Introduction

Real‑world use of Janus kinase (JAK) inhibitors often involves rheumatoid arthritis (RA)  patients intolerant to methotrexate (MTX) or with multiple prior biological disease-modifying antirheumatic drugs (bDMARDs) failures, leading to different clinical profiles than those enrolled in randomized trials.

Aim 

To evaluate efficacy & safety of tofacitinib (TOF), baricitinib (BARI), peficitinib (PEF) and upadacitinib (UPA) in real world clinical settings after minimizing selection bias and adjusting the confounding patient characteristics. 

Patient Profile

Patients with RA diagnosed based on the 1987 RA classification criteria of the American College of Rheumatology (ACR)

Methods

• Multicentre, retrospective study
• The study included 622 patients with rheumatoid arthritis (RA) treated with TOFA, BARI, PEF, UPA from ANSWER cohort database.
• The values of Clinical Disease Activity Index (CDAI), C-reactive protein (CRP), and modified Health Assessment
• Questionnaire (mHAQ) after drug initiation and the remission or low disease activity (LDA) rates of CDAI at 6 months after drug initiation were compared among the four groups.

• Clinical Disease Activity Index(CDAI), C-reactive protein (CRP), Modified Health Assessment Questionnaire (mHAQ), Drug retention and discontinuation [lack of efficacy, adeverse events( AEs)] were measured at baseline, 1, 2, 4, 6 months.

• Additional analyses included
• Predictors of low disease activity (LDA) failure for TOF and BARI
• Subgroup: With vs. without MTX
• Subgroup: First vs. later biological/ targeted synthetic (b/tsDMARD line of therapy

Results

Drug Retention and Safety

• Overall, 6‑month retention rate including TOFA, BARI, PEF and UPA was 85.4%. 
• No significant difference was observed among TOF (87.4%), BARI (89.5%), PEF (82.8%), UPA (83.0%) (log‑rank p=0.527). 
• Discontinuation due to Toxic AEs and lack of efficacy also did not differ significantly.
  1. Toxic AEs were  TOF 8.7%, BARI 5.2%, PEF 6.9%, UPA 5.7%; p=0.714) 
  2. Lack of efficacy was TOF 3.9%, BARI 5.9%, PEF 13.8%, UPA 7.5%; p=0.201) 

Table 1 : Comparison of drug retention rates and discontinuation rates due to toxic adverse events and lack of efficacy until 6 months after drug administrations until 6 months among TOFA, BARI, PEF and UPA treatment

Drug efficacy (overall and post‑IPTW)

• At 6 months, CDAI, mHAQ, and CRP improved from baseline across all groups (p<0.001)
• No significant differences among TOF, BARI, PEF, and UPA after IPTW 
• CDAI remission (TOF 35%, BARI 30%, PEF 46%, UPA 44%) and CDAI‑LDA (TOF 87%, BARI 85%, PEF 89%, UPA 82%) were statistically similar between drugs.

Efficacy of JAK inhibitors in subgroup: with vs without methotrexate (MTX)

• CDAI Remission rates showed no significant difference with vs without MTX for TOF or BARI. (TOFA: OR 1.1, p=0.318; BARI: OR 1.0, p=0.936).
• LDA: modestly higher with MTX for both TOF (OR 1.1; p=0.043) and BARI (OR 1.1; p=0.016).

Predictive factors for resistance to LDA (multivariable)with

• TOF: higher baseline CDAI (OR 1.09; p<0.001) and baseline CRP (OR 1.32; p=0.049) predicted resistance.
• BARI: higher baseline CDAI (OR 1.07; p<0.001), glucocorticoid dose (OR 1.18; p=0.035), and number of prior b/tsDMARDs (OR 1.36; p=0.004) predicted resistance.

Conclusion

• Drug retention, efficacy, and safety were comparable across TOFA, BARI, PEF, and UPA, with no significant differences observed among the four JAK inhibitors for RA treatment. 
• Predictors of reduced LDA response differed by drug: higher baseline CRP and CDAI were linked to poorer response with TOFA, while higher baseline glucocorticoid dose, baseline CDAI, and greater prior b/tsDMARD use predicted reduced response with BARI.

Reference

Rheumatology, 2024; 63: 3033–3041