ENPP1 and IFIT2 in PBMCs as early predictive biomarkers for HBsAg clearance and responses to Peg-IFN-α in HBeAg-negative chronic hepatitis B patients.

OBJECTIVE

To investigate whether early on-treatment expression of ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1) and interferon-induced protein with tetratricopeptide repeats 2 (IFIT2) in peripheral blood mononuclear cells (PBMCs) could serve as biomarkers for predicting hepatitis B surface antigen (HBsAg) clearance and therapeutic responses to pegylated interferon-α (Peg-IFN-α) in HBeAg-negative chronic hepatitis B (CHB) patients.

METHODS

In this prospective study, HBeAg-negative CHB patients received Peg-IFN-α therapy for 48 weeks. Patients were categorized into virological response (VR) or non-virological response (NVR) groups based on changes in HBV DNA and HBsAg levels at week 48, and further stratified into serological response (SR) or non-serological response (NSR) groups based on the occurrence of HBsAg loss or seroconversion. ENPP1 and IFIT2 mRNA levels in PBMCs were quantified using real-time quantitative PCR. Correlations between gene expression and HBsAg reduction were evaluated using Spearman's rank correlation analysis. Factors independently associated with VR and SR were identified using univariate and multivariate logistic regression analyses. The discriminative ability of individual biomarkers and combined predictive models was assessed by receiver operating characteristic (ROC) curve analysis.

RESULTS

Both ENPP1 and IFIT2 mRNA levels were significantly elevated in the VR and SR groups during early treatment. Expression levels of both genes at weeks 12 and 24 positively correlated with the magnitude of HBsAg reduction, and their early induction at week 12 was associated with subsequent HBsAg decline. Multivariate analyses identified ENPP1 and IFIT2 mRNA levels at weeks 12 and 24 as independent predictors of VR and SR. ENPP1 expression at week 12 demonstrated the highest predictive accuracy for VR (AUC = 0.7645), while ENPP1 at week 24 showed moderate predictive value for SR (AUC = 0.7399). In contrast, IFIT2 consistently exhibited superior predictive performance for VR (AUC = 0.8791) and for SR at week 24 (AUC = 0.8879). The combination of ENPP1 and IFIT2 further improved predictive accuracy for both VR (AUC = 0.9098) and SR (AUC = 0.9217).

CONCLUSIONS

ENPP1 and IFIT2 in PBMCs represent valuable biomarkers for early prediction of Peg-IFN-α treatment efficacy and HBsAg clearance in HBeAg-negative CHB patients.