(518) A targeted sequencing panel simultaneously detects Pseudomonas aeruginosa and Staphylococcus aureus species and antimicrobial resistance profiles from sputum

Background: People with CF (pwCF) produce less sputum in the era of highly effective modulators. Thus, it is increasingly important to identify more sensitive approaches to monitor and manage chronic respiratory infections. Targeted culture-independent technologies designed to identify principal CF pathogens may help address needs in the current modulator era. We developed a targeted amplicon sequencing panel able to detect Pseudomonas aeruginosa (PA), Staphylococcus aureus (SA), mixed strain infections, and presence or absence of antimicrobial resistance (AMR) genes from a minimal sample of sputum DNA. We hypothesized that targeted amplicon sequencing would have several advantages over traditional culture in terms of improved sensitivity, ability to detect emerging infections, and prediction of microbiologic and clinical outcomes.

Methods: We designed a custom amplicon sequencing panel with amplicon targets for over 200 genes including species detection markers and AMR genes for PA, SA, and 16 other bacterial species. Amplicon libraries were sequenced on the Ion S5XL. As part of panel optimization, we performed extensive testing on DNA from pure bacterial strains to confirm primer specificity between species and in mixtures of strains. We also tested the panel on total DNA from sputum samples using a range of starting DNA inputs from 5-100ng. Finally, we evaluated the panel on a pilot set of 13 sputum samples from pwCF, and compared amplicon target detection (i.e. read counts) with traditional culture results and antimicrobial susceptibility testing (AST) results.

Results: Target read counts obtained from pure bacterial samples showed that the custom panel accurately identified PA, SA, and mixtures of species and strains. SA strain types (i.e. MSSA vs. MRSA) were distinguished by the presence or absence of AMR genes such as mecA, mecI and mecR1, and PA strains were distinguished by gene deletions such as exsA, CrpP, soxR, and the entire psl operon. Analysis of read counts from sputum samples with various DNA inputs (5-100ng) revealed highly similar AMR gene profiles regardless of starting concentration. Results from the pilot set of 13 CF sputum samples showed nearly perfect correspondence between detection of PA and/or SA read counts and traditional culture results. However, in a subset of samples, low levels of reads were detected for a species (PA or SA) in which the subject had cultured negative suggestive of an emerging infection. In SA-positive sputum, all samples with detection of the mecR1 gene had corresponding isolates that were resistant to oxacillin by AST while sputum samples with no mecR1 detection were susceptible (p=0.03). The panel also accurately separated MRSA from MSSA isolates cultured from sputum based upon AMR gene detection of mecR1 and ANT(4')-Ib from sputum DNA (p< 0.001).

Conclusions: Preliminary data suggest that our custom amplicon sequencing panel is highly sensitive and specific for simultaneous detection of PA, SA, and AMR gene profiles, even from extremely low quantities of total sputum DNA. Global analysis of gene targets allows for population-level detection of mixed species, multiple strains, and emerging infections in clinical samples. Examination of a larger cohort of CF sputum samples in the context of AST and pulmonary exacerbation treatment outcomes is ongoing and will be summarized for NACFC 2022.

Acknowledgements: This work was supported by the Cystic Fibrosis Foundation (SAAVED19A0)

References: