Background/Purpose: Observational studies are critical in assessing medication safety and effectiveness in the real world. Nonrandom assignment can provide insight to how and when medications are prescribed. Since the US introduction of TNF inhibitors (TNFi) in 1998, several newer biologics with varying mechanisms of action are available to patients and physicians. Our objective was to compare characteristics and serious adverse events (serious infections, malignancies and autoimmune diseases) of patients who received abatacept (ABA) compared to those who received other biologic DMARDs (BDC).

Methods: Participating patients with RA in the National Data Bank for Rheumatic Diseases (NDB) provided treatment and other characteristics (sociodemographics and comorbidities) through self-reported biannual questionnaires. Only responses from 2005-2015 were used for greater comparability. Initiators of ABA were matched to patients initiating other DMARDs in a 1:3 ratio. All patient-reported outcomes were verified by medical record review except non-melanoma skin cancer. For cancer and autoimmune outcomes, ITT analyses were performed, holding the exposure constant from the start of treatment until the end of follow-up, and for infections, through its duration plus 90 days. Marginal structural models were used to estimate the effect of treatment on the outcome by an appropriate control for the effects of time-dependent confounders using stabilized weights. Variables included sex, employment, smoking, education, income, BMI, insurance, comorbidity index, RA severity, and co-medications. These results are reported as hazard ratios (HR), estimated through a pooled logistic regression. The analysis was also stratified by prior number of exposed biologics. The primary comparison group were RA patients starting biologic DMARD (BDC), and we also compared with non-biologic DMARD (DC).

Results: A total of 1456 RA patients initiated ABA and 3222 a BDC during the study, representing total patient years of drug exposure (total follow-up time) of 2396.6 (4613.9) and 8576.5 (10462) respectively. The ABA patients were older, had longer RA duration, worse disease severity, more comorbidities and more exposure to prior DMARD use (Table 1). HRs are presented in Table 2. The results were similar when comparing with those initiating DC (1663 patients with 4340.0 patient years of drug exposure, 5215.7 follow-up time). Table 1. Baseline mean (SD) characteristics of ABA and biologic comparison cohorts.

Table 2. Hazard ratios (95% CI) of incident events comparing ABA with other biologics.

Conclusion: We found a trend of channeling with ABA, where patients starting ABA had worse disease outcomes and higher prior use of DMARDs. After adjustment there appeared to be less difference in the adverse events studied for both comparisons cohorts with exception of increased lupus co-diagnosis and statistically significant decreased hospitalized infections for ABA.