A comprehensive model to better screen out antiviral treatment candidates for chronic hepatitis B patients.

BACKGROUND

Chronic hepatitis B virus (HBV) infection is a serious human health threat given its high morbidity and mortality. Timely and effective antiviral treatment can postpone liver disease progression and reduce the occurrence of HBV-related end-stage liver disease. At present, the antiviral treatment criteria are mainly based on alanine transaminase (ALT) levels, HBV DNA levels and HBV e antigen levels according to the American Association for the Study of Liver Diseases treatment guidelines. However, some chronic hepatitis B (CHB) patients not meeting the above criteria still experience liver disease progression without antiviral treatment. It is urgent to identify a more comprehensive tool to screen out more antiviral treatment candidates as soon as possible.

METHODS

Considering the vital role of the immune response in the development of HBV infection and CHB cure, we collected data from 335 treatment-naïve CHB patients and comprehensively analysed their clinical and immune traits (including innate and adaptive responses). The immune parameters were obtained by flow cytometry. Finally, we established a model that can better distinguished CHB patients who need treatment through machine learning and LASSO regression of serological and immune parameters.

RESULTS

Through a series of analyses, we selected four important clinical parameters (ALT, HBV DNA, the Fibroscan value, and the A/G ratio) and four immune indicators (NKbright + NKp44+, NKbright + NKG2A+, NKT+GranzymeB+, and CD3 + CD107a + ) from more than 200 variables and then successfully established a mathematical model with high sensitivity and specificity to better screen out antiviral treatment candidates from all CHB patients.

CONCLUSIONS

Our results developed a refined model to better screen out antiviral treatment candidates from all CHB patients by combining common clinical parameters and important immune indicators, including innate and adaptive immunity. These findings provide more information for improving treatment guidelines and have potential implications for the timing of antiviral therapy to achieve better virus control and reduce the occurrence of end-stage liver disease.