Biomarkers identification of the efficacy of anti-TNF α agent (Adalimumab) in ankylosing spondylitis patients using a proteomic approach

• Autor(a) principal: Fernandes, Ana Filipa Ribeiro
• Data de Publicação: 2020
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10348/10725
• topic: TNF inhibitor therapy; radiographic axial spondyloarthritis

Axial Spondyloarthritis (axSpA) is amongst the most common forms of inflammatory arthritis. Widely used in the treatment of AS (the most frequent axSpA), adalimumab is an engineered antibody holding only human peptide sequences that bind with high affinity and specificity to soluble and transmembrane TNF α, blocking its interaction with receptors p55 and p75. However, the response to TNF inhibitors (TNFi) is heterogeneous in terms of efficacy and might be linked with serious side effects. A priori identification of patients more propense to respond to TNFi is critical in clinical practice. Therefore, the aim of this work was to identify serum proteins for which the level of variation is tightly associated with the clinical response to adalimumab in AS patients. The proteomic analysis involved 33 patients with AS, of which 19 responders and 14 nonresponders after 14 weeks treatment with adalimumab, according to ASAS criteria. Serum samples were collected at baseline, 3-5 days, 2 weeks and 14 weeks after treatment. The experimental workflow combined immunoaffinity depletion of the high-abundant serum proteins, tryptic digestion of the proteins extracted from the depleted serum, HPLC separation of tryptic peptides, MS/MS based identification and MS quantification of the detected proteins. Uni- and multivariate statistical analysis of the differentially abundant proteins was used to select the more sensitive and specific serum biomarkers related with therapeutic response. Protein function association network analysis of differential proteins was performed with STRINGdb. LC-MS/MS method allowed the identification of 333 proteins with at least 2 non-ambiguous peptides. A set of new putative biomarkers was identified with 8 proteins displaying differences between R and NR (p < 0.05) at baseline. Of these, 3 proteins were highly linked with a good clinical response, while the other 5 proteins were strongly related with a non-response to adalimumab therapy at W14. This set of proteins was confirmed to be predictive of the type of response to treatment with an area under the Receiver Operating Characteristics (ROC) curve of 1. Additionally, our analysis showed that the pathways dysregulated in both groups were relatively similar. Proteomic approaches constitute a very promising strategy to the identification of biomarkers to predict a therapeutic response. These results provide evidence that a panel of biomarker proteins might be identified even before the beginning of treatment which is of utmost importance for clinical practice.