Long COVID is associated with lower percentages of mature, cytotoxic NK cell phenotypes, 2024, Tsao et al.

[SNT Gatchaman]

Long COVID is associated with lower percentages of mature, cytotoxic NK cell phenotypes
Tasha Tsao; Amanda M. Buck; Lilian Grimbert; Brian H. LaFranchi; Belen Altamirano Poblano; Emily A. Fehrman; Thomas Dalhuisen; Priscilla Y. Hsue; J. Daniel Kelly; Jeffrey N. Martin; Steven G. Deeks; Peter W. Hunt; Michael J. Peluso; Oscar A. Aguilar; Timothy J. Henrich

No abstract.

We performed high-dimensional NK cell phenotyping on PBMCs collected 4 months following SARS-CoV-2 infection from people with LC (N=21) and those who had prior COVID-19 but fully recovered (N=14) in the UCSF LIINC cohort. Samples were collected in the first year of the pandemic prior to the emergence of vaccination, antiviral treatment, or frequent reinfection, and most participants (74%) had mild/moderate outpatient COVID-19

mature, cytotoxic CD56dim/CD16+ NK cell percentages were significantly lower in participants with LC compared to those who had fully recovered (P=0.037); a more pronounced significant reduction was observed in those with severe LC (P=0.036).

Link | PDF (The Journal of Clinical Investigation) [Open Access]

 

[Yann04]

This matches previous findings in ME/CFS of lower than normal NK cell cytotoxicity.

Also random but it would be a cool feature to be able to add tags to someone else’s post. In this case would it be possible to add “cytotoxic” as a tag? cheers.

 

[forestglip]

So this is about how many cytotoxic NK cells there are and the recent "Meta-analysis of Natural Killer (NK) cell cytotoxicity" was about average cytotoxicity of NK cells, right? Not necessarily related?

 

[SNT Gatchaman]

would it be possible to add “cytotoxic” as a tag?

Done.

 

[Yann04]

So this is about how many cytotoxic NK cells there are and the recent "Meta-analysis of Natural Killer (NK) cell cytotoxicity" was about average cytotoxicity of NK cells, right? Not necessarily related?

That’s possible, I’m not familiar with how “cytotoxicity” is measured, is it in relation to a single cell (or an average thereof)? If yes, then you are correct.

Though I would guess a lower number of cytotoxic NK cells would correlate with an average of lower cytotoxicity among NK cells.

 

[forestglip]

That’s possible, I’m not familiar with how “cytotoxicity” is measured, is it in relation to a single cell (or an average thereof)? If yes, then you are correct.

Not really sure, hard to understand their description of measuring cytotoxicity in the meta analysis.

Though I would guess a lower number of cytotoxic NK cells would correlate with an average of lower cytotoxicity among NK cells.

Why would you think that? I can see the possibility that with fewer NK cells, cytotoxicity of each one might be upregulated to compensate.

 

[Yann04]

Why would you think that? I can see the possibility that with fewer NK cells, cytotoxicity of each one might be upregulated to compensate.

That’s a good point!

I’m basing this on no immune knowledge at all, but I was under the impression that cytotoxic NK cells were the ones who mainly produced cytotoxicity as opposed to other types, but yes I had not considered other types may still produce it or adapt. The body is a lot more complex than the models brainfoggy yann can come up with.

 

[Hutan]

LC was defined as the presence of COVID-attributed symptoms at least 3 months following the initial infection, severe LC was defined as the presence of >5 symptoms, and all participants were confirmed to have consistent presence or absence of LC symptoms for up to 8 months.

An odd definition of severity, not related to impact on function, only number of symptoms.

Age was similar across groups (Fig 1A), but a larger proportion of people with LC were female (Fig 1B)

It doesn't sound as though the control group was well matched on sex. It's possible that what is reported here is just the result of sex differences.
Figure 1B is difficult to interpret, but I think the presence of any LC symptom qualifies a person to be in the LC group, so around 47% of the combined male participants had LC and 72% of the combined female participants had LC. That suggests that the majority of controls were men, while the majority of the LC group were female.

 

[forestglip]

It doesn't sound as though the control group was well matched on sex. It's possible that what is reported here is just the result of sex differences.

Given the percentages of CD56dim /CD16+ NK cells were significantly lower in females than males (Fig 1J), we repeated comparisons of NK cell phenotypes including only females and observed a significantly lower percentage of CD56 dim /CD16+ (47.1% vs 63.5%) and CD56dim /KIR2D(L1/S1/S3/S5) + (16.2% vs 29.2%) NK cell populations in those with severe LC versus those who fully recovered (P=0.045 and 0.011, respectively; Figs 1K, S1). These data suggest that the differences observed implicate underlying factors between NK cell phenotypes and LC in addition to potential sex differences.

They found this correlation too, though sex may play a role:

we observed a significant negative correlation between the percentage of these cells and the number of reported LC symptoms (Spearman r=-0.5, P=0.022, Fig 1H).

Tim Henrich also discussed this finding at the PolyBio symposium.

 

[Hutan]

Thanks @forestglip. Chart 1k confirms what I had calculated: 9/14 fully recovered people were men (64%); 8/21 LC were men (38%).

The finding is a bit interesting, but the sample sizes are too small and the definition of LC too loose to say much. Only 5 fully recovered women, and the percentages of mature NK cells as shown in Figure 1k overlap a lot.