A spectroscopic diagnostic for rheumatoid arthritis using liquid biopsies

Abstract

Rheumatoid arthritis (RA) possess not only a substantial degree of clinical heterogeneity but is diagnosed on a diverse array of clinical criteria. The lack of a single marker predictive methodology means that the timely diagnosis and treatment of these patients proves challenging. With the advent of targeted therapies, it is becoming increasingly important to accurately diagnose RA at an early stage of disease in order to ensure effective and timely disease management which can minimise long term sequelae such as joint tissue damage. Raman spectroscopy has recently gained increasing clinical recognition as a non-invasive and label-free method for obtaining a complete biochemical fingerprint of the content of biological samples. This study explored the application of Raman spectroscopy coupled with multivariate data analysis, as an adjunct or alternative tool for the differential diagnosis of RA using peripheral blood mononuclear cells (PBMCs) and purified primary immune cell subsets. High performance partial least square discriminant analysis (PLSDA) classification models constructed in this study enabled identification of spectroscopic discrimination of RA patients and healthy controls without influence from confounding factors. Spectral fitting analysis identified potential spectral biomarkers such as Proteinase K and TNF-α that elucidate the spectral classification between healthy controls and RA patients. These results demonstrate the capability of Raman Spectroscopy in RA diagnosis.