Article Text

AB1105 Epidemiological Assessment of Baseline Characteristics of the Abatacept-Treated Patients in Select Real-World Rheumatoid Arthritis Data Sources
  1. T. Simon1,
  2. K. Ghandi1,
  3. J. Askling2,
  4. F. Wolfe3,
  5. K. Michaud4,
  6. V. Thyagarajan5,
  7. N. Lin6,
  8. P. van Riel7,
  9. M. van de Laar8
  1. 1Bristol-Myers Squibb, Hopewell, United States
  2. 2Karolinska University Hospital Solna, Stockholm, Sweden
  3. 3National Data Bank for Rheumatic Diseases, Wichita
  4. 4University of Nebraska Medical Center, Omaha
  5. 5Optum Epidemiology, Ann Arbor
  6. 6Optum Epidemiology, Waltham, United States
  7. 7Radboud University, Nijmegen
  8. 8Medisch Spectrum Twente, Enschede, Netherlands


Background Registries in North America and Europe have reported that patients with RA initiating abatacept are in poorer health, more likely to be previously exposed to several biologics and to have more comorbidities than initiators of TNF inhibitors.1,2 This along with the need to determine the optimal comparison cohort for post-marketing safety studies, warrants a more comprehensive evaluation of the baseline characteristics of the abatacept-initiators.

Objectives To present baseline characteristics of patients receiving abatacept among RA data sources used to support the ongoing post-marketing safety evaluation of abatacept.

Methods At present, four registries and one administrative claims database contribute data on patients with RA treated with abatacept to the post-marketing safety program.

Baseline demographics and clinical characteristics for patients with RA are collected and recorded in each of the data sources. These data are provided to Bristol-Myers Squibb (sponsor) annually as part of the post-marketing regulatory commitment. Comparison groups of non-biologic DMARD users and biologics-treated patients (other than abatacept) are also provided from 2 of the 5 registries.

Results Through July 2013, a total of 3,278 abatacept-treated patients with ∼7,600 person-years of cumulative follow-up have been reported from the 5 data sources. Abatacept users were similar in age and gender distribution across the different registries. Greater than 80% of all abatacept users across all registries were previously exposed to one or more biologics (median =2; see table). However the proportion of abatacept users previously exposed to one or more biologics was slightly lower (48%) in the claims database and may be attributable to shorter follow-up period.

Conclusions Our evaluation of baseline characteristics shows that most abatacept-treated patients have been exposed to at least 2 biologics prior to initiating abatacept. In addition, abatacept-treated patients in these data sources have longer disease duration when compared with the IV abatacept trials3 and published data.1,2 These data suggest that careful evaluation of baseline characteristics in abatacept real world-users and other real world-treated biologic and non biologic patients is necessary to determine the optimal comparison cohort prior to evaluating real-world safety outcomes.


  1. Michaud K et al. Ann Rheum Dis 2013;72(Suppl3):458.

  2. Askling J et al. Ann Rheum Dis 2013;72(Suppl3):455.

  3. Weinblatt ME et al. J Rheumatol. 2013;40:787-97.

Disclosure of Interest T. Simon Employee of: BMS, K. Ghandi Employee of: BMS, J. Askling Consultant for: Wyeth, Schering-Plough, Abbott, Speakers bureau: BMS, F. Wolfe: None declared, K. Michaud: None declared, V. Thyagarajan: None declared, N. Lin: None declared, P. van Riel Consultant for: Roche, M. van de Laar: None declared

DOI 10.1136/annrheumdis-2014-eular.2529

Statistics from

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.