Pattern-based diagnosis and screening of differentially expressed serum proteins for rheumatoid arthritis by proteomic fingerprinting.

The objective of this study was to search for proteomic patterns distinguishing patients with rheumatoid arthritis (RA) from healthy controls, biomarker candidates specific for early RA, and proteins reflecting disease activity, by profiling of serum proteins using magnetic bead-based (MB) separation and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). Magnetic chemical affinity beads were used to differentially capture serum proteins prior to MALDI-TOF analysis. Seventy serum samples from patients with RA and 50 from healthy controls were analyzed. The samples were randomly allocated to the training set or test set to develop a pattern by means of decision tree algorithm. ANOVA test was utilized to search for biomarkers for early RA. Pearson correlation analysis was employed to estimate the overexpressed peaks in relation to Disease Activity Score (DAS)28. The algorithm identified a pattern based on 3 peaks (m/z 2,490, 5,910.07, 6,436.73) that, in the training set, separated patients with RA from healthy controls with a sensitivity and specificity of 87.5 and 96.7%, respectively. Blind test data indicated a sensitivity of 86.7% and specificity of 90%. The peaks of m/z 1,014.92 and 1,061.38 were raised significantly in early RA group (disease duration <12 months) compared with those in non-early RA group (disease duration ≥12 months) and healthy controls. A positive correlation was found between the intensity of peak 9,591.47 and DAS28 scores. Using MB separation followed by MALDI-TOF-MS enabled rapid diagnosis of RA according to fingerprint pattern, a method which might also help to assess disease activity and identify early RA.