our data reveals that IFX is associated with a significantly higher withdrawal rate for both loss of efficacy over time and the development of serious adverse events, in particular tuberculosis.
Disclosure of Interest None Declared