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.
Serious Infection in Rheumatoid Arthritis Patients Switching Anti-Tumor Necrosis Factor Drugs.
Jones2, Judith K., Napalkov1, Pavel, Dong1, Wei, Alexander1, Kimberly, Goehring Jr2, Earl L., Nguyen-Khoa2, Bao-Anh
Rheumatoid arthritis (RA) patients commonly change drug therapy for reasons of safety or effectiveness. This study describes rates of first serious infection among RA patients who switch from one anti-Tumor Necrosis Factor (aTNF) drug to another aTNF compared to those who do not switch.
Subjects with RA who received only an aTNF drug (infliximab, etanercept, or adalimumab) were observed in a large health claims database from 1/1/2001 to 12/31/2007. Non-switchers (NS) remained on one aTNF during the study period, Switchers (S) had at least one change to another aTNF. Index dates were defined as the day of the first aTNF claim for NS, and the day of the treatment switch for S. Baseline data were collected for the 365 days of enrollment prior to the index date. Serious infections included those which required IV antibiotics or hospitalization; only the first event was included in incidence rate estimates. Two attributable risk periods were used: 1) infection occurring <= 90 days from a prior claim for an aTNF (90-Day) and 2) any infection occurring after the index date (Ever-Treated). Data were stratified by 1-year and >= 2 years post-index. Cox regression was used to compare serious infection rates, adjusting for age, gender, selected comorbidities, Charlson comorbidity score, hospitalizations, and other RA treatments.
There were 13,752 RA patients in the NS cohort, and 2,293 in the S cohort. In the 90-Day model, unadjusted rates of first serious infection was nonsignificantly lower for NS vs S (6.31/100 PY, 95% CI: 6.016.62 vs. 6.78/100 PY, 95% CI: 5.957.67). Rates of first serious infections declined from 8.59/100 PY and 8.72/100 PY in the first year post-index for NS and S, to 2.66/100 PY and 2.64/100 PY >= 2 years post-index.
In the Ever-Treated model, NS also had nonsignificantly lower unadjusted first infection rates than S (8.45/100 PY 95% CI: 8.108.80 vs. 9.10/100 PY, 95% CI: 8.1510.12). Rates of first serious infection declined from 10.15/100 PY and 10.11/100 PY in the first year post-index for NS and S, to 4.18/100 PY and 4.44/100 PY >= 2 years post-index.
Regression analysis showed no significant difference between NS and S cohorts in the risk of serious infection for either attribution model (90-Day HR=0.93, 95CI: 0.741.17; Ever Treated HR=0.94, 95CI: 0.781.15). First and second year rates were not different between NS and S. Significant predictors for increased risk of serious infection included: age >= 50 years; positive history of serious or opportunistic infection, diabetes, respiratory disease; baseline Charlson score >= 2, or increasing number of hospitalizations. The effect of methotrexate on infection was conditionally related to several other risk factors.
Conclusions The risk of a serious infection was not different between RA patients that switched aTNF drugs and those who did not. Indications for switching an aTNF drug could not be captured in this study. Rates in the Ever-Treated model were generally higher than the 90-Day model for overall, 1-year, and >= 2 years post-index. Regardless of switching status, the rate of first infection 1-year post-index was up to 3 times greater than 2+ years post-index.
To cite this abstract, please use the following information:
Jones, Judith K., Napalkov, Pavel, Dong, Wei, Alexander, Kimberly, Goehring Jr, Earl L., Nguyen-Khoa, Bao-Anh; Serious Infection in Rheumatoid Arthritis Patients Switching Anti-Tumor Necrosis Factor Drugs. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :698