Article Text

PDF
FRI0168 Drug discontinuation in 6,657 patients with ra starting their first tnf inhibitor: Comparison of adalimumab, etanercept & infliximab
  1. M. Neovius,
  2. E. Arkema,
  3. H. Olsson,
  4. J. Eriksson,
  5. J. Simard,
  6. J. Askling
  7. and ARTIS
  1. Dept of Medicine, Karolinska institutet, Stockholm, Sweden

Abstract

Background Biologics are likely to perform differently in different patient segments and between-drug differences may exist. Changing patient populations, practice and treatment alternatives may have led to calendar period differences in drug survival within and between TNF inhibitors.

Objectives To compare drug discontinuation rates on adalimumab, etanercept, and infliximab in patients with RA overall and by calendar period.

Methods Using the Swedish Biologics Register (ARTIS), adult patients with RA (n=6,657; 76% women; mean/median age 55/57y, HAQ 1.2/1.1, DAS28 5.2/5.3) initiating TNF inhibitor therapy between 2003 and 2009 were identified. Data on drug initiation and discontinuation of first anti-TNF prescription were collected through Dec 31, 2010. Drug discontinuation rates over a maximum of 5 years of follow-up for adalimumab, etanercept, and infliximab were compared overall and by calendar period (2003-2005/2006-2009; n=3,019/3,638). Adjustment was made for age, sex, education level, period, baseline HAQ, concomitant treatment (DMARD and steroids), and general frailty (using outpatient visits and hospital days in the previous two years as proxies).

Results Over 13,808 person-years [pyrs; mean/median 2.1/1.6y] of follow-up, 3,141 patients discontinued their first line biologic (23/100pyrs; 43% due to inefficacy and 32% to adverse events). In adjusted analyses for the full period, infliximab (hazard ratio [HR] 1.7, 95%CI 1.5-1.8) and adalimumab (1.4, 95%CI 1.2-1.5) had higher discontinuation rates than etanercept, while infliximab had higher discontinuation rate than adalimumab (1.2, 95%CI 1.1-1.4; Figure). These findings were consistent across periods, but varied with HAQ and follow-up time (difference between etanercept and adalimumab present only during the first year of follow-up). The overall discontinuation rate was lower for starters in the 2003-2005 than the 2006-2009 period (crude rate 20 vs 27/100pyrs; adjusted HR 0.8, 95%CI 0.8-0.9). This held true also for each individual drug. The composition of 1y discontinuations also changed: discontinuations due to adverse events decreased from 40% to 32%, while inefficacy increased from 38% to 47% (both p<0.001).

Conclusions We found greater drug discontinuation rates in infliximab vs both adalimumab and etanercept, and adalimumab vs etanercept. The differences existed in both periods investigated, but varied with follow-up time and subgroup.

Disclosure of Interest M. Neovius Grant/Research support from: This research received funding from Astra Zeneca through the Swedish public-private-partnership program on chronic inflammation (COMBINE), E. Arkema: None Declared, H. Olsson: None Declared, J. Eriksson: None Declared, J. Simard: None Declared, J. Askling: None Declared

Statistics from Altmetric.com

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.