Abstract: Objective To construct a nomogram predictive model for the prognosis of patients with primary biliary cholangitis (PBC) based on autoantibody markers. Methods A total of 112 PBC patients admitted to our hospital from January 2018 to March 2022 were retrospectively selected and included in the study. The patients were followed up for at least 2 years to assess the prognosis, analyze the prognostic factors, and construct a nomogram predictive model. The area under the receiver operating characteristic (ROC) was used to analyze the prediction efficiency. Results The proportion of age ≥65 years (52.78% vs. 28.95%), total bilirubin (TBil) level ≥22.2 μmol/L (72.22% vs. 42.11%), ascites (55.56% vs. 32.89%), Child-Pugh grade B/C (41.67%/30.56% vs. 32.86%/15.79%), anti-gp210 antibody positive (52.78% vs. 15.79%) and anti-sp100 antibody positive (47.22% vs. 19.74%) in the poor prognosis group were higher than those in the good prognosis group. The proportion of ursodeoxycholic acid (UDCA) response (50.00% vs. 80.26%) in the good prognosis group was lower than that in the good prognosis group (P<0.05). Logistic regression analysis showed that TBil (OR=3.652, 95%CI:1.315~10.139), anti-SP100 antibody positive (OR=4.716, 95%CI:2.018~11.019), UDCA response (OR=0.185, 95%CI:0.099~0.342) and anti-gp210 antibody positive (OR=6.245, 95%CI:3.035~12.846) were independent prognostic factors (P<0.05). The C-index of the nomogram model established based on the results of logistic regression was 0.826 (95%CI:0.741~0.910), and the calibration curve was close to the ideal curve (P>0.05). ROC showed that the specificity was 85.50%, the sensitivity was 86.10%, and the area under the curve (AUC) was 0.872 (95%CI:0.800~0.945) (P<0.05). Conclusion The nomogram model based on positive anti-gp210 antibody and positive anti-sp100 antibody can better predict the risk of poor prognosis of PBC patients.

Key words: Primary biliary cholangitis, Autoantibodies, Prognosis, Prediction model, Column chart