Identification and surveillance of antibiotic resistance in Helicobacter pylori in the UK
Publication Date
July 23, 2025
Creator
Abstract
Helicobacter pylori is the leading cause of peptic ulcers and gastric cancer. Treatment is becoming more problematic due to increasing antimicrobial resistance (AMR) rates globally. There is no routine surveillance of H. pylori antimicrobial sensitivities in the UK, and published data is lacking as culture-based sensitivity testing is time-consuming and technically involved. This study aimed to characterise antimicrobial sensitivities of isolates collected in Nottingham since 2001, and to compare levofloxacin phenotypic and genotypic resistance rates. Additionally, whole genome sequencing methods for determining antibiotic resistance in clinical isolates were investigated, and how that data can be used with online resistance detection tools to predict phenotypic resistance.
Gastric biopsy samples were collected, with informed written consent and ethics approval, from patients attending the Queen’s Medical Centre in Nottingham for an upper GI tract endoscopy. Antibiotic sensitivity was assessed using disc diffusion and E-test methods, with strain NCTC (National Collection of Type Cultures) 11637 as a control. Of 241 isolates tested, 27.8% were resistant to clarithromycin, 61.8% to metronidazole, and 2.5% to amoxicillin, which are used in first-line therapies. For those used in second-line therapies, 0% of isolates were resistant to tetracycline, and 4.1% to levofloxacin. Multi-drug resistance was found in 27.4% of isolates. Resistance to clarithromycin increased dramatically between 2001-2005 and 2011-2018 (15.6% to 40.0%; p=0.011).
Inappropriate prescription of antibiotics leads to treatment failure and the development of further resistance. In localities with high levels of clarithromycin resistance, alternatives such as levofloxacin should be considered, although use should be safeguarded to prevent the further development of resistance. Genetic determination of levofloxacin resistance in the gyrA gene has previously been investigated; however, genotypic sensitivity testing does not reliably predict phenotypic resistance. To investigate levofloxacin resistance mechanisms, the quinolone resistance determining region in the gyrA gene of 37 H. pylori isolates were amplified via PCR and Sanger sequenced. The minimum inhibitory concentration (MIC) and sequencing data were compared. Of the 9 phenotypically resistant isolates, 88% had a substitution at gyrA amino acid position 87 or 91. Every isolate containing a substation at position 91 or an Asn87Lys substitution had a resistant MIC of ≥32 to levofloxacin, evidencing that those mutations are predictors for resistance. One phenotypically resistant isolate did not contain any gyrA mutations.
To further investigate this, whole genome sequencing (WGS) was undertaken. However, H. pylori has a highly mutable genome which makes the correct assembly of short-read sequences problematic. Conversely, the higher error rate of long-read sequencing can obscure the presence of resistance polymorphisms. Therefore, a hybrid approach was tested. 13 isolates with a range of levofloxacin MICs were selected for whole genome sequencing using Illumina and Oxford Nanopore Technologies. The long-read sequencing data were used to construct de novo genome assemblies. Polishing using long- and short-read data increased the genome length and mean quality while decreasing sequencing gaps. Polished genome sequences were analysed for resistance genes using the Comprehensive Antibiotic Resistance Database (CARD). The agreement rate between genotypic and phenotypic resistance to both levofloxacin and tetracycline was 100%, metronidazole was 61%, while clarithromycin and amoxicillin had low accordance with 23% and 0% respectively.
In summary, resistance rates to clarithromycin and metronidazole were higher than had previously been estimated for UK isolates. Based on the resistance profiles, treatment failure is more likely to occur when patients are given first-line therapies without amoxicillin. Levofloxacin resistance can be predicted from sequencing of the gyrA gene, and polishing of whole genome sequences can be used to confidently identify levofloxacin resistance genes using the online tool CARD. Based on this data, it is recommended that resistance rates of H. pylori should be more extensively monitored, and alternative eradication therapies should be considered. Further advancement of PCR and WGS-based AMR detection methods should be explored.
Item Type
ethesis
Thesis Type
PhD
Supervisors
Subjects (LC)
Associated Schools / Departments
School of Medicine
eprints ID
80902
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
Thesis_EMGarvey_20004962_Final Submission.pdf
Type
Full-text
Description
Examined. Thesis with corrections completed
Size
4.49 MB
Format
Adobe PDF
Checksum (MD5)
6e14483ae002011e44698d64ee199929