Machine-learning based patient classification using Hepatitis B virus full-length genome quasispecies from Asian and European cohorts

Abstract

Chronic infection with Hepatitis B virus (HBV) is a major risk factor for the development of advanced liver disease including fibrosis, cirrhosis, and hepatocellular carcinoma (HCC). The relative contribution of virological factors to disease progression has not been fully defined and tools aiding the deconvolution of complex patient virus profiles is an unmet clinical need. Variable viral mutant signatures develop within individual patients due to the low-fidelity replication of the viral polymerase creating ‘quasispecies’ populations. Here we present the first comprehensive survey of the diversity of HBV quasispecies through ultra-deep sequencing of the complete HBV genome across two distinct European and Asian patient populations. Seroconversion to the HBV e antigen (HBeAg) represents a critical clinical waymark in infected individuals. Using a machine learning approach, a model was developed to determine the viral variants that accurately classify HBeAg status. Serial surveys of patient quasispecies populations and advanced analytics will facilitate clinical decision support for chronic HBV infection and direct therapeutic strategies through improved patient stratification.

Application of Artificial Intelligence for the Diagnosis and Treatment of Liver Diseases

“…Recently, ultra‐deep sequencing of HBV genome across two Asian and European populations was applied to a random‐forest algorithm to classify seroconversion status with high accuracy. ( 48 )…”

Recurrent HBV Integration Targets as Potential Drivers in Hepatocellular Carcinoma

“…Lastly, because it is known that HBV behavior differs in Asian and Western cohorts [60], the geographic region where patient specimens were collected is indicated for each study. The majority of studies (72.7%) collected patient specimens in Asian populations, while only 6.0% examined Western cohorts.…”

Critical Updates on Chronic Hepatitis B Virus Infection in 2021

“…Since carcinogenesis is an important aspect in the natural history of HBV-related infection, the novel HBV DNA quantitation-time index (HDQTI), comprising HBV DNA quantitation and follow up, was found to predict HBV associated liver cancer prognosis identified a cut-off value at 34. The HDQTI also predicted cancer recurrence and the need for shorter surveillance intervals with appropriate imaging in patients with a high score [ 122 ]. Liver biopsy can be avoided in a significant number of patients with use of the combined ELF™ (based on Fibroscan®) algorithm.…”