Preventing necrotizing enterocolitis with a precision phage cocktail
A prophylactic, orally administered bacteriophage cocktail targeting the pre-symptomatic Klebsiella/Enterobacteriaceae bloom that precedes NEC in very-low-birth-weight infants — intercepting the disease days before the bowel is lost.
Working title · Aligned to NIH/NICHD & NIAID priorities · Mechanism: R01 / U-series translational
Illustrative figures for framing; the proposal would cite center-specific incidence and the GRAM/cohort literature precisely.
Specific Aims
The one-page core
Problem. Necrotizing enterocolitis is the most lethal acquired gastrointestinal emergency of prematurity. Despite decades of effort there is no approved preventive therapy, and the standard of care once NEC begins — broad-spectrum antibiotics and surgery — arrives after the damage is underway and further deranges the fragile preterm microbiome.
Opportunity. NEC does not strike at random. Strain-resolved metagenomics shows a reproducible Klebsiella and fimbriated-Enterobacteriaceae bloom — with elevated bacterial replication rates — in the days before onset. In preterm piglets, transferring a bacteria-free fecal virome prevents NEC, and UV-inactivating the phages abolishes that protection. The active agent is bacteriophage.
Hypothesis. A defined, GMP-grade lytic phage cocktail targeting the dominant preterm-gut Klebsiella lineages, administered prophylactically to high-risk VLBW infants, will suppress the pre-NEC bloom and reduce NEC incidence — without the collateral microbiome damage of antibiotics.
Build & characterize the cocktail
Assemble a panel of strictly lytic phages against contemporary NICU Klebsiella pneumoniae/variicola and K. aerogenes isolates from partner units. Characterize host range across a banked isolate library, sequence for absence of toxin/lysogeny/AMR genes, and select a 4–6 phage cocktail that maximizes coverage and suppresses resistance (independent receptors). Define stability and dosing in simulated infant-gut conditions and with human milk.
Establish efficacy & safety preclinically
Test the cocktail in established preterm-piglet and gnotobiotic/humanized-mouse NEC models colonized with patient-derived Klebsiella. Primary outcomes: NEC incidence/severity (histology) and target-strain suppression by replication-rate metagenomics. Safety: systemic phage exposure, immune activation, and — critically — off-target microbiome integrity (16S/shotgun) to demonstrate the commensal-sparing advantage over antibiotics.
Biomarker-guided first-in-infant safety study
Under an FDA emergency/expanded-access IND framework, run a phase 1 safety and pharmacokinetic (gut-persistence) study in stable VLBW infants, layered on a surveillance program that tracks the Klebsiella replication-rate biomarker. The design positions a subsequent biomarker-triggered prophylaxis trial: treat when the bloom signature appears, before symptoms.
Significance
NEC mortality has barely moved in 30 years. A preventive that is precise enough to use prophylactically in well infants — because it spares the microbiome — would be a category change, not an increment.
Innovation
First defined phage cocktail aimed at NEC prevention; first use of a replication-rate microbiome biomarker to trigger a phage intervention before symptoms; a model for precision antibacterials in the most vulnerable patients.
Risks & mitigations
Phage resistance → multi-receptor cocktail + monitoring. Strain mismatch → regional isolate banking + adaptive composition. Regulatory novelty → magistral/eIND precedent and Eur. Ph. 5.31 quality framework. Immune clearance → short prophylactic windows.
Team & partners (to assemble)
A translational consortium
- Neonatology / NICU network — isolate banking, surveillance, eventual trial sites.
- Phage biology lab — isolation, engineering, host-range and resistance work (e.g. a TAILOR-style at-cost academic pipeline).
- Microbiome metagenomics — the replication-rate biomarker (Banfield/Morowitz-style strain resolution); a natural role for Microbiome Medicine.
- GMP phage manufacturing — titre, stability, release testing to Eur. Ph. 5.31.
- Regulatory / bioethics — neonatal eIND, consent in a vulnerable population.
Open science
Steal this grant
We believe the fastest way to catalyze the field is to give the proposal away. The full grant and a reusable, funder-agnostic template are free to read, copy, and adapt for your own indication — no permission required.