Bacterial Signatures of Paediatric Respiratory Disease: An Individual Participant Data Meta-Analysis

Abstract

Introduction: The airway microbiota has been linked to specific paediatric respiratory diseases, but studies are often small. It remains unclear whether particular bacteria are associated with a given disease, or if a more general, non-specific microbiota association with disease exists, as suggested for the gut. We investigated overarching patterns of bacterial association with acute and chronic paediatric respiratory disease in an individual participant data (IPD) meta-analysis of 16S rRNA gene sequences from published respiratory microbiota studies.Methods: We obtained raw microbiota data from public repositories or via communication with corresponding authors. Cross-sectional analyses of the paediatric (<18 years) microbiota in acute and chronic respiratory conditions, with >10 case subjects were included. Sequence data were processed using a uniform bioinformatics pipeline, removing a potentially substantial source of variation. Microbiota differences across diagnoses were assessed using alpha- and beta-diversity approaches, machine learning, and biomarker analyses.Results: We ultimately included 20 studies containing individual data from 2624 children. Disease was associated with lower bacterial diversity in nasal and lower airway samples and higher relative abundances of specific nasal taxa including Streptococcus and Haemophilus. Machine learning success in assigning samples to diagnostic groupings varied with anatomical site, with positive predictive value and sensitivity ranging from 43 to 100 and 8 to 99%, respectively.Conclusion: IPD meta-analysis of the respiratory microbiota across multiple diseases allowed identification of a non-specific disease association which cannot be recognised by studying a single disease. Whilst imperfect, machine learning offers promise as a potential additional tool to aid clinical diagnosis.

Realising respiratory microbiomic meta-analyses: time for a standardised framework

“…Access to published data varies greatly among studies. For our meta-analysis [ 19 ], just over half of the 20 included studies had publicly available sequence data, and, at the time of the initial literature search, only two contained sufficient metadata to enable meaningful linking with the sequence data. There was also variability in terms of the types of samples that had been uploaded to public repositories: some publicly uploaded datasets included samples which were not explained or mentioned in the original article, potentially representing mock communities, sequencing controls, negative extraction controls, samples excluded from the original publication or samples from different studies that were uploaded under the same accession number.…”

“…Our meta-analysis [ 19 ] identified two broad approaches to reporting antibiotic usage that are common among respiratory microbiota studies: either the time between a patient’s last antibiotic use and sampling is reported or, alternatively, no samples are taken from individuals with antibiotic exposure more recent than a specified cut-off date. The former can be challenging to accommodate in meta-analyses, as a standardised definition of “recent antibiotic use” is lacking.…”

“…Having recently conducted a large-scale, 16S rRNA gene-based IPD meta-analysis focused on the paediatric respiratory microbiota that involved 2624 children from 20 studies [ 19 ], we identified several challenges around both data sharing and methodological heterogeneity within this research niche. Here, we examine limitations of respiratory microbiota datasets that emerged from our study [ 12 ] and propose strategies to achieve the higher inter-study standardisation needed to support future large-scale respiratory microbiome meta-analyses.…”

Microbiomes of native Aotearoa New Zealand animals

“…Each of the aforementioned approaches brings its own biases, including (but not limited to) differences in DNA extraction efficiency between different microbial cell types, PCR bias whereby the rRNA genes of some organisms are preferentially amplified over others (sometimes due to selection of PCR primers), and choice of database for assigning taxonomic identities to obtained sequence data (Pollock et al 2018). Differences in methodological approaches among different research labs can make it difficult to compare between studies, and there are increasing calls for standardisation of methods within the microbiome field (Costea et al 2017;Broderick et al 2021;Bodawatta et al 2022). While global initiatives such as the Earth Microbiome Project (Thompson et al 2017) speak to the power of methods standardisation, one must also acknowledge the challenges with attempting to develop a one-size-fits-all approach; indeed, in the authors' laboratory alone, several different DNA extraction protocols are used when working with different host animals.…”

MADAME: a user-friendly bioinformatic tool for data and metadata retrieval in microbiome research

“…Meta-analyses minimize this variation, providing a more precise estimation of microbiome effects. In this way, robust, generalizable microbiome-health relationships become identifiable (Broderick et al 2023). Moreover, meta-analyses unveil elusive trends and associations in combined data, facilitating links recognition between microbial groups or compositions and health conditions.…”