Patient-level clinical and proteomic data supporting predictive models of TNF inhibitor response in psoriatic arthritis

This dataset supports the development and independent evaluation of clinical, proteomic, and combined predictive models of response to tumour necrosis factor inhibitors (TNFi) in patients with psoriatic arthritis. The dataset includes patient-level measurements from a discovery cohort (n = 47) and an independent evaluation cohort (n = 26). Variables comprise four proteomic predictors (CD200, PDGFB, CLSTN2, and BANK1), two clinical predictors (erythrocyte sedimentation rate and dactylitis), age, sex, and treatment-response status assessed using DAPSA50 at six months.