| Objective:In previous studies,it has been shown that autophagy is involved in various aspects of the pathogenesis of primary biliary cholangitis(PBC)and may have a relationship with the progression and prognosis of PBC.Therefore,the aim of this study was to analyze and clarify the correlation between autophagy and the clinical features and disease progression of primary biliary cholangitis,and to investigate and discuss the predictive value of autophagy on the response to ursodeoxycholic acid(UDCA)therapy in patients with PBC.Methods:A retrospective case-control study was conducted on patients with PBC who were admitted to the Department of Gastroenterology of the First Affiliated Hospital of Air Force Military Medical University from January 2005 to January 2022 and first diagnosed with hepatic puncture electron microscopy data;466 patients were screened and included as an autophagy-related cohort,and divided into autophagy and non-autophagy groups based on whether autophagosomes were observed under hepatic puncture electron microscopy;in the autophagy Patients with incomplete post-treatment data and treatment irregularities were excluded from the relevant cohort,and 260 patients were further included as aresponse-related cohort,which was divided into good response group and poor response group based on ALP≥1.67 ULN as an important criterion for poor response to UDCA treatment;relevant clinical data indicators at baseline and one year after treatment were collected,mainly including:PBC patients’ demographic characteristics,hematological indicators,immunoglobulins,autoantibodies,histological stage of liver biopsy and electron microscopy results of liver penetration,liver fibrosis score,and UDCA treatment response risk score.The relevant indexes were compared between the two groups,and the risk factors influencing patients to develop poor response to UDCA treatment were screened by binary logistic regression analysis.Based on the above factors,a clinical prediction model was constructed to predict the occurrence of poor UDCA response in PBC patients,and the model was evaluated.The results of the prediction model were further visualized by introducing column line plots to enhance the clinical application of the model.In addition,the correlation between autophagy and PBC and the applicability of the constructed model were further analyzed and validated by subgroup analysis.Results:1.The levels of ALT,AST,TBIL,ALP,GGT and IgM in the autophagy group were higher than those in the non-autophagy group;the levels of TP,ALB and IgG were lower than those in the non-autophagy group(P<0.05);the proportion of patients with early pathological stage in the autophagy group was lower than that in the non-autopha gygroup,and the proportion of patients with late pathological stage in the autophagy group was higher than that in the non-autophagy group(P<0.05).2.The proportion of patients with autophagosomes detected by electron microscopy was significantly higher in the poor responder group than in the good responder group,and the levels of ALT,AST,TBIL,ALP,GGT and the scores of APRI,FIB-4,GLOBE and UK-PBC were significantly higher in the poor responder group than in the good responder group,and the levels of ALB were significantly lower than in the good responder group(P<0.05);the proportion of patients with early pathological stage was lower than that of the good responder group.The proportion of patients with early pathology was lower than that in the good-response group,and the proportion of patients with late pathology was higher than that in the good-response group(P<0.05).3.A clinical prediction model was developed to predict the occurrence of poor response to UDCA treatment in patients with PBC.A total of four clinical variables were included in the model,namely autophagy,maximum upper normal value of ALT,maximum upper normal value of ALP,and GLOBE score.Among the patients,autophagosomes were detected(OR=3.019;95%CI:1.237-.366;P=0.015),ALT levels were elevated(OR=1.717;95%CI:1.059-2.783;P=0.028),ALP levels were elevated(OR=2.633;95%CI:1.848-3.942;P<0.001),and high level of GLOBE score(OR=3.876;95%CI:2.317-6.482;P<0.001)were independent risk factors for poor treatment response in patients with PBC.4.after constructing the UDCA poor treatment response prediction model based on the above factors,the ROC curve and calibration curve were used to evaluate the discrimination and calibration of the model.the AUC value of the area under the ROC curve was 0.861 with 95%confidence interval of 0.818-0.905,which demonstrated a fairly high diagnostic value;the calibration curve showed that the predicted values of the model were in good agreement with the actual observed values.In addition,a decision curve was used to evaluate the utility of the model,and the results showed that the model had good clinical validity.Column line plots were drawn to demonstrate the probability of poor response to UDCA treatment in patients with PBC,making the results more intuitive.5.The model cohort had a maximum Jorden index of 0.620 when the cutoff value of 0.205 was used,and the sensitivity and specificity of the model were 0.795 and 0.825,respectively,and the positive likelihood ratio and negative likelihood ratio were 4.547 and 0.249,respectively.0.205 group,and the differences between groups were compared.Age,autophagy detection rate,ALT,AST,TBIL,ALP,GGT,APRI,FIB-4 levels,GLOBE score,UK-PBC score,and UDCA poor response rate were lower in the model score<0.205 group and ALB levels were higher in the model score≥0.205 group;the proportion of patients with early pathology was higher than that in the model score≥0.205 group,and the proportion of patients with advanced pathology was lower than that of the model score≥0.205 group(P<0.05).6.Further exploration by subgroup analysis,stratified analysis of autophagy detection rate for response status and stratified analysis of predictive model score UDCA poor response rate,respectively,showed significant differences in autophagy detection rate between patients with good response and poor response in early and early-mid pathological stages,and the predictive model had good predictive The predictive model had good predictive value at different pathological stages and different stages of disease progression(P<0.05).Conclusion:Autophagy was found to correlate with the degree of disease and UDC A response in PBC patients,and patients with autophagosomes observed at the baseline of treatment by liver puncture electron microscopy had more severe disease and could be used as a risk factor to predict poor UDCA response.The autophagy-related prediction model constructed on the basis of the previous model can facilitate the early screening of patients with poor response to UDCA treatment and provide important clinical decision guidance for early initiation of second-line treatment combination intervention... |
PDF Full Text Request