Identification of Immunogenic Epitopes for Lyme Disease Using Machine Learning-Based Comprehensive Reactivity Profiling of Patients Antibodies
Abstract
Rationale: Lyme disease (LD) is the most prevalent tick-borne zoonotic disease in the United States caused by the Borrelia burgdorferi bacterium, with estimated 476,000 new yearly cases and rapidly increasing incidence. While most early LD are diagnosed by the presence of erythema migrans (EM) or bullseye rash at the bite site, 20%-30% of patients do not present with an EM complicating early diagnosis of the disease when treatment is most effective. The molecular diagnostic test recommended by the American Centers for Disease Control and Prevention is based on the detection of LD-specific antibodies in patient’s blood and has a number of marked limitations. Low test efficacy combined with similarity of symptoms to other febrile illnesses such as seasonal flu, Epstein-Barr virus, and others highlights an urgent need for improved diagnostic tests capable of resolving LD from other diseases. To address this need, this project is focused on identifying a panel of biomarkers for distinguishing LD from other febrile diseases with similar symptoms. To this end, we are using an approach which, in contrast to other studies conducted so far, profiles an entire antibody response of each individual patient and characterizes it through advanced machine learning to generate a mathematical model of the response for each patient. These models are then used to identify short segments in the pathogen proteome that can then be used as candidate biomarkers for the disease. In the proposed research, we will conduct the response profiling experiments, characterize the response through machine learning and identify candidate biomarkers to distinguish LD from other febrile illnesses followed by their confirmation in a prototype clinical test. Objective: The proposed research is focused on the identification and validation of a panel of biomarkers for distinguishing an improved diagnostic test for acute Lyme disease capable of distinguishing LD from other febrile. It is expected that the candidate biomarkers discovered by this effort can form the basis of a clinical lab test and have the potential to be developed into a point-of-care test. The proposed research is relevant to two Fiscal Year 2021 Tick-Borne Disease (TBD) Research Program Focus Areas: (1) Diagnostic biomarker panel for Lyme disease and/or other TBDs that distinguishes tick-borne infection from other febrile illnesses and (2) Innovative approaches that provide diagnosis for a single or multiple tick-borne infections. The proposed study will: (1) Serve as a stepping-stone toward accurate differential diagnosis of LD in its early stage ,for which there is currently no approved tests, thus allowing for marked reduction in misdiagnosed and untreated patients, (2) Provide a deeper insight into LD by enabling better understanding about humoral immune system response at the single-patient level, and (3) Enable a direct comparison of humoral immune system response in LD and other febrile diseases. This has the potential of opening new avenues for future research not only in LD, but also in other infectious diseases. Impact: The test will reduce the burden of both the general public and military personnel by preventing disease progression to chronic, late stages for which there are currently no treatment options available. When fully developed, the test will have the potential to attain broad clinical utility due to its superior performance and relatively low cost. When developed into a diagnostic, the results of this work will directly benefit patient population by markedly improving treatment outcomes and reducing public health care burden. It is anticipated that following completion of the work, a diagnostic test based on the identified candidate biomarkers will be developed over 1-2 years. The proposed work is a discovery stage effort to identify candidate biomarkers for the differential diagnosis of Lyme disease from other febrile illnesses. Part of the proposed work is t
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Dec 28, 2022
- Source ID
- W81XWH2210204
Entities
People
- Neal W. Woodbury
Organizations
- Arizona State University
- United States Army