Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

Authors: Ferretti et al. 

Journal/ Pre-Print:Medrxirv 

Tags: Bioinformatics, Clinical, Immunology/Immunity 

Research Highlights  

  1. Mapping SARS-CoV-2 T cell epitopes presented by six of the most common Class I HLA types reveals that memory CD8+ T cells from recovered COVID-19 patient target a limited set of shared immunodominant epitopes (3-8 for each HLA type).  

  1. These epitopes are not on high variable regions and are largely unique to SARS-CoV-2. Only 10% of those correspond to spike protein (only 3 of the 29 epitopes). This highlights the need to design vaccines that mimic natural CD8+ T cell responses and tempers enthusiasm related to potential cross-reactivity with common cold coronaviruses 

  1. Single-cell sequencing revealed that many different T cell clones target the same epitope, although it is the same TCR Va region that targets these epitopes. 


Ferretti et al. performed an unbiased and detailed genome-wide screen (T-Scan) on memory CD8+ T cells against SARS-CoV-2 epitopes from 25 convalescent COVID-19 patients, focusing on epitopes presented by the most commonly expressed Class I HLASARS-CoV-2 CD8+ T cell memory establishment appears to be dominated by few antigenic epitopes that are shared among the same HLA typeMoreover, analysis of one specific HLA type (A*02:01) confirmed the presence of memory CD8+ T cells and revealed that many different T clonotypes target the same epitope. Specific fragments were recurrently recognised by the T cells of multiple patients that shared the same dominant TCR Va geneThe epitopes were compared to SARS-CoV-1 and common cold hCoV epitopes but found to be largely unique to SARS-CoV-2 and not on high variable regions. As many other studies have recently shown, only a small proportion (10%) of these epitopes correspond to the spike protein, highlighting the importance of the immune response against other parts of the virus such ORF1ab and N protein. 

Impact for SARS-CoV2/COVID19 research efforts  

Understand the precise targets of the CD8+ cellular immune response establishing immune memory and epitope specificity and frequency to SARS-CoV2/COVID19 

Study Type  

  • Bioinformatics study 

  • Clinical Cohort study  

Strengths and limitations of the paper 


Comprehensive mapping of the COVID-19 convalescent CD8+T memory response by prevalent HLA types. Shows an oligoclonal response- with the same Va fragmentthat largely targets a limited set of shared immunodominant epitopes.  

Standing in the field:  

Implications for vaccine development. In line with current literature for analysis of convalescent patients regarding immunodominance hierarchy (Griffoni et al. or Le Bert et al); and in contrast with in silico studies predicting epitopes presented by HLA alleles and other studies showing cross-reactivity to hCoV under different experimental setups  

Appropriate statistics: The data are well presented and mostly descriptive but there are no statistical tests for significance.  

Viral model used: Clinical samples of recovered COVID-19 patients 

Translatability: This study does not aim for translatability per se but provides important information to help direct vaccine strategiesMoreover, there may be the potential for testing the presence of SARS-CoV2 specific memory CD8+ T cell IFNg secretion after ex vivo stimulation to identify previous infection. 

Main limitations:  

  • While the study suggests that these specific epitopes are result in CD8+ T cell memory, this does not necessarily mean that they were the efficient effector clones during the immune response. It would be interesting to look at affinity for the recognised epitope. 

  • Statistics are not clearly shown