Long COVID (LC), characterized by persistent and debilitating symptoms following acute SARS-CoV-2 infection, affects a significant portion of individuals and its etiology remains unclear. Several mechanisms are proposed, including aberrant immune responses, microvascular dysregulation, and viral reactivation. While higher SARS-CoV-2 Spike-specific antibody levels are often observed in those with persistent symptoms, studies indicate adaptive immune dysregulation and exhaustion are also involved. Neutralizing antibody capacity wanes faster than total Spike IgG levels, and the association between longitudinal antibody dynamics or the breadth of the neutralizing response and LC phenotypes is poorly understood. Previous studies have shown a potential link between a broader antibody response to other coronaviruses and Long COVID, suggesting that the breadth of the initial response could be crucial in LC development. Given the emergence of Omicron variants that evade neutralization from previous strains, understanding the relationship between the initial antibody response's breadth and durability and Long COVID is critical.

The literature review section highlights existing research on the various aspects of Long COVID and immune responses following SARS-CoV-2 infection. Studies have explored the role of aberrant autoreactive immune responses, microvascular dysregulation, and the reactivation of latent viruses in the pathogenesis of Long COVID. The persistence of SARS-CoV-2 subgenomic RNA and proteins in tissues of some individuals with Long COVID has also been documented. While some studies have noted higher antibody levels in individuals with persistent symptoms, others have reported associations between weaker antibody responses over time and Long COVID. A recent preprint indicated a possible correlation between an expanded antibody response to OC43 coronavirus and Long COVID, suggesting that the breadth of the immune response might play a significant role in the development of this condition.

This study employed a longitudinal design, enrolling participants infected with SARS-CoV-2 during the early pandemic waves before widespread vaccination. Plasma samples were collected at approximately 1, 2, and 4 months post-infection. The researchers used a PhenoSense assay to measure neutralizing and cross-neutralizing antibody responses against various SARS-CoV-2 variants (original strain, Alpha, Delta, Omicron BA.1, and BA.5). A subset of participants with high or low Omicron BA.5 responses were further tested with an expanded panel of Omicron subvariants (BA.4.6, BQ.1.1, XBB.1.5, and BA.2). Statistical analyses included non-parametric tests (Mann-Whitney and Friedman tests) for cross-sectional comparisons and mixed-effects models for longitudinal analyses, accounting for demographic and clinical factors (age, sex, hospitalization, diabetes, BMI, Long COVID symptoms). Logistic regression was used to assess the association between neutralizing antibody responses and the presence of Long COVID and specific symptom phenotypes.

Several key relationships between SARS-CoV-2 antibody neutralization and Long COVID were identified. While neutralizing antibody responses to the original infecting strain were not associated with Long COVID in cross-sectional analyses, cross-neutralization ID50 levels to Omicron BA.5 at approximately 4 months post-infection were significantly and positively associated with increased odds of Long COVID overall and with specific symptom phenotypes (neurocognitive and gastrointestinal). Longitudinal modeling revealed additional significant associations between overall neutralization levels and decay rates with Long COVID phenotypes. Participants with gastrointestinal and cardiopulmonary symptoms exhibited significantly higher SARS-CoV-2 neutralizing responses to the original strain across all time points, but with faster decay. A higher proportion of participants had antibodies capable of neutralizing Omicron BA.5 compared to BA.1 or XBB.1.5. Analysis of a subset of participants showed consistently higher cross-neutralization to BA.2, BA.4.6, and BQ.1.1, but lower neutralization to XBB.1.5, similar to BA.1.

The findings suggest a complex interplay between immune responses and Long COVID, with the breadth of antibody neutralization potentially being a key factor. The lack of association between neutralization to the original strain and Long COVID in cross-sectional analyses, contrasted with the significant association observed with Omicron BA.5 cross-neutralization, highlights the importance of considering the breadth of the antibody response. The faster decay of neutralizing ID50 in those with certain Long COVID phenotypes (cardiopulmonary, musculoskeletal) suggests that the temporal dynamics of the immune response might also be crucial. Persistent SARS-CoV-2 antigen presentation in tissues could potentially explain the higher antibody neutralization levels observed over time, potentially contributing to broader responses to subsequent variants. The study supports the idea that different analytical approaches (cross-sectional vs. longitudinal) can provide complementary insights into the complex relationships between immune responses and Long COVID.

This study provides novel insights into the relationship between neutralizing antibody responses and Long COVID, indicating that the breadth of the antibody response, particularly against subsequent viral variants, may be a predictor of Long COVID development and specific symptom phenotypes. Further research is needed to validate these findings in other cohorts and to explore the underlying mechanisms driving these associations. Investigating the role of persistent antigen presentation and the temporal dynamics of immune responses could provide a more complete understanding of Long COVID pathogenesis.

This study's limitations include the use of a convenience sample collected before widespread vaccination and the emergence of Delta and Omicron variants. This focus limits the generalizability of the findings to later pandemic phases. The assay used Spike protein pseudoviruses rather than live virus, although such assays have shown comparable results in other studies. The lack of participants infected with more recent variants prevents the extension of these observations to later pandemic waves. Although the study addressed potential biases by assessing both Long COVID and non-Long COVID participants identically, caution should be exercised when extrapolating findings to all individuals with prior COVID-19.