Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.

Authors:Anna E. Oja et al.  

Journal/ Pre-Print:bioRxiv 

Tags: Clinical, Immunology/Immunity 

Research Highlights

  1. ICU-admitted COVID-19 patients exhibit an overall decreased frequency of CD3+ cells in circulation but with an increase in T follicular helper frequencies relative to unexposed, mild, and severe cohorts  

  1. SARS-CoV-2-specific CD4+ T cells conformed to a type 1 cytokine profile, where ICU patients produced less IFN-y, IL21 and IL4 compared to the mild cohort.  

  1. S1-specific CD4+ T cell responses correlated with anti-SRBD IgG titres in mild patients only.   


The paper investigates the T cell and humoral response within PBMC of 56 Covid-19 patients with varying severities relative to unexposed controls.  The authors found that ICU patients have diminished frequencies of SARS-CoV-2-specific CD4+ T cells.  Overall there was minimal correlation between severity and cytokine profile of the cells, albeit less potent in ICU patients.  Bronchoalvelor lavage fluid (BALF) of ICU patients contained effector-memory (EMRA) and naïve T cells, but few tissue-resident T cells.   

They found that severe patients, which do not require ICU treatment, exhibit higher SARS-CoV-2-specific T cell responses alongside high titers for S-RBD and N IgG.  Despite a poor T cell response in critical ICU patients, they generated a strong antibody response already early on.  

Impact for SARS-CoV2/COVID19 research efforts  

Understand the immune response to SARS-CoV2/COVID19: the paper expands our knowledge of the T cell composition in peripheral blood and BALF of Covid-19 patients with variable disease severity.   

Study Type  

  • In vitro study 

  • Patient Case study 

Strengths and limitations of the paper 

Novelty: There are increasing reports looking at T-cell phenotypes in COVID patients and in particular CD4 responses. The novelty in this study is that it is the first we are aware of that makes a direct link between the T and B cell responses rather than a correlative/associative link.  

Standing in the field:Interesting to see the results regarding the Tfh in circulation especially in light of the paper by Meckiff et al (Biorxiv2020.06.12.148916v1), using scRNA seq and identifying cytotoxic Tfh 

Appropriate statistics: The statistics used are described appropriately but they are not used properly. In some graphs they don’t seem to perform proper multi-comparison testing between all the groups. T-tests in the supplementary not properly explained. Number of patients per group was also comparable.  

Viral model used:SARS-CoV-2 peptide pools 

Translatability:Limited; no direct translational implications 

Main limitations:  

  • The correlation between SARS-CoV-2-specific T cell response and antibody titres was only seen for mild cases, requiring further follow-up for the severe and ICU cohorts.  Furthermore, the authors do not look specifically at Tfh correlations with antibody response.  The variability in time points at collection between patients make these comparisons difficult.   

  • cTfh were identified as CXCR5+PD1+, it would have been useful to include ICOSCXCR3 and CCR6 markers as well to better understand cTfh subsets, especially when linking to B cell help.  Different subsets of human cTfh cells have been distinguished in the literature: Th1-like (CXCR3+CCR6–), Th2-like (CXCR3-CCR6–), and Th17-like (CXCR3–CCR6+) cTfh cells, all of which have altering cytokine production and ability to provide B cell help.  

  • Humoral analysis limited by only analysing IgG responses, and not IgA (which has also been shown to be a dominant isotype during Covid-19) and IgM (ratios would give a better idea of relative isotype switching & B cell response – ie. extrafollicular response).  They miss out on an opportunity to characterise memory B cell or plasma cells.  The authors could have strengthened the paper by more robustly characterising both the cTfh and B cell populations.   

  • Time points for blood collection are different for different groups, this can contribute to some of the differences observed. Ideally one would like at least a sample from the peak of the response and one after recovery for all patient groups 

  • Control group missing for BALF experiments, also – inclusion of total numbers would be helpful to get a sense of the scale in naïve T cell frequency.  

  • Bar graphs difficult to read (figure 1c, d) as well as FACS plots labels (Figure 3a). Some unclear population definitions (e.g Tregs as CD137+, effector cells as CD27-CD45RA+)