A Model to Identify Grey-Zone Patients with Chronic Hepatitis B Requiring Antiviral Therapy: A Multicentre, Retrospective Study

Abstract

Background Individuals who do not match any specific stage in the natural history of chronic hepatitis B are classified into the grey-zone, and appropriate management for these patients remains unclear. This study aimed to develop and validate a non-invasive model to identify grey-zone patients requiring antiviral therapy (AVT). Methods We retrospectively collected data on 200 grey-zone patients not requiring AVT (according to assessment by non-invasive parameters from 2010 to 2023 in six hospitals) and randomised them into development (n=140) and validation (n=60) cohorts. Univariable and multivariable regression analyses were performed to identify independent variables for establishing a nomogram to predict the probability of requiring AVT by liver biopsy, which was assessed using the area under the receiver operating characteristic curve (AUC), calibration plot analysis and decision curve analysis. Results Seventy-eight patients (n=39%) were identified as requiring AVT. Age [odds ratio(OR) 1.06, 95% confidence interval(CI) 1.01–1.11], alanine aminotransferase (OR 2.43, 95%CI 1.08–5.59), lymphocyte percentage (OR 6.43, 95%CI 1.23–33.64), platelet count (OR 0.99, 95%CI 0.98–0.1.00) and international normalised ratio per (0.01) (OR 0.99, 95%CI 0.98–0.1.00) were identified as independent variables for constructing the nomogram, which showed good discriminability (development dataset: AUC=0.755; validation dataset: AUC=0.707), calibration and clinical applicability. Patients with nomogram scores >197 and ≤132 were considered to have a high and low probability of needing AVT, respectively. Conclusions Grey-zone patients requiring AVT should be identified, and the model developed here is a promising tool. Trial registration ClinicalTrials.gov, NCT06041022