Background/Purpose:

While studies have assessed the efficacy of switching among biologic disease-modifying antirheumatic drugs (bDMARD), there is a lack of knowledge regarding the patterns of bDMARD use and switching in a real-world setting. The objective of this study was to describe and compare the characteristics and treatment patterns associated with RA patients who maintain their initial bDMARD (non-switchers) to: [1] patients that cycle from one anti–tumor necrosis factor (AT) to another and [2] patients that switch to a non-anti-TNF agent (NAT) in the year following bDMARD initiation.

Methods:

Patients aged  ≥18 with ≥1 diagnosis of RA between January 2004 and August 2011, from an administrative claims database, were included if they initiated a new anti-TNF treatment, did not have any bDMARD claims in the 12 months prior to bDMARD initiation, and were continuously enrolled for 12 months following bDMARD initiation. Patients who had ≥1 different bDMARD following initiation were classified as AT or NAT study group, depending on mechanism of action (MOA). AT and NAT groups were compared to non-switchers, using t-tests for continuous data and chi-square tests for categorical data. Cox Proportional Hazards model was used to examine time-to-discontinuation/switch while controlling for confounding factors.

Results:

Among the 7,719 naïve bDMARD patients identified, 87% [n=6,698] maintained their initial bDMARD, 10% [n=792] cycled to a second AT and 3% [n=229] switched to a NAT.

Comorbidity profile of study groups was varied; the Deyo- Charlson Comorbidity Index for non-switchers [1.59] was higher than the AT group [1.49; P=0.033] and lower than the NAT group [1.78; P=0.038].  At baseline, the AT group had significantly lower proportions of cardiovascular disease, hypertension, dyslipidemia, and malignancy than the non-switchers [P<0.05]. The NAT group had higher rates of hypertension, diabetes and dyslipidemia than non-switchers [P<0.05].

At baseline, the AT group had significantly greater use of background medications than non-switchers [methotrexate (MTX): 75.5% vs 61.9%; P<0.0001, other non-biologic DMARDs (nbDMARDs):48.4 % vs 40.3%; P<.0001]. The NAT group also had greater use of MTX compared to non-switchers (61.9% vs 54.6%; P<0.0001) but not nbDMARDs.

The most common index bDMARD was etanercept [45%] and adalimumab [20%].  Among the AT group, the majority of patients switched to adalimumab [49%], followed by etanercept [25%], infliximab [20%] golimumab [4%], and certolizumab [3%]. For the NAT group, most common switch therapies were abatacept [43%], tocilizumab [38%], and rituximab [19%].

Mean duration of therapy on index bDMARD was 248 days, 145 days, and 127 days for non-switchers, the AT group and NAT groups, respectively.  In adjusted analyses, the AT and NAT groups had similar discontinuation/switch rates (hazard ratio=1.13; P=0.15).

Conclusion:

In the 12 months following bDMARD initiation, most RA patients maintained their initial bDMARD. Compared to non-switchers, the NAT group had higher comorbidities, suggesting that sicker patients are more likely to be switched to a different MOA. The use of MTX and nbDMARDs at baseline was not universal.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/biologic-switching-among-patients-with-rheumatoid-arthritis-in-the-united-states-2004-2011/