NIH-Funded Study Targets Saliva-Based Long COVID Diagnosis

Long COVID presents with heterogeneous, persistent symptoms that complicate clinical assessment and case definition. A lack of a definitive diagnostic test forces clinicians to exclude alternative etiologies before assigning a diagnosis. Saliva offers a non-invasive sample type that can be collected repeatedly across disease stages, making it well suited for biomarker research aimed at clarifying disease status over time. Building on this potential, a new project will examine salivary biomarkers to support diagnosis and risk stratification in Long COVID.

UNC Adams School of Dentistry (University of North Carolina; Chapel Hill, NC, USA) has received an R01 grant from the National Institutes of Health (NIH) to determine whether biomarkers in saliva can diagnose long COVID and predict which acutely infected patients are likely to develop persistent symptoms. The effort will also analyze saliva collected during acute COVID-19 to evaluate signatures associated with future long COVID. The goal is to translate salivary changes into clinically actionable markers for patient classification.

The team will analyze saliva for immune and host-response signals, including proteins, metabolic mediators, and microbes, and use a machine learning pipeline to identify oral signatures of disease. By processing large patient datasets, the models are expected to improve understanding of Long COVID pathogenesis and help distinguish recovery from persistent illness. The project builds on earlier work in which the group evaluated antiseptic rinses to reduce salivary viral load in dental settings and, after collecting saliva from hundreds of patients, began investigating how SARS-CoV-2 affects the oral immune environment.

Study plans include recalling additional participants and enrolling newly infected patients who will be followed through serial saliva collection. At least three months after infection, participants will complete validated symptom questionnaires and be grouped as fully recovered or showing signs of Long COVID. Their saliva samples, linked with health histories, will then be analyzed to determine how salivary proteins, metabolic mediators, and microbes are associated with Long COVID outcomes.

Preliminary analyses of banked early-infection samples showed altered salivary protein and metabolite profiles among individuals who later developed Long COVID. Proteomic testing also identified dozens of differentially expressed proteins related to oral inflammation, oxidative stress responses, and immune dysregulation. If validated, the researchers note, these salivary biomarkers could support diagnostic and prognostic tools and help identify higher-risk patients for more proactive management during acute infection.

“Based on preliminary results, patients who develop long COVID had significantly altered salivary protein and metabolite profiles relative to patients who recovered. Proteomics results identified 47 differentially expressed proteins between severe long COVID patients and controls. There were several unexpected highly abundant proteins associated with oral inflammation, oxidative stress response, and immune dysregulation. These are interdependent mechanisms perpetuating a chronic cycle of cellular damage that may drive Long COVID pathology that we could identify in saliva during acute infection- which is pretty amazing!” said Laura Jacox, DMD, PhD, MS.

“For patients who have long-term long COVID symptoms, it would be wonderful to provide a non-invasive diagnostic test. This would save efforts and costs that are currently being spent ruling out other potential causes, and a test provides comfort to patients and informs providers,” said Jacox.

Related Links
UNC Adams School of Dentistry