Rapid, Point-of-Care, Host Gene Expression Test for Presymptomatic Viral Infection and to Distinguish Bacterial from Viral Illness
Abstract
As COVID-19 has shown, tests to detect emerging viral threats are absolutely essential for the public health response. However, there is a lag of many months before tests are readily available to detect new and emerging viral pathogens. This dangerous diagnostic void cripples containment efforts and allows disease to propagate. One way to fill this void is to utilize a pathogen-agnostic test that identifies patients with viral infection when they are acutely ill or ideally, when pre-symptomatic. Measuring the host’s response to infection offers the most advanced solution to this problem. Exposure to a pathogen leads to robust immunological responses. In later stages of illness, this immunological response can be measured in a person’s antibodies. However, there are earlier host responses that occur within hours of exposure. We have discovered gene expression biomarkers that identify the presence of a viral infection (PreV test). This viral response signature can be detected days before symptoms occur. The signature itself can be used to identify the presence of ANY virus, making it ideally suited to detect new pathogens. The PreV test can also be modulated to be virus specific, as we have done for SARS-CoV-2 and influenza. Another major challenge in dealing with emerging viral infections is the need to accurately distinguish viral from bacterial disease. The absence of reliable diagnostic tests contributes to inappropriate antibacterial use, which fuels the antimicrobial resistance crisis. This disproportionately impacts respiratory health because providers cannot confidently discriminate bacterial from viral infection and because antimicrobial resistance is making it harder to treat bacterial infections when they do occur. Our solution to this problem relies on the same machine learning approaches used to develop the PreV signature. Specifically, we discovered a host gene expression signature that discriminates bacterial from viral infection (B/V). Our B/V test performs better than any currently available diagnostic strategy for bacterial/viral discrimination. These two signatures, PreV and B/V, are only clinically meaningful if they can be measured at the point of need. We have therefore partnered with Biomeme, a leader in point-of-care molecular diagnostic technology. Together, we will develop the PreV and B/V tests on their state-of-the-art testing platform known as the Franklin-ISP. This rapid, sample-to-answer platform is hand-held, battery-powered, smartphone-operated, and offers CLIA-waivable simplicity, meaning it can be operated nearly anywhere. The system also offers enough flexibility to combine our host response strategy with direct pathogen detection. The result would be a fully comprehensive assessment of the patient, from both the immunological and pathogen perspectives. The successful completion of the proposed research will deliver two tests. PreV can detect emerging viral infections without having to detect the virus itself. Moreover, this test is effective during the pre-symptomatic phase when people are unknowingly contagious. The second test, B/V, will empower clinicians to know when a bacterial infection is present and when antibiotics are needed. Consequently, we can improve respiratory health through better patient management and a decrease in antimicrobial resistance.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Dec 05, 2021
- Source ID
- W81XWH2110741
Entities
People
- Ephraim L Tsalik
Organizations
- Duke University
- United States Army