Lung microbiome in children with hematological malignancies and lower respiratory tract infections

Yun Zhang, Haonan Ning, Wenyu Zheng, Jing Liu, Fuhai Li, Junfei ChenFrontiers in Oncology 2022 · DOI

TL;DR

Children with hematological malignancies and lower respiratory tract infections (LRTIs) exhibit a significantly altered lung microbiome, characterized by reduced diversity and an increased abundance of potentially pathogenic bacteria. These microbial shifts are linked to poorer clinical outcomes, suggesting potential diagnostic and therapeutic targets.

What Problem Does This Paper Address?

Children undergoing treatment for hematological malignancies are particularly vulnerable to severe respiratory infections, which remain a leading cause of morbidity and mortality. While the lung microbiome is known to play a crucial role in respiratory health, its specific composition and function in this vulnerable pediatric population, especially when complicated by LRTIs, remain poorly understood. This study aims to characterize the lung microbiome in children with hematological malignancies and LRTIs, comparing it to children with LRTIs but without underlying malignancies, to identify key differences and their clinical implications.

Methods

The researchers conducted a prospective cohort study, collecting bronchoalveolar lavage (BALF) fluid samples from two groups of children: 16 with hematological malignancies and moderate-to-severe LRTIs, and 21 age-, gender-, weight-, and infection-severity-matched controls with LRTIs but no underlying malignancies.

Key techniques employed include:

  • 16S rRNA gene sequencing: This standard method was used to analyze the bacterial composition of the BALF samples by sequencing the V3-V4 hypervariable region.
  • Quantitative PCR (qPCR): Used to determine the bacterial load in the samples.
  • Bioinformatic analysis: This involved several steps:
  • Quality control and filtering: Raw sequencing data was processed to remove low-quality reads and chimeric sequences.
  • Operational Taxonomic Unit (OTU) clustering: Sequences were clustered into OTUs based on a 97% identity cutoff to represent microbial diversity.
  • Species annotation: OTUs were annotated to identify the microbial taxa present.
  • Alpha diversity analysis: Metrics like ACE, Chao1, Simpson, and Shannon index were calculated to assess diversity within individual samples.
  • Beta diversity analysis: Principal Component Analysis (PCA) and UPGMA clustering were used to assess differences in microbial community composition between samples.
  • Functional prediction: PICRUSt2 and BugBase were utilized to predict the functional capabilities of the microbial communities based on their genetic potential.
  • Statistical analysis: Fisher's exact test, chi-square tests, Wilcoxon rank-sum test, general linear models, and Spearman correlation were used to compare groups and identify associations.

Datasets

The study utilized BALF samples collected from pediatric patients at Qilu Hospital of Shandong University between January 2020 and October 2021. No pre-existing public datasets were used for comparison or training in this specific study.

Key Insights and Ideas

  • Distinct Microbial Signatures: Children with hematological malignancies and LRTIs have a markedly different lung microbiome compared to those with LRTIs alone. This isn't just a minor shift; it's a fundamental alteration in community structure.
  • Reduced Diversity, Increased Pathogenicity: The most striking finding is the significantly decreased alpha and beta diversity in the malignancy group. This reduction in microbial richness and evenness is coupled with an increase in the abundance of potentially pathogenic bacteria like Proteobacteria, Escherichia_Shigella, and Klebsiella. This suggests a less resilient and more pro-inflammatory lung environment.
  • Functional Shifts: Beyond just who is there, what the microbes are doing is also different. The lung microbiome in the malignancy group shows enhanced functions related to infectious disease pathogenesis (bacterial and parasitic), antimicrobial drug resistance, and survival in harsh environments. This implies the microbiome is actively contributing to the severity and persistence of infections.
  • Predictive Power of Microbiome: The study identifies specific microbial features (alpha diversity, beta diversity, Proteobacteria, Firmicutes, Escherichia_Shigella, unclassified_Enterobacteriaceae) that are significantly correlated with hospitalization course. This opens the door for using microbiome profiles as biomarkers to predict patient outcomes.

Key Results

  • Diversity: Significantly decreased alpha (Shannon index) and beta (Bray-Curtis dissimilarity) diversity were observed in children with hematological malignancies and LRTIs compared to controls (p < 0.01).
  • Phylum-level Abundance: The malignancy group showed a reduced proportion of Firmicutes, Bacteroidota, and Actinobacteriota, and an increased proportion of Proteobacteria.
  • Genus-level Abundance: Notably elevated genera in the malignancy group included Parabacteroides, Klebsiella, Grimontia, Escherichia_Shigella, and unclassified_Enterobacteriaceae.
  • Functional Differences: Functions related to infectious disease (bacteria/parasite), antimicrobial drug resistance, and human pathogenesis were significantly increased in the malignancy group.
  • Clinical Correlations:
    • Alpha diversity (Shannon), beta diversity (Bray-Curtis), Proteobacteria, unclassified_Enterobacteriaceae, and Escherichia_Shigella were negatively associated with hospitalization course.
  • Firmicutes were positively associated with hospitalization course.

Limitations and Future Work

The authors acknowledge several limitations:

  • Sample Size: The study involved a relatively small number of patients (16 in the malignancy group), which limits the generalizability of the findings. Larger, multi-center studies are needed.
  • Single Center: Data was collected from a single institution, potentially introducing site-specific biases.
  • Lack of Gut Microbiome Data: The study focused solely on the lung microbiome, but the gut-lung axis is increasingly recognized. Future research should integrate gut microbiome data to explore potential cross-talk.
  • Confounding Factors: Children with hematological malignancies are undergoing chemotherapy and have compromised immune systems, making it challenging to isolate the specific impact of the malignancy itself versus these treatments on the microbiome.

Future work suggested includes:

  • Longitudinal studies tracking the microbiome over time, especially during chemotherapy.
  • Investigating the gut-lung axis and its role in these infections.
  • Exploring the specific mechanisms by which the altered lung microbiome contributes to disease pathogenesis and influences clinical outcomes.
  • Developing microbiome-based diagnostic and therapeutic strategies.

Summary updated

Abstract

Children with hematological malignancies and LRTIs showed obviously decreased α and β diversity, significantly increased function in infectious disease pathogenesis, antimicrobial drug resistance, and unfavorable environment tolerance. Moreover, α diversity (Shannon), β diversity (Bray-Curtis dissimilarity), and <i>Proteobacteria</i> may be used as negative correlated predictors for hospitalization course in these children whereas <i>Firmicutes</i> may be utilized as a positive correlated predictor.
BibTeX
@inproceedings{zhang2022lung,
  author = {Yun Zhang and Haonan Ning and Wenyu Zheng and Jing Liu and Fuhai Li and Junfei Chen},
  title = {{Lung microbiome in children with hematological malignancies and lower respiratory tract infections}},
  booktitle = {Frontiers in Oncology},
  year = {2022},
  doi = {10.3389/fonc.2022.932709},
}
Figure