Antimicrobial susceptibility assays for Neisseria gonorrhoeae: a proof-of-principle population-based retrospective analysis

Background Neisseria gonorrhoeae treatment guided by molecular antimicrobial susceptibility assays could improve treatment options and antimicrobial stewardship; however, few commercial assays are available. We aimed to investigate antimicrobial susceptibility of N gonorrhoeae isolates in New South Wales, Australia, and estimate the potential usefulness of hypothetical combinations of rapid molecular antimicrobial susceptibility assays.Methods In this proof-of-principle, population-based, retrospective analysis, we assessed N gonorrhoeae susceptibility data for ceftriaxone, azithromycin, ciprofloxacin, and penicillin. Isolates were previously collected as part of the Australian Gonococcal Surveillance Programme between Jan 1, 2008, and Dec 31, 2019. All cultured N gonorrhoeae isolates with susceptibility data to all four antimicrobials were included. However, only one isolate was included if several isolates originated from the same individual within 13 days of the previous isolate originating from that individual, and there were less than two standard double-dilution minimum inhibitory concentrations between the isolates. We assessed the use of different combinations of hypothetical antimicrobial susceptibility assays and treatment combinations in terms of their ability to minimise overall ceftriaxone use, and use specifically in isolates with decreased susceptibility to ceftriaxone, compared with standard non-assay-guided empirical ceftriaxone treatment.Findings We included 23 089 N gonorrhoeae isolates. The prevalence of antimicrobial sensitivity fluctuated significantly during the study. Isolates with decreased susceptibility to ceftriaxone were more likely to be resistant to one or more antimicrobials than isolates without decreased susceptibility (782 [986%] of 793 vs 10 661 [478%] of 22 296), particularly ciprofloxacin (p<00001) and penicillin (p<00001). Compared with empirical ceftriaxone treatment, we estimated that strategies based on the use of hypothetical antimicrobial susceptibility would reduce ceftriaxone use (p<00001). However, because of co-resistance, most assay-directed treatment strategies, including those involving use of assays for two antibiotics, would result in only moderate reductions in ceftriaxone use among isolates with decreased susceptibility to ceftriaxone.Interpretation Individualised treatment guided by molecular antimicrobial susceptibility diagnostics could help to reduce overall ceftriaxone use in gonorrhoea. However, the use of these assays needs to be informed by the non-random nature of co-resistance among circulating N gonorrhoeae strains.