Screening of Hepatocellular Carcinoma Patients with High Risk of Early Recurrence After Radical Hepatectomy Using a Nomogram Model Based on the γ-Glutamyl Transpeptidase-to-Albumin Ratio

Background and purpose The present study aimed to establish a γ-glutamyl transpeptidase-to-albumin ratio (GAR)-based nomogram model to predict early recurrence of hepatocellular carcinoma (HCC) after radical surgery. Methods Patients enrolled in this study were randomly allocated into a train and validation cohort in a ratio of 7:3. The Least Absolute Shrinkage and Selection Operator (LASSO) proportional hazards model and cox regression model were combined to identify independent risk factors related to HCC recurrence. Based on these risk factors, a predictive nomogram was constructed and validated in both inner and outer test cohorts. The performance of the nomogram was evaluated by C-index, the area under the receiver operating characteristic curve (AUC), the calibration curve and decision curve analysis. Results The tumor size, tumor number, BCLC stage, microvascular invasion (MVI) and GAR value were identified as independent risk factors related to HCC recurrence and used to construct the predictive nomogram. AUC of the nomogram showed satisfactory accuracy in predicting 1-, 3- and 5-year disease-free survival. The calibration curve showed agreement between the ideal and predicted values. The risk score more than 72 as calculated by the nomogram was related to early recurrence of HCC after radical surgery. DCA plots showed better clinical usability of the nomogram as compared with the BCLC staging system in all three included cohorts. Conclusion The nomogram based on the GAR value may provide a new option for screening of the target HCC cohort of patients who need anti-recurrence therapy after surgery.