Higher relative abundance of several lung microbiome genera not commonly recognized as pathogens was associated with lower airway inflammatory markers across chronic airway diseases, according to a systematic review and meta-analysis published in npj Biofilms and Microbiomes.
The analysis also identified modest associations between some genera and improved lung function, although the findings differed by disease state and should be interpreted cautiously.
Researchers reviewed 34 observational studies, including both cross-sectional and longitudinal designs, published from 2008 to 2025. The studies included 4,119 patients with asthma, bronchiectasis, cystic fibrosis, chronic obstructive pulmonary disease (COPD), or asthma-COPD overlap syndrome. Most studies evaluated sputum samples using 16S ribosomal RNA sequencing to characterize airway microbial composition.
Across pooled analyses, Prevotella and Veillonella abundance was positively associated with forced expiratory volume in 1 second (FEV1), although the reported effect sizes were modest and of uncertain clinical significance. Several genera were also associated with lower markers of neutrophilic airway inflammation.
Specifically, Prevotella, Rothia, Streptococcus, and Veillonella were associated with lower neutrophil elastase activity, while Neisseria, Prevotella, and Rothia were linked to lower interleukin-1 beta and interleukin-8 concentrations. Lower tumor necrosis factor alpha concentrations were associated with Actinomyces, Granulicatella, Lautropia, Prevotella, and Veillonella.
The researchers emphasized that the findings were hypothesis-generating rather than practice-changing. Twenty-six of the 34 included studies were cross-sectional, limiting causal inference and preventing determination of whether bacterial abundance contributed to improved airway health or simply reflected healthier airway environments.
Most studies also did not adequately adjust for important confounders such as smoking, corticosteroid exposure, antibiotic use, age, and sex.
The investigators pooled data across asthma, COPD, cystic fibrosis, and bronchiectasis largely because disease-specific sample sizes were insufficient for most genus-biomarker combinations. Disease-specific meta-analysis was feasible for only three combinations, and only one — Prevotella abundance and lung function in COPD — remained statistically significant in disease-specific analysis.
Importantly, some asthma studies showed the opposite relationship, with Prevotella associated with lower lung function, suggesting that microbiome associations may differ substantially across chronic airway diseases.
The paper also highlighted important methodological limitations in current airway microbiome research. Most included studies used 16S ribosomal RNA sequencing, which generally identifies bacteria only at the genus level rather than the species or strain level. The authors noted that genera such as Streptococcus contain both potentially beneficial commensals and recognized pathogens, limiting interpretation of genus-level associations alone.
Another challenge is that several genera associated with favorable biomarkers — including Prevotella, Veillonella, Rothia, Neisseria, and Streptococcus — are common oral commensals. However, the researchers noted that studies using bronchoalveolar lavage and protected bronchial brushings, which reduce upper-airway contamination, demonstrated similar associations.
The authors also cautioned that most studies reported relative bacterial abundance rather than absolute bacterial load. Without quantitative approaches such as polymerase chain reaction or droplet digital polymerase chain reaction, it remains unclear whether these genera truly expand in biomass or simply appear proportionally increased when pathogenic organisms decline.
“Thus, there is currently insufficient evidence to classify nonpathogenic bacteria of the lung microbiome as pathobionts,” wrote Lucia Grassi, of Ghent University, and colleagues. The researchers said species- and strain-level longitudinal studies and mechanistic investigations are needed to clarify whether these bacteria actively influence disease biology and whether they could eventually serve as therapeutic targets.
The review was not preregistered, and no study protocol was prepared prior to the analysis. The manuscript is currently published as an article in press pending final editing.
Disclosures: Aurélie Crabbé, of Ghent University, reported a pending patent application related to anti-inflammatory bacterial compositions and therapeutic applications.
Source: npj Biofilms and Microbiomes