Thread by @ZdenekVrozina

Long COVID may not begin only after the acute infection has passed. In at least some patients, the immune system appears to go off track from the very start. A breakthrough study.🧵

What makes this study especially interesting is that it does not just describe what patients with long COVID look like later on. Instead, it looks for differences already during acute hospitalization and then again three months later. That is its main strength - it tries to capture whether a risk immune signature is already present early.

The authors used mass cytometry with 40 markers and compared 10 patients who later developed long COVID with 13 similarly ill patients who did not.

The standard analysis of the main immune cell populations found almost no clear differences between the groups. If you only look at how many of each immune cell type are present, long COVID is not predicted very well. The differences only became visible when they used analysis of cell phenotype, including markers of migration, activation, functional state.

During the acute phase of infection, the early abnormalities in patients who later developed long COVID were seen mainly in innate immunity. The authors especially highlight cDC2 dendritic cells and gamma delta T cells.

In cDC2 cells, they found increased CCR5 expression together with decreased CD56 and CD45RA, which they interpret as a sign of altered migration and possibly weaker early antiviral effector function. In gamma delta T cells, lower CD43 expression may point to impaired migration and activation within inflamed tissue.

At the same time, lower HLA-DR signaling was seen in naive CD8 T cells in the long COVID branch, which the authors link to weaker activation of cytotoxic responses.

After three months, the picture shifts more toward adaptive immunity.
In long COVID patients, naive CD4+ and CD8+ T cells showed a persistent phenotype marked by lower CD62L and higher CXCR3. That is quite interesting, because CD62L is associated with homing back to lymph nodes and normal immune surveillance, whereas CXCR3 is more of a marker for trafficking into inflamed tissues.

The authors interpret this as meaning that these T cells may be reprogrammed more toward movement into inflamed peripheral sites rather than normal circulation and priming. In addition, naive CD4+ T cells had lower IL-7R/CD127, which may suggest impaired T-cell homeostasis and survival.
Central memory CD4+ T cells also showed lower CD161 expression.

Taken together - the study supports a model in which long COVID does not begin only after the infection. Instead, in at least some patients, an atypical immune response is already developing during the acute phase.

First, innate immunity and the interface between innate and adaptive immunity are affected. Later, a more persistent abnormal T-cell state remains. That fits quite well with the idea that long COVID may partly result from an immune response that does not fully resolve properly.

These were patients from the first wave of the pandemic, so the data are not complicated by vaccination or by later viral variants.

Perez-Mazzali at al., Persistent T cell phenotypic alterations and early innate immune dysregulation as potential biomarkers of Long COVID. journalofinfection.com/article/S0163-…

Gamma delta T cells are especially interesting here. They are early innate-like effectors involved not only in the response to SARS-CoV-2, but also in the control of herpesviruses, including EBV.

Reviews of EBV immunity describe gamma delta T cells as contributing to the control of EBV infection, cellular transformation, and lytic replication.

In this study, the authors found an altered gamma delta T-cell phenotype already during the acute phase in patients who later developed long COVID.

At the same time, they emphasize early dysregulation of dendritic cells and gamma delta T cells as a possible predisposing factor for persistent symptoms. In the highlights, they essentially argue that early immune imbalance may predispose patients to persistent post-viral pathology.

If gamma delta T cells are functionally off balance, that could mean weaker early control of latent or reactivating viruses, and EBV is one of the most plausible candidates. This would fit a model in which part of LC is not just a direct consequence of SARS-CoV-2 itself, but also a downstream consequence of impaired control over latent infections.

Reviews on EBV do indeed suggest that weakened cellular immunity is linked to poorer EBV control, and gamma delta T cells are part of that defense.

The Shahbaz et al. study pushes this interpretation further. It suggests that in long COVID with an ME/CFS phenotype, the problem is not just a few altered markers, but broader immune remodeling - changes in monocytes and NK cells, signs of T-cell exhaustion, and depletion of gamma delta T cells and MAIT cells, which the authors associate with the Galectin-9/TIM-3 axis. x.com/zdenekvrozina/…

That matters, because this no longer looks like nonspecific inflammation. It looks more like a state in which several layers of immune dysfunction are combined - altered innate immunity, weaker cytotoxic control, more exhausted T cells, and loss of cell populations that normally help monitor infection at the interface of innate and adaptive immunity.

This kind of immune environment could plausibly mean worse control over latent viruses or more frequent episodic reactivation which fits well with a model of immune suppression or dysregulation.

These findings together support a model in which, in some patients, acute infection disrupts immune regulation, and that disturbance then evolves into a longer-lasting state with weaker cellular control, especially at the level of cytotoxic and innate-like populations.

In that kind of setting, the hypothesis of EBV reactivation - or simply poorer EBV control - is very biologically plausible.