Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

Baseline Factors Related to Response to Conventional DMARD Treatments.

Sterling1,  Kimberly L., Carter1,  Gebra Cuyun, Messing6,  Susan, Reed5,  George, Chen6,  Rui, Kaukeinen6,  Kimberly, Tu6,  Xin

Eli Lilly and Co, Indianapolis, IN
Eli Lilly and Co
Eli Lilly and Co, Indianapolis, IN
New York University School of Medicine, New York, NY
University of Massachusetts, Amherst, MA
University of Rochester, Rochester, NY


Heterogeneity of treatment response exists in rheumatoid arthritis (RA), while there is a lack of understanding of genetic and non-genetic factors that predict treatment response. Identifying baseline factors associated with patient response to RA treatment will assist tailoring therapies for patients. This study aimed to determine non-genetic baseline factors associated with treatment response at one year among RA patients initiating conventional disease-modifying antirheumatic drugs (DMARD) therapy with no previous biologic use.


The study included RA patients previously naïve to biologic therapy who initiated a conventional DMARD and had at least one year of follow-up in the Consortium of Rheumatology Researchers of North America (CORRONA) registry. Treatment response was assessed using the Clinical Disease Activity Index (CDAI) at one year (±3 months), evaluated in both a continuous (change from baseline to endpoint) and categorical (endpoint score <=19.6 and improvement from baseline > 4.3) manner. The registry allowed for exploration of a broad set of baseline factors including socio-demographics, clinical factors, patient reported measures (i.e. behaviors, comorbidities, function), and prior healthcare use. Multivariate stepwise regression using backward elimination was conducted to determine variables significantly associated (p<0.05) with improvement.

Summary of the Results:

A total of 1,350 patients met the entry criteria, and 893 (66%) of those had complete data and were included in the multivariate analyses. Of those with complete data, the mean age was 59.7 years, 662 (74%) were female, 750 (84%) were white, and the mean initial CDAI score was 14.5. The following baseline factors were associated with greater CDAI improvement (continuous) (R2=0.39): higher initial CDAI score (b=-0.598, p<0.0001), lower BMI (b=0.086, p=0.05), lower modified Health Assessment Questionnaire (mHAQ) score (b=3.738, p<0.0001), no history of infection (b=-1.336, p=0.03), no use of analgesics (b=1.303, p=0.04), and not being disabled (b=-3.839, p=0.0002). For the categorical CDAI improvement outcome, the odds of improvement in disease activity is greater for those at baseline with a higher initial CDAI score (odds ratio [OR]=1.101, p<0.0001), younger in age (OR=0.986, p=0.03), light to no alcohol consumption history (moderate/heavy vs none/light, OR=0.302, p=0.01), and are not disabled (yes vs no, OR=0.413, p=0.002) (R2=0.22).


This study identified that higher baseline disease activity and not being disabled were consistently associated with treatment response (continuous and categorical CDAI). Additional exploration is needed to understand the predictive role of variables such as BMI, physical function, history of infections, and analgesic use, which were inconsistently associated with response; and differences in sets of risk factors between medication classes (e.g. conventional vs biologic DMARDs). With a broad set of predictors, clinicians will be able to tailor treatment based on baseline characteristics to improve patient outcomes.

To cite this abstract, please use the following information:
Sterling, Kimberly L., Carter, Gebra Cuyun, Messing, Susan, Reed, George, Chen, Rui, Kaukeinen, Kimberly, et al; Baseline Factors Related to Response to Conventional DMARD Treatments. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :52
DOI: 10.1002/art.27821

Abstract Supplement

Meeting Menu