Background/Purpose: Methotrexate (MTX) is the first line treatment for Rheumatoid Arthritis (RA) in Sweden, but one third of patients do not experience satisfying treatment response. Delay in effective treatment cause increased inflammation, handicap and decreased quality of life. The prescription of the most appropriate treatment, in a stratified manner, is thus crucial to provide better and more cost effective health care. MTX is an inhibitor of protein and nucleic acid synthesis that is known to inhibit immune cells. The exact mechanism of MTX is unknown. Here we aimed to:

• Detect the cellular processes elicited by MTX treatment and/or connected to treatment response.
• Predict MTX treatment response using data collected prior to treatment initiation.

Methods: We modeled treatment response based on gene expression and flow cytometry data in a well-characterized routine clinical care cohort. The 57 RA patients underwent clinical evaluation by rheumatologist at the Karolinska Hospital at baseline as well as three months into their treatment. Both times patients contributed blood samples in which broad panels of cell type and activation markers were assayed using flow cytometry and RNA was extracted and sequenced.

Results: We first detected genes whose expression changed upon treatment and/or in connection with a better EULAR treatment response. Treatment had a large effect on gene expression where 132 genes were regulated with a false discovery rate (FDR) < 5%. The gene expression changes that were associated to treatment response (25 genes with FDR < 5%) overlap with the genes regulated by treatment, but in patients that respond well the induced changes were larger. Thus, if MTX is able to elicit stronger effects the patient is more likely to respond well to treatment. Pathway analysis revealed that MTX treatment down-regulated the antibody-driven immune response (p:1×10^-26). The proportion of plasma cells in peripheral blood was indeed decreased by MTX treatment (p: 0.008). Next we detected genes whose expression in treatment-naïve patients is associated to a later good or moderate EULAR response. We adjusted our analysis for potential confounding factors: percentage of B-cells, ancestry, age, sex and anti-CCP. Our analysis was restricted to genes that we detected as modulated by treatment (50% FDR, 1.3k genes). We observe an enrichment of small p-values, indicating the presence of true associations. Two immunoglobulin genes had a higher expression in those who later respond to treatment (IGLV8-61, IGLV7-46, FDR: 5.4%). Patients who later respond well to treatment indeed had a larger proportion of plasma cells in their peripheral blood (p:0.007). CXCL9 had a lower expression among patients that later respond to treatment (p: 2×10^-5, FDR: 2.1%). CXCL9 was in fact up-regulated by treatment (p: 0.007), and this up-regulation was stronger in patients who respond well (p: 0.001).

Conclusion: We have detected that MTX treatment is effective in patients with a larger proportion of plasma cells, and that the presence of these cells is significantly diminished by MTX treatment. Patients who respond well express less CXCL9 at baseline, but up-regulates this gene strongly during treatment.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/antibody-mediated-immunity-drives-response-to-methotrexate-treatment-in-rheumatoid-arthritis-patients/