Helicobacter Pylori and Vonoprazan Dual Therapy

Upper gastrointestinal endoscopy examinations were performed for all participants and biopsies from gastric antrum and body were obtained. All the collected gastric biopsy specimens were sent to Jiangxi Provincial Key Laboratory of Digestive Diseases, First Affiliated Hospital of Nanchang University for susceptibility testing of antibiotics. The minimum inhibitory concentrations of antibiotics (amoxicillin, metronidazole, clarithromycin, levofloxacin and tetracycline) were determined by E-test. The inhibition zone for furazolidone was determined by the Kirby-Bauer disc diffusion method. Detailed procedure for detecting antibiotics resistance were consistent with our previous study. A strain was considered as resistant if minimum inhibitory concentration >0.125 μg/mL for amoxicillin, >8 μg/mL for metronidazole, ≥1 μg/mL for clarithromycin, >2 μg/mL for levofloxacin, ≥2 μg/mL for tetracycline, the inhibition zone was ≤7mm for furazolidone.

The detailed demographics and characteristics (sex, age, nationality, height, weight, education status, dwelling area, history of smoking and alcohol etc.) were recorded. The treatment-related adverse events (TEAEs) were recorded. We defined adherence as good if the participants took ≥80% drugs of the regimen during the consecutive 14 days. The H. pylori status after therapy was evaluated by ¹³C-UBT at least 6 weeks after completion of treatment. Proton pump inhibitors and antibiotics were stopped at least 2 and 4 weeks before ¹³C-UBT, respectively.

The stool samples were collected at baseline (before treatment), week 2 (after eradication) and week 8-10 (confirmation of H. pylori status). We sent the stool samples from subjects with successful eradication for metagenome DNA extraction and shotgun sequencing to avoid the influence of H. pylori eradication failure on gut microbiota. Briefly, OMEGA Mag-Bind Soil DNA Kit (Omega Bio-Tek, Norcross, GA, USA) was used to extract total microbial genomic DNA samples. Metagenome shotgun sequencing libraries from extracted microbial DNA was constructed by Illumina TruSeq Nano DNA LT Library Preparation Kit, which was then sequenced by Illumina NovaSeq platform (Illumina, USA) with PE150 strategy at Personal Biotechnology Co., Ltd. (Shanghai, China).

Raw sequencing reads were subjected to processing to yield quality-filtered reads suitable for further analysis, including removal of adapter and low-quality reads. Subsequently, minimap2 was utilized to align reads to the host genome of human and eliminate host contamination. Gene prediction was carried out on the generated contigs from each sample, whose translated protein sequences were subsequently pooled and clustered using mmseqs2. The lowest common ancestor taxonomy of the non-redundant genes was ascertained using mmseqs2 in "taxonomy" mode, by aligning them against a customized database comprising protein sequences of bacteria from GTDB (release 207: https://data.ace.uq.edu.au/public/gtdb/data/releases/), fungi from NCBI-nr (https://ftp.ncbi.nlm.nih.gov/blast/db/FASTA/), and viruses from RVDB (version 24.1: https://rvdb.dbi.udel.edu/download/). In order to assess the abundance of genes, high-quality reads from each sample were mapped onto the contigs using minimap2 and read counts were computed using htseq. Abundance values in metagenomes were normalized using copies per kilobase per million mapped reads. Clean high-quality reads were processed and profiled with ARGs-OAP (version 2.0) by querying against the SARG (version 3.0-F) database, which is a structural antimicrobial resistance genes database containing 32 types and 2842 subtypes of antibiotic resistance genes. In order to perform the quantification and downstream analysis of diversity indices, resistome profiling, and prevalence ranking, the abundance of antibiotic resistance genes at type and subtype level were normalised to the number of 16S rRNA genes.