Session : Innovations in Diagnostics

152 - blood2bac: Species ID and AMR Prediction of Bacterial Pathogens at Low Concentrations in Blood Using a Rapid Ultra-high Enrichment Process and Nanopore Sequencing

Oral Abstract

Location: On Demand

Background: Each year in the United States there are over 1.7 million cases of sepsis that account for a third of hospital deaths. A key to reducing morbidity and mortality rates is early, appropriate antibiotic therapy. Most new diagnostic approaches still suffer from insufficient sensitivity to low bacterial loads in blood and limited sets of detection targets for bacterial species identification (ID) and antimicrobial resistance (AMR) determination. As such, blood culture remains the gold standard for diagnosing bacteremia despite limitations such as > 2-day turnaround time (TAT), incompatibility with fastidious organisms, and frequent inability to recover causative pathogens.

Methods: 31 clinically relevant bacterial pathogens, made up of 17 gram-positive and 14 gram-negative bacterial species, were spiked into 2 to 4 healthy donor blood samples at 1 to 5 CFU/mL. The samples were run through our proprietary Blood2Bac™ pipeline, sequenced on a nanopore platform, and data were passed through Keynome®, our proprietary machine learning algorithm to determine species ID and AMR.

Results: By assessing the efficiency of pathogen DNA enrichment and genome coverage post sequencing, we report high performance of 3 CFU/mL for 3 bacterial species and ≤ 2 CFU/mL for the 28 remaining species, which includes S. aureus, E. coli, and Streptococcus spp., three of the leading causes of sepsis.
For all 31 bacterial species tested, Keynome called species ID with 100% accuracy. In addition, Keynome also predicted the AMR profile of pathogens with 100% accuracy for 19 drug/species AMR combinations, including ciprofloxacin for E. coli, clindamycin for S. aureus, and aztreonam for K. pneumoniae.

Conclusion: Blood2Bac is able to enrich a wide range of bacterial pathogens directly from blood and enable bacterial whole genome sequencing with an estimated TAT of 12 hours. When coupled with Keynome, our process provides accurate species ID and AMR calls for key BSI pathogens even at single-digit CFU/mL concentrations. Our species-agnostic and culture-free process enables detection of a diverse range of bacterial species with high sensitivity, providing a robust and comprehensive diagnostic.

Presenting Author(s)

Chiahao Tsui

Senior Research Scientist
Day Zero Diagnostics
Somerville, MA

Disclosure: Day Zero Diagnostics (Employee, Shareholder)

Slides Slides and Audio

Co-Author(s)

LC

Lisa S. Cunden, PhD

Research Scientist
Day Zero Diagnostics
Cambridge, Massachusetts

Disclosure: Day Zero Diagnostics (Shareholder)
NB

Nicole Billings, PhD

Director, Lab Services
Day Zero Diagnostics
Boston, Massachusetts

Disclosure: Day Zero Diagnostics (Employee)
IN

Imaly A. Nanayakkara, PhD

Research Scientist
Day Zero Diagnostics
Boston, Massachusetts

Disclosure: Day Zero Diagnostics (Employee, Shareholder)
IH

Ian Herriott, BS

Manager of Sequencing Technologies
Day Zero Diagnostics
Boston, Massachusetts

Disclosure: Day Zero Diagnostics (Employee, Shareholder)
MK

Michael E. Kumcu, MS

Research Associate II
Day Zero Diagnostics
Boston, Massachusetts

Disclosure: I do not have any relevant financial / non-financial relationships with any proprietary interests.
RM

Rachel R. Martin, n/a

Research Associate
Day Zero Diagnostics
Watertown, Massachusetts

Disclosure: Day Zero DIagnostics (Employee)
MC

Michelle Chen, MS

Research Associate
Day Zero Diagnostics
Boston, Massachusetts

Disclosure: Day Zero Diagnostics (Employee, Shareholder)
FP

Febriana Pangestu, n/a

Genomics Sequencing Technician
Day Zero Diagnostics
Boston, Massachusetts

Disclosure: Day Zero Diagnostics (Employee, Shareholder)
PK

Paul Knysh, PhD

Research Scientist, Machine Learning
Day Zero Diagnostics
Boston, Massachusetts

Disclosure: Day Zero Diagnostics (Employee)
CM

Cabell Maddux, n/a

Senior Software Engineer
Day Zero Diagnostics
Providence, Rhode Island

Disclosure: Day Zero Diagnostics (Employee, Shareholder)
ZM

Zachary Munro, n/a

Software Engineer
Day Zero Diagnostics Inc.
Somerville, Massachusetts

Disclosure: Day Zero Diagnostics Inc.: Employee, Shareholder
MH

Miriam Huntley, PhD

Chief Technology Officer
Day Zero Diagnostics
Boston, Massachusetts

Disclosure: Day Zero Diagnostics (Employee, Shareholder)