Microbiome and host response signatures for pneumonia among lung transplant recipients

University Of Pittsburgh At Pittsburgh, Pittsburgh PA

Investigators

Linked publications, trials & patents

Abstract

Project Summary Pneumonia (PNA) is a leading cause of death following lung transplantation (LTx). Tracheobronchitis (TB) is commonly diagnosed in LTx recipients in whom a respiratory tract infection (RTI) is suspected, and diagnostic criteria for PNA are not fulfilled. However, conclusive pathologic evidence for TB in the setting of LTx is lacking, and it is unclear that the diagnosis constitutes a coherent disease entity. Moreover, RTIs in LTx recipients are often difficult to distinguish clinically from colonization in the absence of infection (COL). Recent DNA sequencing-based microbiome profiling studies have revealed that PNA is characterized by loss of microbial diversity in the lower respiratory tract, and emergence of a dominant pathogen. Microbiome and host response markers that are associated with PNA, TB or COL following LTx are undefined. In preliminary studies, we identified multivariable, non-culture-based signatures within bronchoalveolar lavage fluid (BALF) samples from LTx recipients that distinguished PNA and TB from COL, which were comprised of 16S microbiome and cytokine covariates. Compared to COL, PNA was characterized by loss of diversity and a robust pro-inflammatory cytokine response. In contrast, TB was characterized by high microbial diversity and multifunctional cytokine responses that differed from those of PNA. Taken together, the preliminary data suggest that PNA and TB stem from different pathogenic processes, and that microbiome and cytokine profiling may have utility as diagnostic adjuncts to cultures and conventional criteria. The objective of this project is to refine and validate multivariable signatures for PNA, TB and COL among LTx recipients who are not mechanically ventilated. In specific aim 1, we will refine the signatures by performing 16S microbiome profiling for bacteria, assessing the presence of respiratory viruses, and measuring cytokine responses in BALF samples that are banked in our LTx biorepository. Studies will expand upon our preliminary data by including a larger number of samples for each diagnosis and detecting respiratory viruses as well as bacteria, thereby identifying signatures that are more robust and representative of the LTx population. We will employ powerful multivariable regression analysis methodologies developed in our preliminary studies. In specific aim 2, we will validate microbiome and cytokine signatures, using an independent set of BALF samples from the biorepository and samples collected prospectively from LTx recipients. To our knowledge, we are the first group to describe the microbiome and host response underpinnings of PNA, TB and COL in LTx recipients. We anticipate that this study will define optimized, disease-specific signatures for these diagnoses, which are not contingent upon culture results. The project is expected to lead to multi-center trials validating the signatures, and to studies in settings outside of LTx, such as among patients with ventilator-associated RTIs or community-acquired PNA. Furthermore, our results will provide insights into possible pathogenic mechanisms for PNA and TB following LTx, which can inform subsequent mechanistic studies.