Integrative AI-based multiomic and neurophysiological profiling of chronic pain in rheumatoid arthritis: study protocol for the prospective, observational, case-control RA-PAIN-AI study.

introductionRheumatoid arthritis (RA) is a chronic systemic autoimmune disease in which pain remains the most prominent and burdensome symptom from the patient's perspective. Despite effective control of peripheral inflammation with biological and targeted synthetic disease-modifying antirheumatic drugs, an estimated 20-30% of patients continue to experience persistent moderate-to-severe pain after achieving clinical remission. This therapeutic gap reflects the evolution of pain mechanisms from peripheral nociception towards central sensitisation and nociplastic pain, creating a need for integrative approaches to patient stratification. The Rheumatoid Arthritis Pain Artificial Intelligence (RA-PAIN-AI) study aims to identify and characterise clinical, neurophysiological and multiomic signatures associated with chronic pain in patients with RA using integrative artificial intelligence (AI)-based analytical methods. METHODS AND ANALYSIS: RA-PAIN-AI is a prospective, observational, single-centre case-control study integrating clinical laboratory neurophysiological and multiomic data. Two groups of adult patients with RA will be enrolled: patients with active RA qualifying for biological therapy under the Polish national drug programme B.33 and patients receiving biological therapy who are in sustained clinical remission for at least 1.5 years without clinically significant chronic pain. All participants will undergo clinical assessment, validated patient-reported outcome measures and peripheral blood collection at baseline, followed by repeat questionnaires and peripheral blood collection at 3-6 months. The exact interval between visits will be recorded and included in longitudinal analyses where appropriate. Electroencephalography will be performed only in an exploratory subset according to predefined feasibility, safety and eligibility criteria. Biological samples will be analysed using whole genome sequencing (genomics), messenger RNA expression profiling (transcriptomics) and measurement of circulating proteins (secretomics). A minimum of 50 participants per group will be recruited. Because the study is exploratory and high-dimensional, AI and machine-learning analyses will use prespecified feature filtering, dimensionality reduction, regularisation and internal validation procedures, and all predictive models will require external validation in future cohorts. ETHICS AND DISSEMINATION: The protocol received a favourable opinion from the Bioethics Committee at the Hirszfeld Institute of Immunology and Experimental Therapy, Polish Academy of Sciences, Wroclaw, Poland (Opinion No. KB-17/2025, dated 25 September 2025). Written informed consent will be obtained from all participants before any study-related procedure. The results will be disseminated through peer-reviewed publications, conference presentations and communication with relevant scientific, clinical and patient communities. TRIAL REGISTRATION NUMBER: NCT07769619.