A Gaussian process regression model in rheumatoid arthritis (RA) showed promise in guiding anti-TNF drug selections in clinical practice based on primarily clinical profiles with additional genetic information, a new study found. The model was developed and cross-validated in 1,892 patients. It was evaluated on an independent dataset of 680 patients. Researchers found:
- In cross-validation tests, the studied method predicts changes in disease activity scores with a correlation coefficient of 0.406.
- It correctly classified responses of 78% of participants.
- Gaussian process regression effectively remapped the feature space and identified subpopulations that do not respond well to anti-TNF treatments.
- Genetic SNP biomarkers showed small additional contribution in the prediction on top of the clinical models.
Guan Y, et al. Machine learning to predict anti-TNF drug responses of rheumatoid arthritis patients by integrating clinical and genetic markers. [Published online ahead of print July 24, 2019]. Arthritis Rheumatol. doi: 10.1002/art.41056.