Background Treatment guidelines for the management of RA recommend sequential use of biologic therapies, and the majority of patients switch from one anti-TNF agent to another. Previous data from commercial claims databases have shown that patients on a second-line anti-TNF agent are more likely to switch biologic treatment than those on second-line abatacept, a biologic with a different mechanism of action.¹ This finding has not been studied in a Medicare population.

Objectives To compare subsequent switch rates of patients in a Medicare population who had switched between anti-TNFs and patients who had switched from an anti-TNF to abatacept.

Methods A retrospective, observational analysis using research-identifiable Medicare claims data (January 2006–December 2010) was conducted in patients with RA prescribed an initial biologic DMARD (bDMARD; abatacept, adalimumab, etanercept, infliximab), with a minimum of 6-month, pre-index and 12-month, post-index eligibility. Patients were assigned to the first-line switch cohort if they had a claim for a new bDMARD within 200% days supply of the last claim for initial bDMARD during the12-month follow up. Further analysis in a subgroup of the switch cohort compared first-line anti-TNF patients who switched to another anti-TNF (adalimumab, etanercept, infliximab) with those who switched to abatacept. These two cohorts were followed to determine subsequent switch rates to a third bDMARD. Patients were considered second-line switchers if they had a claim for a new bDMARD within 200% days supply of the last claim for their second bDMARD during 12-month follow up. Logistic regression analysis was conducted to determine anti-TNF to anti-TNF and anti-TNF to abatacept regimens as predictors of subsequent switching, while controlling for covariates (age, gender, socioeconomic status, region, severity of RA, comorbidities and non-biologic RA therapies).

Results Of the 33,026 patients with RA identified who initiated bDMARD therapy (81% anti-TNF and 19% abatacept), first-line switch rates were comparable (anti-TNF: 6.5%; abatacept: 5.7%). Of the 1728 initial anti-TNF patients who switched a second time, 535 switched to another anti-TNF and 739 switched to abatacept. Subsequent switch to a third bDMARD occurred in 7.3% of those who had previously switched to abatacept and 18.9% of those who had previously switched to an anti-TNF. Logistic regression demonstrated that, after adjusting for covariates, patients who switched to abatacept had significantly lower odds of switching again than anti-TNF switchers (odds ratio=0.33, 95% CI: 0.22, 0.51).

• Download figure
• Open in new tab
• Download powerpoint

Conclusions In a Medicare population, patients with RA who switch between anti-TNFs have significantly higher rates of subsequent bDMARD switch compared with those who switch from an anti-TNF to abatacept. It is hypothesised that treatment pathways that include switching to a different class of bDMARD may reduce subsequent overall bDMARD switch rates and therefore reduce costs in the management of patients with RA.

References

1. Meissner B, et al. Arthritis Rheum 2011;63(Suppl 10): 2198

Disclosure of Interest L. Rosenblatt Shareholder of: Bristol-Myers squibb, Employee of: Bristol-Myers Squibb, F. Lobo Shareholder of: Bristol-Myers squibb, Grant/research support from: Bristol-Myers squibb, Employee of: Bristol-Myers squibb, P. Cockrum Employee of: Bristol-Myers squibb, L. Wang: None Declared, E. Alemao Shareholder of: Bristol-Myers squibb, Employee of: Bristol-Myers squibb, O. Baser Grant/research support from: Bristol-Myers squibb, H. Yuce: None Declared