The airway microbiome of persons with cystic fibrosis correlates with acquisition and microbiological outcomes of incident Stenotrophomonas maltophilia infection.

Rationale: Chronic infection with Stenotrophomonas maltophilia in persons with cystic fibrosis (pwCF) has been linked to an increased risk of pulmonary exacerbations and lung function decline. We sought to establish whether baseline sputum microbiome associates with risk of S. maltophilia incident infection and persistence in pwCF. Methods: pwCF experiencing incident S. maltophilia infections attending the Calgary Adult CF Clinic from 2010-2018 were compared with S. maltophilia-negative sex, age (+/-2 years), and birth-cohort-matched controls. Infection outcomes were classified as persistent (when the pathogen was recovered in ≥50% of cultures in the subsequent year) or transient. We assessed microbial communities from prospectively biobanked sputum using V3-V4 16S ribosomal RNA (rRNA) gene sequencing, in the year preceding (Pre) (n = 57), at (At) (n = 22), and after (Post) (n = 31) incident infection. We verified relative abundance data using S. maltophilia-specific qPCR and 16S rRNA-targeted qPCR to assess bioburden. Strains were typed using pulse-field gel electrophoresis. Results: Twenty-five pwCF with incident S. maltophilia (56% female, median 29 years, median FEV1 61%) with 33 total episodes were compared with 56 uninfected pwCF controls. Demographics and clinical characteristics were similar between cohorts. Among those with incident S. maltophilia infection, sputum communities did not cluster based on infection timeline (Pre, At, Post). Communities differed between the infection cohort and controls (n = 56) based on Shannon Diversity Index (SDI, p = 0.04) and clustered based on Aitchison distance (PERMANOVA, p = 0.01) prior to infection. At the time of incident S. maltophilia isolation, communities did not differ in SDI but clustered based on Aitchison distance (PERMANOVA, p = 0.03) in those that ultimately developed persistent infection versus those that were transient. S. maltophilia abundance within sputum was increased in samples from patients (Pre) relative to controls, measuring both relative (p = 0.004) and absolute (p = 0.001). Furthermore, S. maltophilia abundance was increased in sputum at incident infection in those who ultimately developed persistent infection relative to those with transient infection, measured relatively (p = 0.04) or absolute (p = 0.04), respectively. Conclusion: Microbial community composition of CF sputum associates with S. maltophilia infection acquisition as well as infection outcome. Our study suggests sputum microbiome may serve as a surrogate for identifying infection risk and persistence risk.

Staphylococcus aureus in Non-Cystic Fibrosis Bronchiectasis: Prevalence and Genomic Basis of High Inoculum ß-Lactam Resistance.

Rationale: The pathobiology of Staphylococcus aureus in non-cystic fibrosis bronchiectasis (nCFB) is poorly defined. When present at high density or "inoculum," some methicillin-sensitive S. aureus (MSSA) can inefficiently degrade antistaphylococcal ß-lactam antibiotics via BlaZ penicillinases (termed the "inoculum effect" [IE]). Given the high burden of organisms in bronchiectatic airways, this is particularly relevant. Objectives: Drawing from a prospectively collected biobank, we sought to understand the prevalence, natural history, potential for transmission, and antibiotic resistance profiles among nCFB-derived MSSA isolates. Methods: All individuals attending a regional consultancy nCFB clinic with sputum collected between 1981 and 2017 were considered, and those with one or more S. aureus-positive cultures composed the cohort. Each individual's most recent biobank isolate was subjected to whole-genome sequencing (including the blaZ gene), antibacterial susceptibility testing, and comparative ß-lactam testing at standard (5 × 105 colony-forming unit [cfu]/ml) and high (5 × 107 cfu/ml) inocula to assess for the IE and pronounced IE. Results: Seventy-four (35.4%) of 209 individuals had one or more sputum samples with S. aureus (68 MSSA, 6 methicillin-resistant S. aureus). Those with S. aureus infection were more likely to be female. Among 60 of 74 MSSA isolates subjected to whole-genome sequencing, no evidence of transmission was identified, although specific multilocus sequence typing types were prevalent, including ST-1, ST-15, ST-30, and ST-45. Antibiotic resistance was uncommon, except for macrolides (∼20%). Among the 60 MSSA samples, the prevalence of IE and pronounced IE was observed to be drug specific: meropenem (0% and 0%, respectively), cefepime (3% and 5%, respectively), ceftazidime (8% and 0%, respectively), cloxacillin (12% and 0%, respectively), cefazolin (23% and 0%, respectively), and piperacillin-tazobactam (37% and 17%, respectively). The cefazolin IE was associated with blaZ type A (P < 0.01) and ST-30 (P < 0.01), whereas the piperacillin-tazobactam IE was associated with type C blaZ (P < 0.001) and ST-15 (P < 0.05). Conclusions:S. aureus infection was common, although no evidence of transmission was apparent in our nCFB cohort. Although routine susceptibility testing did not identify significant resistance, inoculum-related resistance was found to be relevant for commonly used nCFB antibiotics, including cefazolin and piperacillin-tazobactam. Given previous associations between IEs and negative patient outcomes, further work is warranted to understand how this phenotype impacts nCFB disease progression.