Key clinical point: A machine-learning model combined demographic, clinical, and genetic markers to predict patients’ changes in disease activity scores 24 months after their baseline assessment and identified nonresponders to anti–tumor necrosis factor treatments.

Major finding: Compared with traditional trial-and-error practice, the model helped to predict up to 40% of anti–tumor necrosis factor nonresponders of European descent.

Study details: A machine-earning, model-building study of 1,892 RA patients and an independent dataset of 680 RA patients combined demographic, clinical, and genetic markers to predict patients’ changes in disease activity scores 24 months after their baseline assessment, and identify nonresponders to anti–tumor necrosis factor treatments.

Disclosure: The research was supported by the National Science Foundation and the National Natural Science Foundation of China. Several of the researchers reported financial relationships with pharmaceutical or technology companies.