TREATMENT RESPONSE PREDICTION IN PATIENTS WITH RHEUMATOID ARTHRITIS. Review

Abstract

Relevance. A variety of targeted therapies for rheumatoid arthritis (RA) treatment exist. Therefore, reliable predictors are needed that could be used to accurately predict the efficacy or inefficacy of these therapies in individual patients. This could allow clinicians to improve diagnosis and prognosis, to make the treatment personalized and to reduce healthcare expenses.
 Objectives: to analyze and systemize the predictors of response to treatment in patients with RA.
 Materials and Methods. We analyzed the recently discovered predictors of treatment response in RA patients using papers cited on PubMed, Lilacs, and EMBASE databases from Jan 2005 until Jan 2020. Predictive factors were grouped into four categories: methotrexate (MTX)-treated RA, tumor necrosis factor (TNF)-α inhibitors-treated RA, interleukin (IL)-6 inhibitors-treated RA, and rituximab (RTX)-treated RA.
 Results. Based on the results of several studies, predictors of response to methotrexate were high Disease Activity Score (DAS), concentration of myeloid-related proteins 8/14, high P-glycoprotein levels, low serum calprotectin and leptin levels, baseline serum concentration of tumor necrosis factor (TNF)-α, TNF receptor I, interleukin (IL)-1β, soluble CD163, numbers of CD14+highCD16, vascular cell adhesion molecule, lower expression of hsa-miR-132-3p, hsa-miR-146a-5p, and hsa-miR-155-5p. A positive response to biological therapy was determined by male gender, younger age, lower health assessment questionnaire, erythrocyte sedimentation rate or C-reactive protein, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, tender joint count (or swollen joint count) scores, absence of comorbidities, baseline albumin, IL-34, IL-1β, D-dimer, fibrinogen, matrix metalloproteinase 3, DAS 28 and Simplified Disease Activity Index (SDAI). The plasma interferon (IFN) activity and the IFN beta/alpha ratio, IL-1Ra level were predictive in TNF antagonist-treated patients. Predictors of response to IL-6 inhibitors were anti–citrullinated protein antibody (ACPA)+, baseline Sharp/van der Heijde score, myeloid soluble intercellular adhesion molecule 1, serum levels of sIL-6R, IL-8, calprotectin, and lymphoid activation and bone remodeling markers. The prediction of the best response for rituximab was determined to be a combination of IL-33, rheumatoid factor or ACPA, IgG, and also lower number of previous biological therapies. Genetic factors, such as single-nucleotide polymorphisms at gene locus rs10919563, rs11541076, rs12083537, rs11265618, and rs1801274, and rs396991 can also be used to predict a response to treatment.
 Conclusions. One of the leading problems in the development of predictors remains the collection of high-quality and complete information from a large number of patients. For this, it is necessary to develop an digital program for collecting specific data (depending on the specific disease) and developing new algorithms for predicting the response to treatment.