This study was designed to investigate the effects of polymorphisms in RETN on remission in RA patients receiving TNF-α inhibitors. In addition, machine learning algorithms were trained to predict remission. Ten single-nucleotide polymorphisms were investigated. Univariate and multivariable analyses were performed to evaluate associations between genetic polymorphisms and the efficacy of TNF-α inhibitors. A random forest-based classification approach was used to assess the importance of different variables associated with the efficacy of TNF-α inhibitors. Various machine learning methods were used for finding vital factors and prediction of remission. The eight most significant features included in the multivariable analysis were sex, age, hypertension, sulfasalazine, rs1862513, rs3219178, rs3219177, and rs3745369. T-allele carriers of rs3219177 and males showed approximately 6.0- and 3.6-fold higher remission rates compared to those with the CC genotype and females, respectively. The elastic net algorithm was the best machine-learning method for predicting remission of patients with RA treated with TNF-α inhibitors. On the basis of the results of this study, it may be possible to design individually tailored treatment regimens to predict the efficacy of TNF-α inhibitors.
Keywords: Arthritis; Machine learning; Polymorphism; RETN; Resistin; Rheumatoid; Tumor necrosis factor-alpha.
Copyright © 2022 Elsevier B.V. All rights reserved.