2165. Helicobacter pylori Infections in the Bronx, New York: Whole-Genome Sequencing for Rapid Genotypic Susceptibility Testing

BACKGROUND: Susceptibility-guided treatment of H. pylori is superior to empiric therapy. We determined the accuracy of whole-genome sequencing (WGS) compared with phenotypic testing using CLSI/EUCAST breakpoints. METHODS: Thirty-three clinical isolates of H. pylori cultured from gastric biopsies were sequenced with a coverage range between 40x and 80x using Illumina Miseq platform and the reads were assembled and annotated with PATRIC. Phenotypic susceptibility tests were performed using E-test strips under microaerophilic conditions for 72 hours. Mutations associated with amoxicillin, tetracycline, clarithromycin, levofloxacin, metronidazole and rifampin resistance were examined. RESULTS: Of the 33 isolates, two were phenotypically resistant to amoxicillin: one carried a β-lactamase gene (bla(TEM-116)) and the other exhibited a point mutation pbp2 (A541T). All isolates were tetracycline susceptible phenotypically, but three isolates had point mutations in 16S rRNA that are associated with resistance (A926G). Clarithromycin results showed a good correlation between methods. Nine clarithromycin-resistant isolates demonstrated point mutations in 23S rRNA (A2142G/A2143G). Fifteen isolates were phenotypically resistant to levofloxacin, but resistance mutations were found in only 14 strains (N87I/N87K/D91Y/D91N/D91G/D99N in gyrA). We analyzed our strains for the presence of intact genes rdxA and frxA, either of which convert the prodrug form of metronidazole into the active form. Twenty-four of 33 isolates were tested phenotypically. We found 3 isolates with truncations in both genes. These isolates had metronidazole MICs >256. The presence of one or both intact genes did not always result in low MICs, indicating that there may be significant point mutations that contribute to resistance. Rifampin was not tested phenotypically, but no mutations in rpoB were found. In summary, the correlation of WGS and phenotypic testing was 100% for amoxicillin and clarithromycin, 97% for levofloxacin, 91% for tetracycline (n = 33), and 67% for metronidazole (n = 24). CONCLUSION: WGS provides a detailed analysis of H. pylori resistance and a broader analysis of antimicrobials that may be of clinical value. Additional studies are needed for genotypic prediction of metronidazole resistance. DISCLOSURES: All authors: No reported disclosures.