Airway "Resistotypes" and Clinical Outcomes in Bronchiectasis

Abstract

Rationale: Chronic infection and inflammation shapes the airway microbiome in bronchiectasis. Utilizing whole-genome shotgun metagenomics to analyze the airway resistome provides insight into interplay between microbes, resistance genes, and clinical outcomes. Objectives: To apply whole-genome shotgun metagenomics to the airway microbiome in bronchiectasis to highlight a diverse pool of antimicrobial resistance genes: the "resistome," the clinical significance of which remains unclear. Methods: Individuals with bronchiectasis were prospectively recruited into cross-sectional and longitudinal cohorts (n = 280), including the international multicenter cross-sectional Cohort of Asian and Matched European Bronchiectasis 2 (CAMEB 2) study (n = 251) and two independent cohorts, one describing patients experiencing acute exacerbation and a further cohort of patients undergoing Pseudomonas aeruginosa eradication treatment. Sputum was subjected to metagenomic sequencing, and the bronchiectasis resistome was evaluated in association with clinical outcomes and underlying host microbiomes. Measurements and Main Results: The bronchiectasis resistome features a unique resistance gene profile and increased counts of aminoglycoside, bicyclomycin, phenicol, triclosan, and multidrug resistance genes. Longitudinally, it exhibits within-patient stability over time and during exacerbations despite between-patient heterogeneity. Proportional differences in baseline resistome profiles, including increased macrolide and multidrug resistance genes, associate with shorter intervals to the next exacerbation, whereas distinct resistome archetypes associate with frequent exacerbations, poorer lung function, geographic origin, and the host microbiome. Unsupervised analysis of resistome profiles identified two clinically relevant "resistotypes," RT1 and RT2, the latter characterized by poor clinical outcomes, increased multidrug resistance, and P. aeruginosa. Successful targeted eradication in P. aeruginosa-colonized individuals mediated reversion from RT2 to RT1, a more clinically favorable resistome profile demonstrating reduced resistance gene diversity. Conclusions: The bronchiectasis resistome associates with clinical outcomes, geographic origin, and the underlying host microbiome. Bronchiectasis resistotypes link to clinical disease and are modifiable through targeted antimicrobial therapy.

Keywords: bronchiectasis; metagenomics; microbiome; resistome; resistotype.

Figure 1.

Characterization of the bronchiectasis resistome. (A) Stacked bar plots representing the observed resistome profiles in bronchiectasis (n = 251; Cohort of Asian and Matched European Bronchiectasis 2) compared with a nondiseased comparator cohort (n = 25). Stacked bars represent the relative abundance of resistance genes grouped according to drug class. (B) β-Diversity of resistance gene profiles (i.e., Bray-Curtis distance) assessed by principal coordinate analysis (PCoA) illustrating resistome diversity and colored by cohort: nondiseased in blue and bronchiectasis in red (permutational multivariate ANOVA, P < 0.001). (C) Longitudinal resistome dynamics in stable bronchiectasis assessed at baseline and follow-up time points across varying time intervals (left; 3 mo, center; 6 mo, right; 7 mo). (D and E) Prospective longitudinal analysis of resistome profiles during exacerbations of bronchiectasis (n = 18, cohort 3, “exacerbation” cohort) with sampling at baseline (before exacerbation, during exacerbation, and 2 wk after exacerbation following antibiotic treatment). (F and G) Prospective longitudinal analysis of resistome profiles in the bronchiectasis exacerbation cohort grouped by observed time to the next exacerbation as <12 weeks or >12 weeks, respectively. PCoA of resistome (β-) diversity (i.e., Bray-Curtis distance) comparing (E) time points across an exacerbation and (G) time to the next exacerbation are illustrated below their respective stacked bar plots (D and F). P values for permutational multivariate ANOVA comparing groups are illustrated. Central points intersected by a cross indicate the centroid in PCoA visualizations. B = before exacerbation; BL = baseline; E = during exacerbation; FUP = follow-up; P = 2 weeks after exacerbation following antibiotic treatment; PC = principal coordinate; PERMANOVA = permutational multivariate ANOVA.

Figure 2.

Association between resistome composition and clinical features in bronchiectasis. Stacked bar plots representing resistome profiles in stable bronchiectasis in the Cohort of Asian and Matched European Bronchiectasis 2 (matched and unmatched, n = 251) stratified according to: (A) exacerbation status, i.e., number of exacerbations in the preceding year categorized as nonexacerbators (i.e., no exacerbations), exacerbators (one or two exacerbations), or frequent exacerbators (at least three exacerbations); (B) lung function based on FEV₁% predicted as >70%, 50–70%, 30–50%, and <30% predicted; (C) disease severity based on the Bronchiectasis Severity Index (BSI) as mild (BSI = 0–4), moderate (BSI = 5–9), or severe (BSI ⩾ 9); and (D) bronchiectasis etiology as idiopathic, postinfection, posttuberculosis, or other. Assessment of intergroup differences in relative abundance of antimicrobial resistance gene classes was evaluated using the Kruskal-Wallis test with false discovery rate correction: *P < 0.05 and **P < 0.01. Ex = exacerbators; F.Ex = frequent exacerbators; IP = idiopathic; N.Ex = nonexacerbators; PI = postinfection; PTB = posttuberculosis.

Figure 3.

The Cohort of Asian and Matched European Bronchiectasis 2 (CAMEB 2) cohort and geographic variability in the bronchiectasis resistome. (A) Schematic overview of the matching strategy employed for individuals included in CAMEB 2. Asian patients (Singapore, n = 95; and Kuala Lumpur, Malaysia, n = 35) were compared with European patients (Dundee, Scotland, n = 96; and Milan, Italy, n = 25) to identify a cohort of 209 patients matched for age, sex, exacerbation frequency, and lung function differing in geographic origin. (B) β-Diversity of resistance gene profiles (i.e., Bray-Curtis distance) assessed by principal coordinate analysis (PCoA) illustrating resistome diversity and colored by geographic origin assessing matched patients from CAMEB 2: Singapore (red), Kuala Lumpur (yellow), Dundee (blue), and Milan (green) (permutational multivariate ANOVA, P = 0.006). (C) β-Diversity of resistance gene profiles (i.e., Bray-Curtis distance) assessed by PCoA illustrating resistome diversity and colored by etiology assessing matched patients from CAMEB 2: idiopathic (purple), postinfection (orange), posttuberculosis (blue), and others (gray). Central points intersected by a cross indicate the centroid in PCoA visualizations. IP = idiopathic; PC = principal coordinate; PERMANOVA = permutational multivariate ANOVA; PI = postinfection; PTB = posttuberculosis.

Figure 4.

Characterization of bronchiectasis resistotypes. (A) Heat map illustrating the Cohort of Asian and Matched European Bronchiectasis 2 resistome profiles that associate with the identified resistance gene profile clusters (“resistotypes” [RTs]) indicated as RT1 (blue) and RT2 (purple). The heat map details the resistance gene composition of each resistotype cluster at the drug class level, expressed as log₂ percentage relative abundance. (B) Principal coordinate analysis of gene-level resistome profiles demonstrating the distinct resistotype RT1 and RT2 clusters as defined by β-diversity of resistance gene profiles (i.e., Bray-Curtis distance). Circles intersected by a cross indicate principal coordinate analysis centroids of compared groups (permutational multivariate ANOVA, P < 0.001). (C) Linear discriminant analysis highlighting resistance genes that distinguish the RT1 and RT2 resistotype clusters. (D–F) Analysis of clinical features associated with the resistotype clusters RT1 (blue) and RT2 (purple) including (D) exacerbation frequency in the year preceding recruitment, (E) lung function (as FEV₁% predicted), and (F) disease severity (per Bronchiectasis Severity Index). (G) Prevalence of the resistotypes across the recruited cohorts and respective geographic regions. **P < 0.01. BSI = Bronchiectasis Severity Index; DD = Dundee, Scotland; KL = Kuala Lumpur, Malaysia; LDA = linear discriminant analysis; MI = Milan, Italy; PERMANOVA = permutational multivariate ANOVA; SG = Singapore.

Figure 5.

Microbiome correlates of bronchiectasis resistotypes (RTs) RT1 and RT2. Stacked bar plots illustrating the relative abundance of microbial taxa associated with (A) resistotypes RT1 and RT2 and stratified by (B) continent and (C) country of origin. (D) Analysis of resistance gene content among the top microbial taxa identified in the Cohort of Asian and Matched European Bronchiectasis 2. Horizontal bars indicate the number of antimicrobial resistance genes (x-axis) identified on contigs assigned to the indicated bacterial taxa (y-axis) based on metagenomic assembly. Prevalent bacterial taxa ranked within the top 40 (by average relative abundance) are colored, whereas lower-abundance taxa are grayed. (E) Analysis of resistance gene distribution across bacterial taxa. Horizontal bars indicate the number of bacterial taxa (x-axis) expressing the respective indicated resistance genes (y-axis) based on metagenomic contig analysis. Resistance genes are differentiated by color according to drug class as indicated in the labels on the y-axis (in parentheses). AMR = antimicrobial resistance; DD = Dundee, Scotland; KL = Kuala Lumpur, Malaysia; MI = Milan, Italy; SG = Singapore.

Figure 6.

The effect of Pseudomonas aeruginosa eradication therapy on bronchiectasis resistotypes. (A) Microbiome composition of eight patients with P. aeruginosa culture-positive bronchiectasis (P1–P8) receiving eradication therapy with 14 days of oral ciprofloxacin followed by 3 months of nebulized colistin. Stacked bar plots illustrate the lung microbiome composition before and after eradication therapy. (B) Resistome profiles of patients with P. aeruginosa culture-positive bronchiectasis (P1–P8) receiving eradication therapy with 14 days of oral ciprofloxacin followed by 3 months of nebulized colistin. Stacked bar plots illustrate resistome composition (at drug class level) before and after eradication therapy. (C) Principal coordinate analysis plot illustrating resistome gene profiles incorporating patients undergoing P. aeruginosa eradication therapy (before eradication, green; after eradication, red) oriented to their previously determined resistotype clusters: RT1 (blue) and RT2 (purple). A black arrow indicates the trajectory of the eradication therapy group centroids, illustrating a shifting proximity from the initial RT2 (purple) to RT1 (blue) cluster in the pre- and posteradication state, respectively. (D) Stacked bar plots depicting antimicrobial resistome profiles of patients before and after antimicrobial treatment. The left section illustrates changes in these profiles following the administration of acute antibiotic therapy to treat pulmonary exacerbations; the right illustrates changes observed after P. aeruginosa eradication therapy. *P < 0.05. ns = not significant (P > 0.05 on paired permutational multivariate ANOVA); PERMANOVA = permutational multivariate ANOVA.