Screening Of Differentially Expressed Serum Proteins For Rheumatoid Arthritis By Surface-enhanced Laser Desorption/Inionation-time Of Flight-mass Spectra

Objective:To study the specific proteins by surface-enhanced laser desorption ionization/time of flight mass spectrometry(SELDI-TOF-MS) in serum of patients with rheumatoid arthritis(RA).Methods:Our study included 184 serum samples, 97 were from patients with RA,76 from healthy individuals. Serum was separated from 4 m L of blood by centrifugation at 3,000 rpm for 10 min, aliquoted and immediately frozen at 80°C until thawed for analysis.1.97 RA patients and 77 healthy control,which were randomly allocated to the training set(83 RA patients and 56 healthy controls) or test set(14 RA patients and 20 healthy controls)to develop and verify a pattern by means of decision treealgorithm.2.The 97 RA patients were divided into active phase(68 cases, including low disease activity(n=17),moderate disease activity(n=21), high disease activity(n=30) and inactive(n=15), simple RA patients(n=44) and RA-SS patients(n=18), RA patients(n=44) and RA-ILD(n=22), RA patients(n=44) and RA-ONFH( n=6), early RA(n=24) and non-early RA(n=60), meet the 1987 diagnostic criteria(n=43) and meet the 2009 diagnostic criteria for RA(n=40), antibody-positive group(n=66) and antibody negative group(n=17), anti-CCP positive group(n=51) andanti-CCP-negative group(n=32),RF-positive group(n=60) and RF-negative group(n=23)and other types.Using theBiomarker Wizard and Biomarker Pattern software to establish the diagnosis prediction model to predict RA disease progression and disease activity.3.SELDI-TOF-MS using Biomarker Wizard software and Biomarker Pattern Software established diagnostic model, and calculate sensitivity and specificity.Results:1.The diagnostic model of M/Z3448.857,4716.712,8214.285 and 10645.1 screening between 83 RA patients and 56 healthy people(P <0.05). The sensitivity and specificity is 91.566% and 92.857%, the area under the ROC curve was 0.937, to verify the diagnosis model, we get a sensitivity of 100% and a specificity of 95%.2.The diagnostic model of the M/Z1130.776, M/Z1501.065, M/Z2091.198, M/Z11381.87 screening between non-active RA patients and low disease activity patients(P<0.05). The sensitivity and specificity is 94.118% and 93.333% and the area under the ROC curve was 0.990.3.The diagnostic model of the M/Z1330.019, M/Z1501.065, M/Z4060.468 screening between the patients of inactive disease and moderate disease active(P <0.05). The sensitivity and specificity is 95.238% and 86.667% and the area under the ROC curve was0.973.4.The diagnostic model of the M/Z2013.494, M/Z8765.232 screening between the patients of inactive disease and high disease active(P <0.05). The sensitivity and specificity is 80.645% and 86.667%, and the area under the ROC curve was 0.898.5.The diagnostic model of the M/Z2013.494, M/Z8765.232 screening between the patients of low disease activity and moderate active disease(P <0.05). The sensitivity and specificity is 80.645% and 86.667% and the area under the ROC curve was 0.989.6.The diagnostic model of the M/Z2032.31, M/Z4716.71 screening between the patients of moderate disease activity and high disease activity(P <0.05). The sensitivity and specificity is 74.194% and 76.190% and the area under the ROC curve was 0.786.7.The diagnostic model of the M/Z2013.494, M/Z4825.63 screening between thepatients of low disease activity and high disease activity(P <0.05), consisting of diagnostic model. The sensitivity and specificity is 87.09% and 88.235% and the ROC curve was 0.916.8.The diagnostic model of M/Z10645.1, M/Z12595.86 screening between the patients of RA patients and RA-ILD(P <0.05). The sensitivity and specificity is 86.364% and84.091% and the area under the ROC curve was 0.856.9.The diagnostic model of M/Z6635.623, M/Z33897.72 screening between the patients of RA patients and RA-SS(P <0.05). The sensitivity and specificity is 77.778% and79.545% and the area under the ROC curve was 0.794.10.The diagnostic model of M/Z2071.689 screening between the patients of RA patients and RA-ONFH(P <0.05). The sensitivity and specificity is 66.667% and 95.445%and the area under the ROC curve was 0.811.12.The diagnostic model of M/Z1679.331, M/Z2032.31, M/Z4465.218,M/Z33897.72M/ Z2071.689 screening between the patients of meeting the 1987 criteria and meeting the 2009 criteria(P <0.05). The sensitivity and specificity is 92.5% and81.395% and the area under the ROC curve was 0.909.13.The diagnostic model of the M/Z1221.005, M/Z2385.32, M/Z4825.63, M/Z8765.232 screening between the patients of antibody positive and antibody-negative group(P <0.05). Tsensitivity and specificity is 86.567% of 87.5% and the area under the ROC curve was 0.926.14.The diagnostic model of the M/Z1247.027, M/Z2671.604, M/Z4465.218,M/Z28057.6screening between the patients of anti-CCP positive group and anti-CCP negative(P <0.05). The sensitivity and specificity is 94.118% and 90.625%. The area under the ROC curve of 0.948.15.The diagnostic model of the M/Z1501.065 screening between the patients of RF positive and RF-negative group(P <0.05). The sensitivity and a specificity is 78.333% and65.21 % and the area under the ROC curve was 0.718.ConclusionThe serum protein fingerprinting by SELDI-TOF-MS could identify new biomarkers in AS. The biomarkers may play an important role in pathogenesis of RA. We could diagnose RA in early stage, predict disease progression and determine disease activity by these biomarkers.