2021. An Ultra-Rapid Host Response Assay to Discriminate Between Bacterial and Viral Infections Using Quantitative Isothermal Gene Expression Analysis
Session: Poster Abstract Session: Diagnostics: Biomarkers and Novel Approaches
Saturday, October 6, 2018
Room: S Poster Hall
  • 180926 IDWeek Poster_isothermal.pdf (801.6 kB)
  • Background:

    Accurate diagnosis and treatment of bacterial infection is critical for improving patient outcomes. However, over-prescription of antibiotics has contributed to C. difficile-infections and to the emergence of antimicrobial resistance. As assessing bacterial infection by culture is slow and molecular pathogen detection is limited in scope, an unmet need remains for a rapid test to differentiate between viral and bacterial infections. We have previously identified a set of 7 host response biomarkers demonstrating an AUROC of 0.91-0.93 for separating bacterial from viral infections across multiple independent cohorts. A clinical in-vitro diagnostic test (IVD) using these markers must be very fast to integrate with physician workflows. Loop-mediated isothermal amplification (LAMP) represents a rapid amplification solution with the potential to meet these needs. We describe an ultra-rapid LAMP assay designed to quantitate these markers for applications in point-of-care decision making.


    LAMP primers for gene expression analysis of selected markers with a housekeeping control were designed for mRNA specificity by targeting FIP primers to splice junctions. Assay specificity, sensitivity and linear dynamic range were evaluated using serial dilution of control material. RNA extracted from preserved patient samples was evaluated by LAMP for concordance with NanoString® nCounter® data (nCounter).


    Iterative optimization of primer design resulted in RT-LAMP assays that selectively amplify target mRNA. Assays demonstrate a linear dynamic range spanning 6 orders of magnitude and a quantitative LOD of about 103 copies. Quantitation of relative expression levels showed good concordance with nCounter data in 10 healthy, 6 viral and 6 bacterial patient samples, with average threshold times of <15 minutes.


    Accurate discrimination of bacterial and viral infection can be achieved on a true point-of-care timescale using our LAMP strategy. This assay could be run on a standard thermal cycler, or any quantitative molecular instrument that allows to measure at least 8 targets. An IVD test distinguishing between bacterial and viral infections could aid in antimicrobial treatment decisions and thereby minimize over-prescription of antibiotics.

    David Rawling, PhD, Wensheng Nie, PhD, Melissa Remmel, PhD, Mark Eshoo, PhD, Jonathan Romanowsky, MBA, Oliver Liesenfeld, MD and Timothy Sweeney, MD/PhD, Inflammatix Inc, Burlingame, CA


    D. Rawling, Inflammatix Inc: Employee , Salary .

    W. Nie, Inflammatix Inc: Employee , Salary .

    M. Remmel, Inflammatix Inc: Employee , Salary .

    M. Eshoo, Inflammatix Inc: Employee , Salary .

    J. Romanowsky, Inflammatix Inc: Employee , Salary .

    O. Liesenfeld, Inflammatix Inc: Employee , Salary .

    T. Sweeney, Inflammatix Inc: Employee , Salary .

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 3rd with the exception of research findings presented at the IDWeek press conferences.