Project Summary/Abstract
The complexities of heterogeneity, stratification, and staging of sepsis have contributed to the poor
translatability of current molecular models for diagnosis and treatment of human sepsis. The National Advisory
General Medical Sciences Council advised using discovery approaches to characterize sepsis using human
biospecimens. Sepsis, a life-threatening organ dysfunction caused by a dysregulated host response to a
bacterial, viral, or fungal infection, is a major cause of death in premature (≤32 weeks) and low birth weight (≤
1500g) infants and occurs in up to 25% of such infants.1-3 The ability to collect microbiome and tissue
biospecimens longitudinally from multiple body sites under extremely controlled circumstances prior to, during,
and throughout sepsis (culture proven), sepsis-like (culture negative critical illness), and "normal" growing
premature infant conditions will allow our team to determine optimal sample collection, storage, and processing
protocols using small volume samples to enhance sepsis research rigor, and to develop new strategies for
sepsis detection by discerning pathways that contribute to the pathophysiology of sepsis in the premature
neonatal population.
In the R21 phase we will collect clinical information and stool, blood, and saliva specimens longitudinally from
preterm infants (≤32 weeks and ≤1500g) on days 1, 3, 7, 14, 21, 28, 35, 42, and 49 of life, and additionally
during suspected sepsis events. Standardization and documentation of sample collection, storage, and
processing and sample validation studies will ensure rigor and reproducibility and inform the field.
In the R33 phase we will collect samples from 2 clinical sites providing a test set for machine learning methods.
Utility of the newly collected biospecimens and associated clinical data will be demonstrated by constructing a
multi-omic network for predicting causal mechanisms from genes/metabolites/clinical parameters differentially
expressed between clinically well neonates and those with culture proven sepsis and culture negative systemic
inflammatory illness. We will make establish testable causal inferences and predictive models of microbial/host
gene interactions and biologic mechanisms leading to sepsis that could form the foundation for future
mechanistic studies.
Our team of neonatal providers, immunologists, systems biologists, microbiome specialists, and
bioinformaticians supported by two large Level IV NICUs are uniquely positioned to collect and analyze
biospecimens and clinical data before, during, and after critical illness in patients with extreme prematurity.