Screening Potential Serum Markers For Liver Fibrosis: A Preliminary Study

China has got a lot of concern to the world as the high incidence of hepatitis B, it has also been a scientific consensus among clinical workers that hepatitis can cause liver fibrosis, cirrhosis even liver cancer. Despite a series of measures have been taken to slow down the process of the diseases, the effect is not very satis factory. How can promptly make the prediction and diagnosis of inflammation of the progress to fibrosis is a difficult problem all the time, Traditional diagnostic methods included clinical symptoms and signs, imaging and histopathology, However, they were subject to some restrictions on the clinical application because the sensitivity and specificity of these methods were inadequate and patients dependence were poor. The purpose of this project is to try to use the proteomic approach (Surface-enhanced laser desorption/ionisation time-of-flight mass spectrometry, SELDI-TOF-MS) to compare fibrosis staging samples with the normal ones, screened differentially expressed protein which can reflect the degree of liver fibrosis, ultimately Provide new clues for early clinical diagnosis of liver fibrosis.Part 1:Screen differentially expressed protein peaks between liver fibrosis and the normal controlObjectives To find differentially expressed protein peaks in peripheral blood for each Pathological stage of liver fibrosis.Methods 61 patients with liver fibrosis (31 belonged to S1 stage of fibrosis,25 belonged to S2/3 stage of fibrosis,5 belonged to S4 stage of fibrosis) and 20 normal were recruited, Surface-enhanced laser desorption/ionisation time-of-flight mass spectrometry (SELDI-TOF-MS) was applied to generate the serum protein spectra, furthermore biomarker Wizard software was used to screen differentially expressed protein peaks for the selected serum protein spectra.Results at normal control versus S1 stage of fibrosis,19 differentially expressed protein peaks were found, of which 6 were up-regulated and 13 were down-regulated; at normal control versus S2/3 stage of fibrosis,15 peaks were found,8 were up-regulated and 7 were down-regulated; at normal control versus S4 stage of fibrosis,14 peaks were found,9 were up-regulated and 5 were down-regulated; at S1 versus S2/3 stages of fibrosis,11 peaks were found,5 were up-regulated and 6 were down-regulated; at S1 versus S4 stages of fibrosis, 9 peaks were found,7 were up-regulated and 2 were down-regulated; at S2/3 versus S4 stages of fibrosis,3 peaks were found and they were all up-regulated. M8471 were down-regulated and M5304 were all up-regulated throughout the fibrosis stages.Conclusion The proteins with M8471 and M5304 might be potential peripheral serum markers of liver fibrosis.Part 2:Establish the SELDI diagnostic model for liver fibrosisObjectives To establish the best SELDI diagnostic models for liver fibrosis. Methods Differentially expressed protein peaks of part one were made a linear classification by Biomarker Pattern software, after experimental parameters were optimized the best diagnostic models were determined. Results at normal control versus S1 stage of fibrosis, M4799 and M6125 were considered the best diagnostic model after these two protein peaks were combined together, accuracy rate of diagnosis was 87%, the Specificity was 80% and Sensitivity was 87%; at normal control versus S2/3 stage of fibrosis, M6215 was considered the best diagnostic model, accuracy rate of diagnosis was 84%, the Specificity was 65% and Sensitivity was 84%; at normal control versus S4 stage of fibrosis, M8591 was considered the best diagnostic model, accuracy rate of diagnosis was 92%, the Specificity was 90% and Sensitivity was 100%; at S1 versus S2/3 stages of fibrosis, M6123 was considered the best diagnostic model, accuracy rate of diagnosis was 75%, the Specificity was 67% and Sensitivity was 76%; at S1 versus S4 stages of fibrosis, M4603 was considered the best diagnostic model, accuracy rate of diagnosis was 92%, the Specificity was 84%and Sensitivity was 100%; at S2/3 versus S4 stages of fibrosis, M4797 was considered the best diagnostic model, accuracy rate of diagnosis was 100%, the Specificity was 100% and Sensitivity was 80%.Conclusion the SELDI diagnostic model used to distinguish between normal control and fibrosis may become one of the methods available for diagnosis of liver fibrosis.