Background/Purpose: Despite advances in rheumatoid arthritis (RA) treatment, 20–40% of patients do not respond to biologic or targeted synthetic DMARDs (b/tsDMARDs). Understanding each drug’s molecular effects may help identify subgroups of patients likely to respond to specific therapies. Objectives: 1-Characterize the ex vivo molecular and cellular effects of JAK (JAKi) and TNF (TNFi) inhibitors on RA immune cells. 2-Identify, in vivo, RA patient subgroups with distinct baseline levels of drug-specific proteomic signatures and assess their association with clinical response.

Methods: Peripheral blood mononuclear cells (PBMC) and neutrophils from 48 RA patients naïve to b/tsDMARDs were cultured with autologous serum and Baricitinib (JAKi) or Etanercept (TNFi) (10µM) for 24 and 12 hours, respectively. Proliferation, adhesion and NETosis were quantified using specialized kits. Protein changes were assessed via proximity extension assay, analysing a 92-protein inflammation panel in the supernatant. To identify subgroups of patients likely to respond to therapies guided by the identified proteomic signatures, a total of 193 RA patients were included (36 treatment-naïve and 32 previously treated patients initiating JAKi, and 125 treatment-naïve patients starting TNFi) where those proteomic signatures were analysed.

Results: Baricitinib and Etanercept reduced proliferation, adhesion and NETosis in PBMC and neutrophils from naïve RA patients. Each drug modulated distinct inflammatory protein signatures: Baricitinib primarily affected Th17/adaptive immunity while Etanercept targeted acute inflammation.Unsupervised clustering analysis of JAKi–naïve patients, based on the ex vivo–derived JAKi proteomic signature, revealed two distinct subgroups. Patients with high expression of this signature showed a significantly greater reduction in disease activity (ΔDAS28-CRP) at 3, 6, and 12 months. In contrast, applying the same stratification to non-naïve patients did not predict treatment response. In TNFi-treated naïve patients, the JAKi-associated proteomic signature also stratified into distinct subgroups; however, high baseline levels were not associated with clinical response, supporting the specificity of this signature to JAK inhibitor therapy.The ex vivo–derived Etanercept proteomic signature effectively stratified TNFi-treated patients, with high baseline levels correlating with improved clinical response at 6 and 12 months. This signature was specific to TNFi and showed no association with treatment response in JAKi-treated patients.

Conclusion: 1. JAKi and TNFi have specific functional and proteomic effects on RA immune cells. 2. Both ex vivo-derived JAKi and TNFi signatures effectively stratified RA patients into subgroups, with each being specific to its respective treatment and not applicable for predicting the clinical response to the other. This highlights the value of drug-specific molecular signatures to personalize and optimize RA treatment strategies. Funded by Instituto de Salud Carlos III (ISCIII) and co-funded by European Union. Supported by EU/EFPIA IMI-JU 3TR, ISCIII (PI24/00959, CD21/00187, RICOR-21/0002/0033) and RYC2021-033828-I.

ACR Meeting Abstracts - https://acrabstracts.org/abstract/distinct-immune-molecular-signatures-modulated-ex-vivo-by-jak-and-tnf-predict-rheumatoid-arthritis-therapy-outcomes-in-patients-naive-to-biologic-and-targeted-synthetic-dmards/