Worldwide, infectious disease remains the leading cause of neonatal mortality and results in one million newborn deaths each year. Rapid and precise profiling of pathogens is the key to targeted and effective clinical management of infectious disease. However, current diagnostic methods are slow, limited in breadth of detection, and often unreliable. Further, they have limited ability to detect polymicrobial infection and suffer from poor specificity due to an inability to distinguish clinically relevant from non-pathogenic DNA. Genotyping of pathogen genomic sequences using High Resolution Melt (HRM) provides a simple, rapid, and modern alternative to blood culture testing. HRM generates sequence specific melt curves in a closed-tube reaction as PCR-amplified DNA fragments are heated and disassociate. Our team has advanced this technology through the proof-of-concept stage, demonstrating that a novel digital PCR format for HRM can achieve accurate universal genotyping and quantification of DNA targets in under 3 hours. DNA sequences present in mixtures are individually amplified and identified using the principles of microfluidic sample partitioning and machine learning to enable sensitive specific polymicrobial detection of all the targets present in the DNA mixture. However, for clinical integration, critical innovations are needed. This includes optimization of our upstream processing steps to selectively capture, load, and detect DNA sequences from only viable pathogen cells from a small-volume (? 1mL) neonatal blood sample. Further, methods to establish linkages between pathogen identity and antibiotic resistance are needed at the single cell level to fully characterize and accurately quantitate the resistance profile of the pathogen population for precise targeted treatment. In phase I of this proposal, we will focus on refining sample preparation steps to (1) maximize our sensitivity and (2) enable co-localization of resistance markers with identified microbes. We hypothesize that we can efficiently capture pathogen load from blood, reducing interference from human cell contents and enable single cell analysis and quantitation of only viable pathogens while linking species identification and resistance. The processes and tests developed in this proposal will provide: 1) high negative predictive value to rule out bloodstream infection; and 2) specific identification and quantitation of pathogens along with co-localization of relevant antibiotic resistance genes for the targeted treatment of true infections while limiting influence from contaminants or environmental microbes of no clinical significance. In Phase II, the system will be developed to its beta form for testing in a clinical environment. Melio?s goal is to ultimately deliver a low-cost, small-footprint benchtop device compatible with the clinical workflow.

Public Health Relevance

We propose the development of a culture-independent molecular diagnostic test that will enable the rapid and precise detection of infections in newborns. This test uniquely combines the cutting-edge technologies of machine learning, microfluidic partitioning of samples for digital interrogation, and High-Resolution Melting of nucleic acids to create unique pathogen fingerprints for all high-risk bacterial, fungal, and viral pathogens and associated resistance marker(s) present in small-volume blood samples in under 3 hours. Such a timely and accurate test is needed for effective antimicrobial therapy in newborns suspected of infection to reduce antibiotic use in non-infected patients while aggressively treating those in need with narrow spectrum antibiotics targeted towards the offending pathogens.

Agency
National Institute of Health (NIH)
Institute
National Institute of Allergy and Infectious Diseases (NIAID)
Type
Small Business Innovation Research Grants (SBIR) - Phase I (R43)
Project #
1R43AI145567-01
Application #
9778646
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Ritchie, Alec
Project Start
2019-08-13
Project End
2020-07-31
Budget Start
2019-08-13
Budget End
2020-07-31
Support Year
1
Fiscal Year
2019
Total Cost
Indirect Cost
Name
Meliolabs Inc.
Department
Type
DUNS #
081117828
City
Santa Clara
State
CA
Country
United States
Zip Code
95054