Gut microbiota as prognosis markers for patients with HBV-related acute-on-chronic liver failure

Abstract

The gut microbiota in the hepatitis B virus related acute-on-chronic liver failure (HBV-ACLF) is poorly defined. We aim to uncover the characteristics of the gut microbiota in HBV-ACLF and in other HBV associated pathologies. We analyzed the gut microbiome in patients with HBV-ACLF or other HBV associated pathologies and healthy individuals by 16S rRNA sequencing and metagenomic sequencing of fecal samples. 212 patients with HBV-ACLF, 252 with chronic hepatitis B (CHB), 162 with HBV-associated cirrhosis (HBV-LC) and 877 healthy individuals were recruited for the study. CHB and HBV-LC patients are grouped as HBV-Other. We discovered striking differences in the microbiome diversity between the HBV-ACLF, HBV-Other and healthy groups using 16S rRNA sequencing. The ratio of cocci to bacilli was significantly elevated in the HBV-ACLF group compared with healthy group. Further analysis within the HBV-ACLF group identified 52 genera showing distinct richness within the group where Enterococcus was enriched in the progression group whilst Faecalibacterium was enriched in the regression group. Metagenomic sequencing validated these findings and further uncovered an enrichment of Lactobacillus casei paracasei in progression group, while Alistipes senegalensis, Faecalibacterium prausnitzii and Parabacteroides merdae dominated the regression group. Importantly, our analysis revealed that there was a rapid increase of Enterococcus faecium during the progression of HBV-ACLF. The gut microbiota displayed distinct composition at different phases of HBV-ACLF. High abundance of Enterococcus is associated with progression while that of Faecalibacterium is associated with regression of HBV-ACLF. Therefore, the microbiota features hold promising potential as prognostic markers for HBV-ACLF.

Keywords: 16S rRNA sequencing; Hepatitis B; acute-on-chronic liver failure; gut microbiota; metagenomic sequencing.

Figure 1.

Gut microbiota distribution among groups. (a) Alpha diversity analysis (P = 3.57E-06). (b) A t-distributed stochastic neighbor embedding (t-SNE) visualization (P = .001). (c) Lef Se analysis showed predominant gut microbiota. (d) The ratio of cocci to bacilli was compared among the three groups. **P < .01

Figure 2.

A classification model for the healthy, HBV-Other and the HBV-ACLF group. (a) 18 most important taxa in the classification model among the 3 groups. (b) The decomposition visualization of the 18 most important taxa among the 3 groups

Figure 3.

The abundance of the genera with the highest richness Enterococcus and Faecalibacterium. The relative abundance of Enterococcus was significantly elevated in the progression group, and that of Faecalibacterium was significantly elevated in the regression group

Figure 4.

Correlation between the gut microbiota and clinical indicators. The common genera associated with ALT/AST (relevant to liver inflammation), TBIL/INR/MELD (relevant to liver disease severity) and WBC/NEUT%/PCT (relevant to the degree of infection), respectively. Blue square represents the common genera selected by the trained Regressor that with clinical relevant

Figure 5.

Difference of gut microbiota between the progression and regression groups. Relative abundance of genus at day-1 (fecal sample collected at day 1 after admission), day 7 and day 14 between the progression and regression group of HBV-ACLF. (a) Enterococcus and Faecalibacterium, and (b) species Lactobacillus casei paracasei, Alistipes senegalensis, Faecalibacterium prausnitzii and Parabacteroides merdae. *P < .05

Figure 6.

The key species of different gut microbiota were further validated by qPCR. qPCR validation of the relative abundance of Enterococcus faecium, Faecalibacterium prausnitzii, Lactobacillus casei paracasei, Clostridium citroniae and Dorea longicatena between progression and regression group of HBV-ACLF. **P < .01