Background/Purpose: Baricitinib, an oral inhibitor of JAK1/JAK2, improved disease activity
with an
acceptable safety profile in a phase 3 study (RA-BEACON) of patients with active rheumatoid
arthritis (RA) and an inadequate response to tumor necrosis factor inhibitors (TNFi).¹
Oral baricitinib once-daily (QD) produced improvements that were sustained
through 24 weeks. The objective of this analysis was to evaluate the effect of prior
bDMARD exposure on the efficacy and safety of baricitinib in RA-BEACON.     

Methods: 527 patients with moderately-to-severely active RA despite
previous use of ≥1 TNFi for ≥3 months were randomized 1:1:1 to
receive placebo (PBO) or 2 or 4 mg baricitinib QD for 24 weeks.  All bDMARDs
were discontinued ≥28 days prior to treatment. Patients were enrolled
with a history of receiving multiple TNFi including non-TNFi bDMARDs.  Post hoc
subgroups defined by prior biologic use [1 TNFi only, >1 TNFi (but no non-TNFi),
and ≥3 bDMARDs] were assessed for evidence of qualitative and
quantitative treatment interactions utilizing modified intent to treat analysis
(mITT) and nonresponder imputation (NRI).  The treatment-by-subgroup interaction comparing
each baricitinib group to PBO was tested at the 0.1 significance level.

Results: Across the treatment arms, ~38% of patients had a
history of previous treatment with ≥1 non-TNFi bDMARD, including abatacept
20%, tocilizumab 19%, and rituximab 17%. Table 1 shows the ACR20 response rate for
baricitinib vs PBO in patients with prior exposure to a non-TNFi by type of
biologic. Improvements in ACR response rates and proportion of patients achieving
DAS28-CRP ≤3.2 were demonstrated at Weeks 12 and 24 with baricitinib in
subgroups defined by prior biologic use (Table 2). Significant interaction p
values were observed infrequently and inconsistently indicating minimal
treatment heterogeneity across subgroups defined by prior biologic use. As with the overall
study, more treatment emergent adverse events occurred in patients receiving
baricitinib 2 or 4 mg vs PBO with prior exposure to 1 or 2 biologics (65%, 74%,
62%, respectively) or ≥3 biologics (84%, 87%, 70%, respectively). Serious infections were
infrequent, and no TB or opportunistic infections were seen.¹

Conclusion: In this post-hoc, exploratory analysis of
prior bDMARD use in the TNFi-inadequate responder population, a beneficial
treatment effect was observed across subgroups irrespective of prior bDMARD use
(number or nature).  In general, a consistent treatment effect was observed
across the strata with no evidence of a qualitative interaction. 

¹Genovese et al.  Ann Rheum Dis
2015;74(Suppl2):75-76