Patients with weak immune systems rely on antibiotics to survive, but this paper reveals a dangerous cycle where their own lungs become reservoirs for superbugs. Antibody-deficient patients face a high risk of respiratory infections and often receive prophylactic antibiotics or self-administered drugs for breakthrough cases.

The primary goal of this research is to understand the specific landscape of antimicrobial resistance within the respiratory tracts of these vulnerable individuals. Researchers recruited two groups: a cross-sectional cohort providing single samples during clinic visits and a prospective cohort submitting samples every two weeks for three months.

In the initial group, twenty-nine patients provided seventy samples, with Common Variable Immunodeficiency being the most frequent diagnosis among those with known bronchiectasis. Figure 1a displays the frequency of three hundred and forty-three bacterial species identified from seventy sputum samples using MALDI-TOF mass spectrometry.

A dotted line marks a two percent threshold, indicating which organisms were selected for further antibiotic resistance analysis. Meanwhile, Figure 1b uses a Venn diagram to show that the microbial community remains largely stable over time in patients who provided multiple samples, particularly regarding commensal streptococci.

Testing revealed extremely high resistance rates, with over eighty-one percent of isolates resistant to azithromycin and thirty-five percent resistant to tetracyclines. Figure 2 presents phenotypic resistance data for 255 isolates of Streptococcus and H.

influenzae, categorized by the patients' antibiotic prophylaxis history. Panel (a) highlights that azithromycin resistance is particularly prevalent across all groups, including those not receiving macrolide prophylaxis.

Panel (b) further illustrates overall susceptibility, showing that while cefotaxime maintains high sensitivity rates, azithromycin exhibits a significant proportion of resistant isolates. These findings underscore the widespread nature of antibiotic resistance in this patient population regardless of specific treatment regimens.

Molecular testing found macrolide resistance genes erm(B) and mef(A) in over forty percent of samples, regardless of whether patients were currently taking prophylaxis. Figure 3 breaks down the prevalence of specific macrolide resistance genes, erm(B) and mef(A), across sputum samples from patients on different antibiotic prophylaxis regimens.

Panels (a) and (b) show that these resistance markers appear in over forty percent of all tested samples, regardless of whether the patient was receiving macrolide treatment or no prophylaxis at all. Panel (c) reinforces this finding at the individual level, indicating that the majority of patients carried at least one of these resistance genes in their respiratory tract.

Phylogenetic analysis showed pairs of Haemophilus influenzae and Pseudomonas aeruginosa isolates from different patients clustering closely, suggesting possible transmission between them. Figure 4 presents a detailed analysis of Gram-negative isolates, specifically *H.

influenzae* and *P. aeruginosa*, by combining phylogenetic clustering with resistome data and phenotypic antibiograms. The authors map specific antibiotic resistance genes, such as variants of PBP3 in *H.

influenzae* or beta-lactamases like OXA and PDC in *P. aeruginosa*, against the patients' prophylactic therapy status. By aligning these genetic markers with disc diffusion test results for drugs like ampicillin and meropenem, the figure visually connects the presence of specific resistance determinants to observed clinical susceptibility patterns across different patient samples.

Analysis of plasmid sequences revealed conjugative plasmids carrying resistance genes clustering across different patients, which may suggest cross-transmission of these genetic elements. This phylogenetic tree illustrates how specific plasmids, such as repUS43 and repUS38, cluster together with streptococcal isolates that carry resistance genes for macrolides and tetracyclines.

The color-coded clades reveal a concerning pattern where identical plasmid sequences appear in different patients (pink), serial samples from the same patient (yellow), or even within the same sputum sample (blue). These groupings suggest that conjugative plasmids are actively moving between bacteria and potentially spreading across individuals through cross-transmission.

A significant negative relationship was found between neutrophil elastase concentration and alpha-diversity markers, while a positive relationship existed with microbial dominance indices. Table 2 reports Spearman correlation coefficients between elastase concentration and four alpha-diversity metrics, alongside correlations for the change in these values between consecutive samples.

The authors observe significant negative correlations with diversity measures like Shannon entropy and Simpson e, while noting a positive relationship with the Berger–Parker dominance index. These statistical associations suggest that higher elastase levels are linked to reduced microbial diversity and increased dominance by specific taxa within the community.

Elastase levels rose significantly with the abundance of the pathogen Haemophilus influenzae but dropped when commensal Streptococcus species were more abundant. Reviewing eleven antibiotic courses taken by nine patients, researchers classified five as appropriate but four as inappropriate due to no change in microbiome or elastase.

The study demonstrates extremely high rates of antibiotic resistance driven by therapeutic use, with macrolide resistance remaining prevalent even in patients not currently on those drugs. The authors conclude that sputum neutrophil elastase is a useful indicator of dysbiosis and acute infection, potentially guiding better antibiotic treatment decisions.

The study proves that high rates of antibiotic resistance exist even without current drug use, suggesting sputum elastase could be a vital marker to stop unnecessary prescriptions.