Dataset Information

Pilot Study Using Machine Learning to Identify Immune Profiles for the Prediction of Early Virological Relapse After Stopping Nucleos(t)ide Analogues in HBeAg-Negative CHB.

ABSTRACT: Treatment with nucleos(t)ide analogues (NAs) may be stopped after 1-3 years of hepatitis B virus DNA suppression in hepatitis B e antigen (HBeAg)-negative patients according to Asian Pacific Association for the Study of Liver and European Association for the Study of Liver guidelines. However, virological relapse (VR) occurs in most patients. We aimed to analyze soluble immune markers (SIMs) and use machine learning to identify SIM combinations as predictor for early VR after NA discontinuation. A validation cohort was used to verify the predictive power of the SIM combination. In a post hoc analysis of a prospective, multicenter therapeutic vaccination trial (ABX-203, NCT02249988), hepatitis B surface antigen, hepatitis B core antigen, and 47 SIMs were repeatedly determined before NA was stopped. Forty-three HBeAg-negative patients were included. To detect the highest predictive constellation of host and viral markers, a supervised machine learning approach was used. Data were validated in a different cohort of 49 patients treated with entecavir. VR (hepatitis B virus DNA ≥ 2,000 IU/mL) occurred in 27 patients. The predictive value for VR of single SIMs at the time of NA stop was best for interleukin (IL)-2, IL-17, and regulated on activation, normal T cell expressed and secreted (RANTES/CCL5) with a maximum area under the curve of 0.65. Hepatitis B core antigen had a higher predictive power than hepatitis B surface antigen but lower than the SIMs. A supervised machine-learning algorithm allowed a remarkable improvement of early relapse prediction in patients treated with entecavir. The combination of IL-2, monokine induced by interferon γ (MIG)/chemokine (C-C motif) ligand 9 (CCL9), RANTES/CCL5, stem cell factor (SCF), and TNF-related apoptosis-inducing ligand (TRAIL) was reliable in predicting VR (0.89; 95% confidence interval: 0.5-1.0) and showed viable results in the validation cohort (0.63; 0.1-0.99). Host immune markers such as SIMs appear to be underestimated in guiding treatment cessation in HBeAg-negative patients. Machine learning can help find predictive SIM patterns that allow a precise identification of patients particularly suitable for NA cessation.

SUBMITTER: Wubbolding M

PROVIDER: S-EPMC7789842 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Many factors are associated with post-treatment relapse in CHB patients, and there are no effective factors to predict relapse. In this study, we investigate the influence factors associated with post-treatment relapse and their predictive value in HBeAg positive CHB (eP-CHB).The factors associated with post-treatment relapse were analyzed firstly by a retrospective study in eP-CHB. Variables included age, sex, regimen, baseline HBeAg and HBV DNA level, total course of treatment as well as duration of consolidation therapy after HBeAg seroconversion. The predictive effects of the influence factors were evaluated in an eP-CHB prospective cohort.89 patients were enrolled in the retrospective study, 42(47.2%) relapsed after discontinuation of treatment. Factors related to post-treatment relapse were total course of treatment, duration of consolidation therapy and baseline HBV DNA level. Relapse rates in patients with total course >36 months, consolidation duration >12 months and baseline HBV DNA level < 1.0E+5IU/ml were lower than those of total course <24 months (P = 0.002), consolidation duration?12 months (P = 0.011) and baseline HBV DNA level > 1.0E+7IU/ml (P = 0.01) respectively. Patients with HBV DNA?1.0E+7IU/ml plus HBeAg<200COI at baseline had the highest relapse rate and cumulative relapse rate than the other three arms (P = 0.048 and 0.008 respectively). Logistic regression analysis demonstrated that baseline HBV DNA level, duration of consolidation therapy and combination of baseline HBV DNA and HBeAg (IgDNA/IgHBeAg) were independent factors to predict post-treatment relapse. The model based on baseline IgDNA/IgHBeAg and consolidation duration worked well in predicting post-treatment relapse in the prospective study and the accuracy, specificity, sensitivity, PPV and NPV for prediction were 80.3%, 81.1%, 79.2%, 73.1% and 85.7% respectively.Virological factors including baseline HBV DNA, HBeAg and treatment course were major influence factors associated with post-treatment relapse in eP-CHB. Patients with higher HBV DNA and lower HBeAg levels at baseline, shorter total course as well as consolidation therapy were more likely to develop relapse after discontinuation of therapy. The antiviral therapy in eP-CHB patients should be individually managed at different levels. It is better to treat those with higher viral load and lower HBeAg levels at baseline for a longer course, especially longer consolidation duration so as to decrease the relapse rate.

Project description:BackgroundGut dysbiosis is present in chronic hepatitis B virus (HBV) infection. In this study, we integrated microbiome and metabolome analysis to investigate the role of gut microbiome in virological response to nucleos(t)ide analogues (NAs) treatment.MethodsChronic HBV patients were prospectively recruited for steatosis and fibrosis assessments via liver elastography, with full-length 16S sequencing performed to identify the compositional gut microbiota differences. Fasting plasma bile acids were quantified by liquid chromatography-tandem mass spectrometry.FindingsAll patients (n = 110) were characterized into three distinct microbial clusters by their dominant genus: c-Bacteroides, c-Blautia, and c-Prevotella. Patients with c-Bacteroides had a higher plasma ursodeoxycholic acids (UDCA) level and an increase in 7-alpha-hydroxysteroid dehydrogenase (secondary bile acid biotransformation) than other clusters. In NAs-treated patients (n = 84), c-Bacteroides was associated with higher odds of plasma HBV-DNA undetectability when compared with non-c-Bacteroides clusters (OR 3.49, 95% CI 1.43-8.96, p = 0.01). c-Blautia was positively associated with advanced fibrosis (OR 2.74, 95% CI 1.09-7.31, p = 0.04). No such associations were found in treatment-naïve patients. Increased Escherichia coli relative abundance (0.21% vs. 0.03%, p = 0.035) was found in on-treatment patients (median treatment duration 98.1 months) with advanced fibrosis despite HBV DNA undetectability. An enrichment in l-tryptophan biosynthesis was observed in patients with advanced fibrosis, which exhibited a positive correlation with Escherichia coli.InterpretationCollectively, unique bacterial signatures, including c-Bacteroides and c-Blautia, were associated with virological undetectability and fibrosis evolution during NAs therapy in chronic HBV, setting up intriguing possibilities in optimizing HBV treatment.FundingThis study was supported by the Guangdong Natural Science Fund (2019A1515012003).

Dataset Information

Pilot Study Using Machine Learning to Identify Immune Profiles for the Prediction of Early Virological Relapse After Stopping Nucleos(t)ide Analogues in HBeAg-Negative CHB.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets