Multi-omics profiling predicts allograft function after lung transplantation.

Area of Science:

• Immunology
• Microbiology
• Genomics

Background:

• Lung transplantation offers a last resort for end-stage respiratory disease but has high mortality due to chronic lung allograft dysfunction (CLAD).
• Mechanisms of CLAD are poorly understood, hindering effective post-transplant management.

Purpose of the Study:

• To elucidate lung adaptation processes post-transplantation.
• To correlate these adaptations with future allograft function.

Main Methods:

• Exploratory cohort study of 78 lung transplant recipients and donors.
• Analysis of bronchoalveolar lavage samples using 16S rRNA sequencing (microbiome), flow cytometry (cellular composition), metabolomics, and lipidomics.

Main Results:

• Distinct temporal dynamics observed across multi-omics data.
• Recipient and environmental factors, not donor microbiome, shape the long-term lung microbiome.
• Bacterial abundance correlated with pre-existing lung diseases.
• Machine learning accurately predicted forced expiratory volume in one second (FEV1) decline using multi-omics data, with microbial profiles showing high predictive power.

Conclusions:

• Bronchoalveolar microbiome, cellular, metabolic, and lipidomic profiles exhibit specific temporal patterns post-lung transplant.
• The lung microbiome accurately predicts future lung function changes.