Background/Purpose: Tofacitinib is an oral Janus kinase inhibitor for the treatment of RA.  Tofacitinib access has been restricted by many US payers to use after treatment with ≥1 injectable biologic DMARD (bDMARD); there are limited real-world data comparing tofacitinib with bDMARDs in biologic-naïve RA patients (pts). This study compared pt characteristics, treatment patterns, and costs in pts initiating tofacitinib vs adalimumab (ADA) or etanercept (ETN) in a US population with commercial healthcare insurance.

Methods: This retrospective cohort study included pts aged 18–64 years with ≥1 tofacitinib or bDMARD healthcare insurance claim (Truven MarketScan® Commercial Claims and Encounters database) during 1/1/2014–1/1/2017, ≥1 MTX claim within 12 months prior to index, and ≥1 diagnosis of RA at index or within 12 months prior. Pts had to be continuously enrolled for ≥12 months pre-/post-index with no pre-index claims for tofacitinib or a bDMARD. Monotherapy was defined as absence of select conventional synthetic DMARDs within 90 days post-index. Outcomes were treatment persistence at 12 months (index refills with <60-day gap after supply expiration), adherence at 12 months (proportion of days covered [PDC]), and 12-month post-index RA-related costs. Statistical analyses were performed using t-statistics (continuous variables) or chi-squared statistics (binary variables) in pairwise comparisons.

Results: Pts who met selection criteria and initiated tofacitinib (n=184), ADA (n=1771), or ETN (n=1472) had similar baseline characteristics except for age which was higher in tofacitinib pts (mean 52.6 [SD 9.3], 49.9 [9.5], and 49.9 [10.0] years, respectively), and tofacitinib pts were more often located in the Northeast vs Southern region vs ADA. Tofacitinib pts had higher 12-month pre-index costs vs ADA (mean $4296 [SD $8159] vs $2880 [$5497], respectively; p=0.0225), more rheumatologist visits 90 days pre-index (mean 1.80 [SD 0.86] vs 1.96 [1.11]; p=0.0435), and higher (worse) Quan Charlson comorbidity index scores (mean 1.65 [SD 0.93] vs 1.50 [0.91]; p=0.0300). Tofacitinib pts had higher monotherapy use (34.2%) vs ADA (21.7%; p=0.0001) and ETN (26.5%; p=0.0263). Persistence at 12 months was similar for tofacitinib pts (46.2%) vs ADA (49.5%) and ETN (50.5%), although more tofacitinib pts restarted index treatment during 12 months post-index (17.9%) vs ADA (9.8%; p=0.0006) and ETN (7.1%; p<0.0001), and fewer tofacitinib pts switched treatment (16.9%) vs ADA (26.5%; p=0.0044) and ETN (29.6%; p=0.0003). Adherence (PDC) at 12 months was also similar between tofacitinib (0.63), ADA (0.67), and ETN (0.67) pts. Total post-index RA-related costs were lower with tofacitinib pts (mean $29,938 [SD $19,287]) vs ADA ($38,733 [$17,096]; p<0.0001) and ETN ($38,534 [$16,261]; p<0.0001).

Conclusion: Among biologic-naïve patients with RA and prior MTX, there was higher baseline monotherapy use among those receiving tofacitinib vs ADA and ETN. Persistence and adherence were similar for tofacitinib, and total RA-related costs were lower vs ADA and ETN, despite tofacitinib pts having higher pre-baseline costs.