Early biological markers of post-acute sequelae of SARS-CoV-2 infection

Abstract

To understand the roles of acute-phase viral dynamics and host immune responses in post-acute sequelae of SARS-CoV-2 infection (PASC), we enrolled 136 participants within 5 days of their first positive SARS-CoV-2 real-time PCR test. Participants self-collected up to 21 nasal specimens within the first 28 days post-symptom onset; interviewer-administered questionnaires and blood samples were collected at enrollment, days 9, 14, 21, 28, and month 4 and 8 post-symptom onset. Defining PASC as the presence of any COVID-associated symptom at their 4-month visit, we compared viral markers (quantity and duration of nasal viral RNA load, infectious viral load, and plasma N-antigen level) and host immune markers (IL-6, IL-10, TNF-α, IFN-α, IFN-γ, MCP, IP-10, and Spike IgG) over the acute period. Compared to those who fully recovered, those reporting PASC demonstrated significantly higher maximum levels of SARS-CoV-2 RNA and N-antigen, burden of RNA and infectious viral shedding, and lower Spike-specific IgG levels within 9 days post-illness onset. No significant differences were identified among a panel of host immune markers. Our results suggest early viral dynamics and the associated host immune responses play a role in the pathogenesis of PASC, highlighting the importance of understanding early biological markers in the natural history of PASC.

Fig. 1. Flow diagram of participants evaluated for post-acute sequelae of SARS-CoV-2 infection (PASC) at month 4.

Flow diagram showing number of enrolled participants, COVID-19 and uninfected participants, and final number of participants who completed follow-up.

Fig. 2. Comparison of plasma N-antigen and IgG Spike antibody levels among those with and without post-acute sequelae of SARS-CoV-2 (PASC) over a 28-day period after symptom onset.

a, b Estimated mean values and 95% confidence intervals are plotted for biological specimen among PASC and non-PASC groups at days 5, 9, 14, 21, and 28 for N-antigen (a) and IgG (b) using generalized estimating equations fit with independent correlation, identity linkage, and Gaussian distribution. Covariates included sex, age, vaccination status, and SARS-CoV-2 variant. Statistics are derived from 80 participants with N-antigen measurements and 48 unvaccinated participants with IgG measurements. Statistical significance was assessed using two-tailed tests. Statistically significant distributions include N-Ag at day 5 (p = 0.01), IgG spike antibody at day 5 (p = 0.04), and IgG spike antibody at day 9 (p = 0.03).

Fig. 3. Proportion of RNA and infectious viral shedding for each day after symptom onset among those with and without PASC.

a, b Bar graphs demonstrating proportion of nasal samples positive for (a) viral shedding and (b) infectious virus by day post-symptom onset from day 3 to day 18 among 32 PASC and 72 non-PASC participants. Two-sided pooled logistic regression demonstrated statistically significant difference in the proportion of positive viral shedding (p < 0.001) and infectious virus (p = 0.02).

Fig. 4. Inflammatory marker levels among those with and without post-acute sequelae of SARS-CoV-2 infection (PASC) over a 28-day period after symptom onset.

a–g Estimated mean values and 95% confidence intervals are plotted for biological specimen among PASC and non-PASC groups at days 5, 9, 14, 21 and 28 for (a) IL-6, (b) IL-10, (c) TNF-alpha, (d) MCP, (e) IP-10, (f) IFN-alpha, and (g) IFN-gamma, using generalized estimating equations fit with independent correlation, identity linkage, and Gaussian distribution. Covariates included sex, age, vaccination status, and SARS-CoV-2 variant. All statistics are derived from imputed data from 80 participants. Statistical significance was assessed using two-tailed tests. No statistically significant differences in estimated values were found when comparing PASC vs. non-PASC at each time point.