Exploration of nucleos(t)ide analogs cessation in chronic hepatitis B patients with hepatitis B e antigen loss

Evaluation of chronic hepatitis B patients who voluntarily discontinued oral antiviral therapy: is there an answer to the controversial topic?

OBJECTIVE

The uncertain treatment duration for nucleos(t)ide analogues (NA) used in the treatment of chronic hepatitis B (CHB) is an important problem for both patients and physicians. The aim of this study was to evaluate the determinants of virologic relapse (VR) and the optimum time of treatment discontinuation in the follow-up of CHB patients who voluntarily discontinued treatment after virological suppression was achieved under NA use.

METHODS

Data from 138 patients from 11 centers were included in this registry-based study. Factors associated with VR were investigated using multivariate Cox regression analysis.

RESULTS

Ninety-nine (71.7%) of the patients were HBeAg (Hepatitis B e antigen) negative. During the 24-month follow-up period after treatment discontinuation, VR occurred in 58.7% (n = 81) of all patients and 57.6% (n = 57) of HBeAg-negative patients. The duration of NA treatment was significantly shorter (cutoff 60 months) in HBeAg-negative patients who later developed VR. In addition, the duration of virologic remission achieved under NA treatment was significantly shorter (cutoff 52 months) in those who later developed VR. In the Cox multivariate regression model of HBeAg-negative patients, having less than 60 months of NA treatment (HR = 2.568; CI:1.280-5.148; P  = 0.008) and the levels of alanine aminotransferase being equal to or higher than twice the upper level of normal at the beginning of treatment (HR = 3.753; CI:1.551-9.081; P  = 0.003) were found to be statistically significant and independently associated with VR.

CONCLUSION

The findings of this study may provide clinical guidance in terms of determining the most appropriate discontinuation time for NA.

Higher HBeAg-reversion virological relapse and lower sustained remission after treatment cessation in tenofovir-treated HBeAg-positive patients

A complete investigation of the clinical outcomes after treatment cessation in HBeAg-positive patients with HBeAg loss is limited. We retrospectively recruited 242 HBeAg-positive patients with HBeAg loss after a median duration of 37.2 months with tenofovir (TDF, n = 77) or entecavir (ETV, n = 165) treatment. There were 77 (31.8%) patients with sustained virological remission (SVR), 85 (35.1%) with HBeAg-reversion virological relapse, 80 (33.1%) with HBeAg-negative virological relapse after treatment cessation, and 23 (9.5%) with HBsAg loss. Clinical data at baseline, on-treatment and during off-treatment follow-up were analyzed. The 3-year cumulative incidences of overall, HBeAg-reversion and HBeAg-negative virological relapse were 70.2%, 54%, and 53.5%, respectively. The common factors associated with HBeAg-reversion and HBeAg-negative virological relapse were tenofovir treatment (hazard ratio [HR] = 5.411, p < 0.001; HR = 2.066, p = 0.006, respectively) and HBsAg at end of treatment (EOT) (HR = 1.461, p = 0.001; HR = 1.303, p = 0.019, respectively). The 5-year cumulative incidence of HBsAg loss in SVR patients was 13.7% and EOT HBsAg was the only associated factor (HR = 0.524, p = 0.024). Compared to that of ETV-treated patients, TDF-treated patients had a significantly higher 3-year cumulative incidence of virological relapse (87.3% vs. 62.8%, p < 0.001), earlier HBeAg-reversion virological relapse (2.9 vs. 7.8 months, p < 0.001), a higher rate of HBeAg-reversion virological relapse (53.2% vs. 26.7%) and a lower SVR rate (15.6% vs. 39.4%) (p < 0.001). In summary, the clinical outcomes after treatment cessation in HBeAg-positive patients with HBeAg loss were composed of HBeAg-reversion virological relapse, HBeAg-negative virological relapse and SVR. TDF was significantly associated with off-treatment virological relapse. EOT HBsAg plays an important role in HBsAg loss among SVR patients and posttreatment virological relapse.

Use of a Deep Learning Approach for the Sensitive Prediction of Hepatitis B Surface Antigen Levels in Inactive Carrier Patients

Deep learning is a subset of machine learning that can be employed to accurately predict biological transitions. Eliminating hepatitis B surface antigens (HBsAgs) is the final therapeutic endpoint for chronic hepatitis B. Reliable predictors of the disappearance or reduction in HBsAg levels have not been established. Accurate predictions are vital to successful treatment, and corresponding efforts are ongoing worldwide. Therefore, this study aimed to identify an optimal deep learning model to predict the changes in HBsAg levels in daily clinical practice for inactive carrier patients. We identified patients whose HBsAg levels were evaluated over 10 years. The results of routine liver biochemical function tests, including serum HBsAg levels for 1, 2, 5, and 10 years, and biometric information were obtained. Data of 90 patients were included for adaptive training. The predictive models were built based on algorithms set up by SONY Neural Network Console, and their accuracy was compared using statistical analysis. Multiple regression analysis revealed a mean absolute percentage error of 58%, and deep learning revealed a mean absolute percentage error of 15%; thus, deep learning is an accurate predictive discriminant tool. This study demonstrated the potential of deep learning algorithms to predict clinical outcomes.