Background/Purpose:
For patients with
rheumatoid arthritis (RA), there are limited real-world data on factors that predict
persistency on biologic therapy or whether use of biologics as monotherapy vs
in combination with conventional synthetic disease-modifying antirheumatic
drugs (csDMARDs) affects persistency. This analysis compared persistency of the
interleukin-6 receptor α inhibitor tocilizumab (TCZ) initiated as
combination therapy vs monotherapy and evaluated predictors of persistency in
patients with RA.

Methods:
Corrona is a
US-based, prospective, observational cohort of nearly 40,000 patients with RA. Patients
who newly initiated TCZ on or after October 1, 2008, within the Corrona
registry and who had ≥ 1 follow-up visit were included in this analysis.
Persistency was defined as time to the switch to a new biologic, regardless of
possible gaps between TCZ discontinuation and switch to a new biologic.
Kaplan-Meier analyses estimated the persistency for all eligible TCZ
initiations and by combination therapy vs monotherapy. Multivariable Cox
proportional hazards models evaluated associations of type of therapy
(combination therapy vs monotherapy) and other baseline factors with
persistency of treatment.

Results: Of the 1001 initiations of TCZ
included in this analysis, 80.7% of patients were female, median (IQR) age was
58 (49-66) years and median (IQR) duration of RA was 9 (5-17) years. At the
time of TCZ initiation, 33.0% initiated TCZ as monotherapy, 52.6% had high
disease activity (Clinical Disease Activity Index > 22) and 90.8% had
received prior tumor necrosis factor inhibitors (TNFis). The overall median
(95% CI) time to switch from TCZ to a new biologic was 33.5 (28.0-41.0) months
(Table). In patients who initiated TCZ as monotherapy, the median (95%
CI) time to switch was comparable with those who initiated TCZ in combination
with csDMARDs (36.4 [29.6-not estimable] months vs 28.0 [22.4-40.0] months,
respectively; P = 0.162). Compared with initiation of TCZ as monotherapy,
initiation of TCZ as combination therapy was not associated with an increased
likelihood of switching from TCZ to a new biologic (adjusted hazard ratio [HR; 95%
CI], 0.81 [0.64-1.04]; P = 0.100). Previous and current smokers were
significantly more likely to switch to a new biologic (HR [95% CI], 1.32 [1.02-1.73]
and 1.50 [1.10-2.06]; P = 0.040 and 0.010, respectively), while prior
use of 1 TNFi was significantly associated with a 47% decreased likelihood of
switching compared with TNFi-naive patients (HR [95% CI], 0.53 [0.34-0.81]; P
= 0.003).

Conclusion: In this real-world setting of
patients with RA, persistency on TCZ was similar when initiated as monotherapy or
as combination therapy. Among TCZ initiators, prior use of 1 TNFi was associated
with a decreased likelihood of switching to another biologic compared with
those with no prior TNFi exposure, whereas smoking was associated with an
increased likelihood of switching to another biologic.