B. Gilbert, D. Courvoisier, D. Mongin, K. Lauper, C. Perrier, R. Muller, A. Finckh
2021 Annals of the Rheumatic Diseases  
Background:When not responding to conventional synthetic DMARDs (csDMARDs), rheumatoid arthritis (RA) patients may receive biological DMARDs (bDMARDs) or targeted synthetic DMARDs (tsDMARDs) including baricitinib (BARI). While BARI has demonstrated efficacy in randomized controlled trials, limited studies have established comparative effectiveness in real world settings, in particular when used in b/tsDMARD-naïve patients.Objectives:To analyze the effectiveness of BARI versus alternative
more » ... , as assessed by drug maintenance over time and by response rates at 12 months.Methods:This is a nested study of RA patients, within the prospective Swiss Clinical Quality Management (SCQM) observational cohort.All treatment courses (TC) with BARI or alternative bDMARDs initiated between 2017-09-01 and 2020-06-01, with at least one follow-up visit, were included. TC with BARI were compared to TC with alternative bDMARDs (non-BARI), including all b/tsDMARDs except rituximab. The non-BARI group was then subdivided into TNF inhibitors (TNFi) and other mode of action bDMARDs (OMA), excluding tsDMARDs. A secondary analysis focusing specifically on b/tsDMARD naïve patients was conducted.Baseline characteristics were compared using ANOVA or χ2 tests. A Cox-model survival analysis assessed drug maintenance. 12-month response rates were estimated using an attrition-corrected, confounder-adjusted approach (1). CDAI score ≤10 defined low disease activity state (LDA), and CDAI score ≤2.8 defined remission.Results:Overall, 1218 eligible TC (from 1028 patients) were initiated during the study period (273 in BARI, 154 other tsDMARD, 473 in TNFi and 318 in OMA). Drug maintenance was significantly shorter for TNFi compared to BARI, even after adjustment for potential confounders (Hazard ratio (HR) for drug discontinuation 1.85 (95% CI [1,40 – 2,43]); p < 0.001). Drug maintenance was also numerically shorter for the OMA group compared to BARI, but the difference was not significant (HR 1.18 (95% CI [0.87 – 1.60]); p = 0.28). These differences were larger when analysing only bDMARD-naïve patients (Figure 1a).All TC taken together, the rates of LDA and remission did not differ significantly between the 3 groups at 12 months. LDA ranged from 63% to 67% (BARI vs OMA p = 0.87; BARI vs TNFi p = 0.81) and remission from 19% to 23% (BARI vs OMA p = 0.30; BARI vs TNFi p = 0.77; Figure 1b).Conclusion:BARI demonstrated a significantly higher overall drug maintenance than TNFi, and a similar drug maintenance to OMA, both in a bDMARD-naïve population and in the overall population. The adjusted 12-month response rates did not differ between BARI, TNFi and OMA groups. These results suggest that prescription of BARI after csDMARD has at least similar outcomes as alternative bDMARDs.References:[1]Lauper, K., et al. Sat0588. Annals of the Rheumatic Diseases 78 (2019).Table 1.Baseline characteristics of studied populationVariableOverall comparison:BARI vs non-BARISubgroup analysis 1:by bDMARD type (tsDMARD excluded)Subgroup analysis 2:In b/tsDMARD-naïve patientsBARI(TC = 273; 273 patients)Non-BARI(altogether)(TC = 945; 755 patients)p valuesTNFi(TC = 473; 408 patients)OMA(TC = 318; 298 patients)p values(vs BARI)BARI(n = 46)TNFi(n = 225)OMA(n = 65)p valuesMean (SD)Otherwise: n %Mean (SD)Otherwise: n %Mean (SD)Otherwise: n %Concomitant csDMARD41 %46 %0.1754 %41 %<0.0150 %61 %48 %0.1Line of Therapy-1st (= bio-naive)17 %35 %<0.0148 %20 %<0.01100 %100 %100 %-2nd20 %23 %23 %23 %-3rd19 %16 %11 %24 %-4th or later44 %26 %18 %33 %Female78%74 %<0.2174 %73 %0.2870 %71 %74 %0.88Age59 (14)56 (14)0.0152 (15)58 (13)<0.0157 (15)51 (14)57 (16)<0.01Disease duration (years)13 (10)10 (9)<0.018 (9)11 (9)<0.015.5 (6.45)5 (7)6.5 (9)0.243CDAI baseline15 (9)15 (10)0.914 (9)16 (11)0.0513 (7)14 (9)14 (10)0.72Previous tsDMARD (non-BARI)33 %8 %<0.014 %11 %<0.010 %0 %0 %TC duration > 12-months37 %29 %0.0127 %30 %0.0237 %29 %34 %0.52BARI = baricitinib, TNFi = TNF inhibitors, OMA = other mode of action, Non-BARI = TNFi + OMA + other tsDMARDs, SD = standard deviation.Disclosure of Interests:Benoit GILBERT: None declared, Delphine Courvoisier: None declared, Denis Mongin: None declared, Kim Lauper Consultant of: Gilead Galapagos, Grant/research support from: AbbVie, Clementine Perrier Shareholder of: Eli Lilly, Employee of: Eli Lilly, Rudiger Muller Consultant of: AbbVie, Novartis, Grant/research support from: GEBRO Pharma, Axel Finckh Speakers bureau: AbbVie, BMS, Pfizer, Eli-Lilly, Consultant of: AbbVie, BMS, Pfizer, Eli-Lilly, Grant/research support from: BMS, Pfizer, Eli-Lilly
doi:10.1136/annrheumdis-2021-eular.1781 fatcat:vcg2en2whvdxrhvv6a64kf3cdi