929 – Age-Dichotomous Associations of Long COVID With Frequency of Public SARS-CoV-2 Memory TCRs, 2025, Akbas et al

929 – Age-Dichotomous Associations of Long COVID With Frequency of Public SARS-CoV-2 Memory TCRs

Selin Akbas, Shuai Li, Hao Zhang, Tamilore Adeagbo, Ni Zhao, Alan Landay, Michael Peluso, Elizabeth Thompson, Andrea Cox, Yukari Manabe, Hongkai Ji, Annukka Antar

[Line breaks added]

Background
Despite its prevalence, the pathobiological mechanisms of long COVID (LC) remain unknown. Persistent SARS-CoV-2 viral antigen may cause LC, and so we sought indirect evidence of persistent viral antigen by assessing for frequency of memory T cells specific for SARS-CoV-2 after SARS-CoV-2 infection in people with and without LC.

Methods
We obtained 122 T cell receptor (TCR) β chain repertoires from sorted memory CD4+ and CD8+ T cells from longitudinal samples drawn at 1- and 4-6-months post-first SARS-CoV-2 infection from people who went on to develop LC (n=17) and those who experienced a full and rapid recovery (n=14, FR).

ImmuneCODE-MIRA-Release002.1, a database of 160,000 public SARS-CoV-2-specific TCRβ sequences with linked antigen specificity, was used to determine the frequencies of public SARS-CoV-2-specific TCRs within memory CD4+ and CD8+ TCRβ repertoires. The frequencies of resting naïve and memory CD4+ & CD8+ T cells, activated CD4+ & CD8+ T cells, and monocyte subsets were also determined using flow cytometry.

The sample was stratified into ‘young’ and ‘old’ groups based on age greater than or less than the median age (50) at time of infection.

Results
The frequency of public SARS-CoV-2-responsive TCRs in memory CD4+ (p=0.54) and memory CD8+ (p=0.83) TCRβ repertoires did not differ between individuals with LC and FR at either 1- or 4-6-months post-infection, nor did they significantly change over time from 1 to 4-6 months post-infection.

Immune cell frequencies did not differ, except that LC had lower frequencies of CD4s among all T cells at 4-6-months post-infection compared to FR (58.5% vs 66.6%, p=0.038).

When stratified by age, older LC had significantly lower frequency of public SARS-CoV-2-responsive TCRs in memory CD4+ TCRβ repertoires than older FR (p=0.016), including public non-spike-(p=0.017) and spike-(p=0.04) specific TCRs compared to FR.

Conversely, younger LC trended towards having a higher frequency of public SARS-CoV-2 TCRs among their memory CD4+ TCRβ repertoires than younger individuals with FR (p=0.088); this trend persisted when exclusively considering overlap with public non-spike, ORF7b, and ORF8-associated TCRs (p=0.055, 0.057, 0.063).

Conclusions
Older individuals with LC had fewer public SARS-CoV-2-specific TCRs in their memory CD4+ T cell repertoires than their peers with full recovery. While these results warrant confirmation by other means, this work suggests that LC may not be associated with persistent antigen in older individuals.

Link (Conference on Retroviruses and Opportunistic Infections) [Abstract Only]