Post

1) Researchers discovered a subgroup of Long Covid patients with more activated killer T cells targeting SARS-CoV-2 spike and proteins from the Epstein–Barr virus (EBV). A brief breakdown of the main findings 👇

2) This was a big study of 134 Long Covid patients recruited at three different clinical sites. The authors also included 54 recovered controls and tested their measurements in 49 patients followed after acute Sars-Cov-2 infection and 51 people post-vaccination.

3) A first finding: the overall T cell landscape was not substantially altered in long COVID. Besides a small increase in the CD39 marker, there were no significant changes in overall T-cell populations in circulation. The difference lay in a very tiny subgroup of T-cells.

4) The study used special proteins (HLA-I and HLA-II tetramers) to isolate specific T cells without altering or stimulating them. This way, they could isolate T-cells that bind to the spike of Sars-cov-2 or other virus particles.

5) They found that Long Covid patients have more activated CD8 T-cells that target Sars-Cov-2 spike than controls. They were characterized by several activation markers (PD-1, CD38, CD39, Ki-67, HLA-DR).

6) The authors tested their assay in longitudinal sample sets collected before, during and after SARSCoV-2 mRNA booster vaccination or breakthrough infection. Spike-specific CD8 T cells were activated but returned to baseline status after approximately 90 days.

7) In a subgroup of LC patients, this wasn’t the case and the measurement remained stable over time. The difference with controls also remained even after controlling for factors such as age, sex, days since last vaccine/infection, number of Sars-Cov-2 infections, etc.

8) The authors used 1 standard deviation from the mean of recovered controls. Using that (not so strict) threshold, 40% of LC samples were activated compared to 16% in controls.

9) Despite a clear difference, the majority of LC patients didn’t have this T-cell abnormality, while many of the recovered and healthy controls had it as well.

10) Next, the authors tested the same for other viruses but found no increase in activated T-cells targeting the flu, cytomegalovirus (CMV), varicella zoster virus (VZV) or Epstein-Barr Virus (EBV). The effect was quite specific.

11) They did find an increase (22% in LC versus 15% in controls) in T-cells targeting lytic antigens of EBV. These are proteins the virus makes when it is active and replicating. Reactivation of EBV has been frequently reported in LC and this is supported by these findings.

12) Next, the authors did several tests on the activated CD8 T-cells and they found that they have: - some markers of T-cell exhaustion - a decreased ability to produce cytokines such as TNF and IFN-γ - they divided less after stimulation

13) The authors argue the findings are consistent with viral persistence in the gut. They say it points to “the existence of a persistent tissue reservoir of SARS-CoV-2 that can be accessed by a fraction of the circulating CD8 T cells at any given time.”

14) One limitation is that the study doesn’t prove that these T-cells are pathogenic and not just a harmless remnant of past infection. They don’t for example show a correlation between having these cells and symptom severity in Long Covid.

15) Non-spike SARS-CoV-2 responses also showed no effect. Another concern is that the findings could reflect batch effects and shifts in experimental conditions. I noticed for example that being in the LIINC cohort had a bigger effect than having LC or not.

16) Another caveat: the data reported are LC samples and some samples come from the same patient. The paper doesn’t clarify how this dependence was taken care of in the statistics (but I assume it wasn’t overlooked given the many researchers involved).