Autoimmune Hepatitis Biomarkers In Pharmaceutical Screening And Intervention-based Metabolomics Research

In recent years, the incidence of autoimmune hepatitis (AIH) is dramatically increasing. AIH is considered as an immune-related disease, including a progressive necroinflammatory and fibrotic process in the liver. Many AIH patients can rapidly progress to cirrhosis and liver failure. It is important to make a quick and accurate diagnosis and treatment to prevent the progression of AIH. However, due to unknown pathogenesis, atypical clinical manifestations and overlap syndrome, the current diagnostic methods for AIH still have many problems, such as complex and time-consuming diagnostic procedures, limited sensitivity and specificity. Therefore, there is an obvious clinical need for new methods that help to diagnose AIH.In this paper, we try to use the method of1H NMR and LC-MS/MS Q-TOF to screen the diagnosis biomarkers of AIH in order to provide new ideas and techniques for the rapid diagnosis of AIH. The main contents and conclusions are as follows:1. The medical database of AILD was established, and analysised the main clinical features of the three kinds of autoimmune liver disease.In this study, the medical records of AILD of the302hospital from2005to2010were systematically organized. The medical database of AILD was established. The database contains a total of1866cases of patients with AILD. which include the patients with AIH676, PBC912and OS278respectively. The proportion of AIH cases in hospital admissions rose from0.65%to0.96%from2005to2010. The clinical symptoms of three autoimmune liver diseases lack of specificity, and the number of liver biopsy examination was relatively small. The establishment of the database would greatly facilitate the epidemiological studies of kinds of AILD.2. The diagnostic models for AIH based on1H-NMR metabonomics of plasma were established. In this part, LC-MS/MS Q-TOF and NMR techniques were employed. Small molecule metabolites of plasma and urine samples from the same group of AIH, PBC and healthy people were detected. At last, diagnostic models for AIH, primary biliary cirrhosis (PBC) and healthy population were established, and the diagnostic models based on1H NMR metabolomics technology showed potential advantages in the diagnosis of AIH.3. By repeated screening,10biomarkers were obtained, and which showed a high sensitivity and specificity in the diagnosis of AIH.First,256variables were screened from1802chemical shift through multi-criteria assessment (MCA) methods, then61variables were eventually locked to establish diagnostic model for patients. In the diagnosis of AIH, PBC, overlap syndromes (OS) and drug-induced liver injury (DILI), the sensitivity, specificity and accuracy of the model were90.5%,98.3%and96.2%respectively; In the diagnosis of AIH and DILI, the sensitivity, specificity and accuracy of the model were95.0%,100%and97.4%respectively; In the diagnosis of AIH and PBC, the sensitivity and specificity were100%, and accuracy was100%; In the diagnosis of AIH and OS, the sensitivity, specificity and accuracy were100%,95.0%and97.3%respectively.Further, identified the61variables corresponding to the10metabolic markers, they were citrate, glutamine, glutamate, acetone, pyruvate, β-Hydroxyisobutyrate, acetoacetate, histidine, dimethylamine and creatinine.4. It was verified that the selected biomarkers could predict the prognosis of AIH by evaluating the therapeutic effect of the Yu Gan granules combined with prednisone.In view of the serum antibody cannot well predict the prognosis of AIH, patients with AIH were treated with Yu Gan granules combined with prednisone. The results showed that the selected biomarkers could well predict the prognosis of AIH. And the therapeutic effect of the combination therapy was better than prednisone monotherapy, especially in lowering aminotransferase and improving quality of life.This is the first report suggesting that an1H-NMR analysis of human plasma samples has potential application in the diagnosis of AIH. The promising predictability of the method suggests that it would augment the currently available diagnostic approaches.