12 Feb T cells part 2

Fri, 02/12/2021 - 21:18


Ep 109-12 a: from RECOVERY, we have now evidence that the IL-6 receptor antagonist Tocilizumab (TCZ) has a beneficial effect in hospitalized hypoxic patients (oxygen saturation <92% on air or requiring oxygen therapy) and evidence of systemic  inflammation (C-reactive protein [CRP] ≥75 mg/L):

  • Less patients died on TCZ within 28 days (28 vs 33 %): this beneficial effect was strongest in those who received corticosteroids
  • More TCZ patients discharged within 28 days (54 vs 47 %)
  • Those on TCZ who did not need mechanical ventilation less likely to require ventilation or to die than the controls (33 vs 38 %)

In Ep 109-12b  these data are put into perspective: YES there is an additional benefit to corticosteroids. But tocilizumab is about 100 times more expensive than dexamethasone, raising questions once again about how to make sure populations across the world can bene􀂦t from scientific progress against COVID-19.


T cells and cellular immunity part 2

Ep 109-13: an in vitro study by Nguyen in Melbourne on lung-derived mononuclear cells from donors over a range of age shows interesting immunological features:

  • The only cell subset that decreases with age is CD8 T memory cells
  • Stimulation with Influenza induces a pro-inflammatory response (IL-6, IL-8, TNF…) that remains constant with age, but both type I (IFN-alpha) and type II (IFN-gamma) responses decline at old age.  More in particular, the IFN-g production by memory CD8 T cells decreases also strongly with age.  This imbalance can explain worse outcome of influenza in old people.
  • There was no IFN production after stimulation with SARS-CoV-2, despite clearly productive infection.  Testifying of a “muted” response in SARS-CoV-2 naive tissue    


Ep 109-14: a very interesting paper by Vibholm from Aarhus investigates people who have recovered from COVID, but remain droplet digital PCR positive in nasopharyngeal swab:

  • Frequency: 13 % after 3 weeks and 5 % after 3 months.
  • None of these transmitted virus to > 750 contacts.
  • There was no clear-cut difference in SARS-CoV-2 antibody levels, but neutralizing Ab were not measured
  • There was an increased breath and magnitude of CD8 T cell responses: these were measured in HLA-A2 individuals with “dextramers” (a technique to label epitope specific T cells). 

These data suggest, but do not prove, that CD8 T cells, rather than antibodies, were instrumental in keeping SARS-CoV-2 infection under control and prevent transmission.    


Ep 109-15:  Tan et al in Singapore investigate the dynamics of T cell responses over the course of disease.  They clearly shows that early appearance of IFN-gamma producing T cells is associated with mild/moderate infection, whereas antibody responses (including neut Ab) show similar kinetics in mild and moderate cases.  Early induction of T cell responses against the “accessory” proteins ORF7/8 was particular for mild disease. The main limitation of this study is the small sample size


Ep 109-16: the group of Nussenzweig in NYC previously showed in a follow up of 61 convalescent patients that, despite slightly decreasing neut Ab over 3 to 6 months, the B cell, memory was intact and there were signs of hypermutation in the antibodies, as a result of adaptation to persistent virus in the gut (see Gaebler 2020).  Here they show that

  • There are also persistent changes in the T cell composition: lower CD4 T and higher CD8 T cells, more signs of T cell activation (e.g. CD25 = IL-2R alpha) and proliferation (Ki67) in vivo.
  • Upon, stimulation in vitro with SARS-CoV-2 peptide pools, the CD4 T cells express Th1 cytokines (IL-2, TNF-A, IFN-G) and the CD8 T cells also cytolytic markers (CD107a). 


Ep 109-17: Tarke et al in Lo Jolla analyze the immunodominance of epitopes recognized by CD4 and CD8 T cells from  99 convalescent patients. Both CD4 and CD8 T cells mainly recognize epitopes in non-structural proteins nsp (non-structural protein) 3, nsp4, nsp12, ORF3a, S, M and N.  In addition CD4 T cells have epitopes in nsp13 and ORF8, while CD non-structural proteins io8 T cells have epitopes in nsp6.  Importantly:

  • Each donor will recognize a different set of epitopes, based on HLA type, but the average is 19 epitopes for CD4 T cells and 17 for CD8 T cells.  This number is higher than for most other RNA viruses (in Dengue it is 11 resp 7).  The authors feel that SARS-CoV-2 therefore might have problems to “escape” from so many T cell epitopes by mutations, without losing fitness.
  • T cell epitopes, induced by SARS-CoV-2 infection are 80 % specific and overlap (cross-react) only 20 % with other “common” Coronaviruses.  
  • CD4 T cell and B cell epitopes are mainly in different regions of S and N proteins, whereas M protein has many CD4 but no B cell epitopes. 
  • Very importantly, whereas the receptor binding domain (RBD) of S is very immunodominant for B cells (and neutralizing antibodies), it contains very few T cell epitopes. This implies that an RBD only construct for vaccine might not induce string T cell “help”.

Limitation: descriptive, trans-sectional study with no insight in protective value of these responses (also no info on cytokine pattern).  Nevertheless important to know which epitopes are most prevalent.  The authors will also make “megapools” of relevant epitopes available for the scientific community.


Ep 109-18 Brahim Belaid from Algers investigated potential prognostic markers between severe and non-severe disease as well as survival and death.  The best markers were IL-6 (increase) and CD8 T cell (decrease). Remarkably, comparing T cell cytokine profile after maximal PMA + Ionomycin stimulation, they see no difference in IFN-G (Th1), but an increase in IL-17 (Th17) and a decrease of IL-4 (Th2) in COVID patients as compared to healthy controls.


Ep 109-19:  Nico Janssens in Nijmegen also explores correlates of severe disease in hospitalized patients:

  • In routine plasma they show that increase of IL-6, CRP, D-dimer and ferritin discriminate between non-severe and ICU patients.
  • Using proteomics, ICU patients are characterized by increased IL-6, hepatocyte growth factor (HGF) and CCL-20 as well as decreased stem-cell factor (and also vascular endothelial growth factor and TRAIL).
  • After stimulation of PBMC with Candida albicans, cells from ICU patients were defective in IFN-g, IL-17 and IL-22 production.


Ep 109-20 Ramin Sami et al studied the prognostic differences with 71 patients of whom 60 survived and 11 died within 10 days on the first day of hospitalization:

  • IL1a and IL-6 were increased in non-survivors (but not TNF, IL1b, IL-10 or TGFb)
  • Two subsets of natural killer cells  were increased in non-survivors: CD56dim CD16+ “cytotoxic” cells and CD56bright CD16dim (cytokine producing)
  • Remarkably, there was no difference in Th1 CD4 T cells, but Th2, Th17 and regulatory T cells were decreased in non-survivors
  • As expected, T cell “exhaustion” marker PD-1 was increased in both CD4 and CD8 T cells in non-survivors, but there was a discrepancy with regard to “activated”  (CD25+ CD69+) T cells: higher in non-survivor CD4 T cells, but lower in their CD8 T cells.


Ep 109-21 Neidleman et al. investigated similar T cell characteristics in survivors versus non-survivors in San Francisco, using a CYTOF, which combines over 40 markers. They also performed not just one, but several staining during follow-up of the patients.  It is a very complex paper and I will limit myself to some highlights:

  • Mild, moderate and severe patients CD4 and CD8 T cells produced IFN-g upon SARS-CoV-2 stimulation
  • CD4+ and CD8+ T cells from severe cases expressed significantly higher levels of activation markers HLADR, CD69, CD25, and PD1, as well as chemokine receptor CXCR4, which directs cells to inflamed and damaged lung tissues.
  • Regulatory T cells were similar in mild, moderate and severe cases, but moderate and severe patients had higher follicular helper cells, which may refer to their tendency to produce more SARS-CoV-2 Ab (but of lower quality).
  • Increased expression of PD1, CD95, and TIGIT in severe cases especially in CD8 T cells points to terminal exhaustion.
  • Similarly increased expression levels of the activation markers HLADR and CD38 and exhaustion markers PD1 and CTLA4, and decreased expression of the IL7 receptor alpha chain CD127 in SARS-CoV-2 stimulated T cells from severe cases: over-abundance of exhausted T cells.
  • Non-survivors more frequently failed to mount a robust Spike specific T cell response, particularly for CD4+ T cells.
  • Non-survicors: highly significant increase in IL6-producing SARS-CoV-2-specific CD8+ T cells.
  • While survivors tended to increase the frequency of SARS-CoV-2-specific T cells over time, non-survivors tended to not have much of these cells to begin with, or the frequency of these cells stagnated or decreased.  


Conclusions: The picture of T cell is complex and some data seem contradictory, but there is accumulating evidence that

  • Early, robust SARS-CoV-2 specific CD4 T cell responses are associated with milder disease, whereas profound exhaustion and possibly IL-6 production by CD8 T cells is associated with poor outcome. 
  • T cell immunity persist in the convalescent phase, it may suppress transmission and it could be useful to prevent re-infection.
  • The precise “landscape” of immunogenic parts of SARS-CoV-2 for CD4 and CD8 T cells according to HLA type is being unraveled and that should help to design better next generation T-cell targeted vaccines that could complement the present predominantly B-cell oriented approach, which seems highly S-variant sensitive….