Long COVID: Immune Signatures and Global Underreporting

Research has identified distinct biological signatures in the blood of individuals with long COVID, potentially providing a path toward objective biomarkers for a condition often defined by subjective symptoms.

A study published in Nature involving 275 individuals utilized multidimensional immune phenotyping and unbiased machine learning to identify biological features associated with long COVID. The research identified marked differences in circulating myeloid and lymphocyte populations when compared to matched controls.

The findings revealed that participants with long COVID exhibited lower levels of cortisol. These individuals showed evidence of exaggerated humoral responses directed against SARS-CoV-2.

The study also observed higher antibody responses directed against non-SARS-CoV-2 viral pathogens among those with long COVID, with a particular note regarding the Epstein-Barr virus.

Symptom-Specific Immune Markers

Some research has focused on linking specific immune markers to particular symptom clusters. In a study published January 29, 2026, in Scientific Reports, researchers analyzed patients with persistent gastrointestinal long COVID two years after severe SARS-CoV-2 infection.

Network analysis from that study suggests that IL-6 may contribute to immune dysregulation in long COVID patients who report chronic gastrointestinal symptoms.

However, other research has presented different findings regarding inflammation. A study published in ScienceDirect assayed a panel of 65 cytokines, chemokines and soluble receptors associated with immune activation and found a lack of systemic inflammation in patients with long COVID-19 syndrome.

Diagnostic Challenges and Underreporting

The effort to identify distinct immunological signatures is critical due to significant diagnostic challenges. A report published November 14, 2025, in ScienceDirect noted that the global deployment of COVID-19 vaccines has introduced difficulties because vaccine-related adverse events can have symptoms that overlap with long COVID.

evidence suggests that long COVID is significantly underreported across the globe.

These diagnostic hurdles are complicated by the variety of symptoms associated with the syndrome, which frequently include:

• Unremitting fatigue
• Post-exertional malaise
• Cognitive dysfunctions
• Autonomic dysfunctions

The integration of immune phenotyping data into machine learning models is intended to help guide future studies into the pathobiology of long COVID and assist in the development of relevant biomarkers to better identify and categorize the syndrome.