Loci for prediction of penicillin and tetracycline susceptibility in Neisseria gonorrhoeae: a genome-wide association study.

Background: Neisseria gonorrhoeae poses an urgent public health threat because of increasing antimicrobial resistance; however, much of the circulating population remains susceptible to historical treatment regimens. Point-of-care diagnostics that report susceptibility could allow for reintroduction of these regimens, but development of such diagnostics has been restricted to ciprofloxacin, for which susceptibility can be predicted from a single locus. We aimed to define genetic variants associated with susceptibility to penicillin and tetracycline.
Methods: We collected publicly available global whole-genome sequencing data (n=12 045) from clinical N gonorrhoeae isolates, with phenotypic resistance data for penicillin (n=6935), and tetracycline (n=5727). Using conditional genome-wide association studies, we defined genetic variants associated with susceptibility to penicillin and tetracycline. We excluded isolates that could not be classified as either susceptible or resistant. To validate our results, we assembled 1479 genomes from the US Centers for Disease Control and Prevention (CDC)'s Gonococcal Isolate Surveillance Project, for which urethral specimens are collected at sentinel surveillance sites across the USA. We evaluated the sensitivity and specificity of susceptibility-associated alleles using Clinical & Laboratory Standards Institute breakpoints for susceptibility and non-resistance in both the global and validation datasets.
Findings: In our conditional penicillin genome-wide association study, the presence of a genetic variant defined by a non-mosaic penA allele without an insertion at codon 345 was associated with penicillin susceptibility and had the highest negative effect size (β) of significant variants (p=5·0x10 ^-14 , β -2·5). In combination with the absence of bla _TEM , this variant predicted penicillin susceptibility with high specificity (99·8%) and modest sensitivity (36·7%). For tetracycline, the wildtype allele at rpsJ codon 57, encoding valine, was associated with tetracycline susceptibility (p=5·6x10 ^-16 , β -1·6) after conditioning on the presence of tetM. The combination of rpsJ codon 57 allele and tetM absence predicted tetracycline susceptibility with high specificity (97·2%) and sensitivity (88·7%).
Interpretation: As few as two genetic loci can predict susceptibility to penicillin and tetracycline in N gonorrhoeae with high specificity. Molecular point-of-care diagnostics targeting these loci have the potential to increase available treatments for gonorrhoea.
Funding: National Institute of Allergy and Infectious Diseases, the National Science Foundation, and the Smith Family Foundation.
Competing Interests: Declaration of interests YHG is on the scientific advisory board of Day Zero Diagnostics; has consulted for Quidel and GSK; has received grant funding from Merck, Pfizer, and GSK; and has received payments for participating in National Institutes of Health (NIH) study sections and speaking at the Association for Molecular Pathology conference. All other authors declare no competing interests.
(Copyright © 2022 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license. Published by Elsevier Ltd.. All rights reserved.)