Response to Daungsupawong/Wiwanitkit LTE

To the Editor:

We are glad to provide more information about the rationale of our work published online by Viral Immunology on October 30, as requested by Daungsupawong and Wiwanitkit.

We agree with the observations the authors made in their letter, some of them highlighted at the end of the Discussion section. All the procedures we utilized in our work to identify memory subsets and Spike-restricted T cell clones were performed using a well-standardized, close-to-clinical diagnostic protocol (see, e.g., Landay and Muirhead, 1989). In other words, samples were processed using a typical “lyse no wash” procedure, which minimizes sample manipulation. We chose this approach for several reasons: (1) to avoid the loss of spike-specific T cell clones when blood samples are subjected to centrifugation and/or Ficoll separation since we expected these clones are rare in circulation; (2) to avoid any preferential expansion of Spike-restricted immunodominant T cell clones following in vitro stimulation, using overlapping peptide library as in Grifoni et al. (2020); and (3) to set up a cytometric protocol easily extended to any diagnostic laboratory equipped with an “entry-level” flow cytometer.

Ficoll separation and in vitro stimulation/expansion of T cell are certainly preferable when the aim is to functionally characterize the complexity of immunological response to a specific antigen and/or to perform T cell receptor (TCR) sequencing. On the other hand, our main aim was, instead, to capture an instant picture of the T cell compartment using the simplest and less handled sample directly starting from a blood withdrawal. Both approaches (i.e., direct staining vs. peripheral blood mononuclear cells, PBMCs, isolation and in vitro stimulation/expansion) have advantages and limitations. We dealt with them, and we chose to perform a study using a direct staining procedure, taking into account its limits.

Regarding the criteria for selecting the immunodominant epitopes, we resorted to the already available tetramers/pentamers complexes provided by Proimmune and we checked these peptides using the SYFPEITHI algorithm (www.syfpeithi.de). We verified that these complexed peptides had the highest score indicated by the manufacturer and that they were conserved in all the SARS-CoV-2 variants.

Regarding the possibility to perform a deeper longitudinal study, we dealt with the problem to recall the subjects belonging to the study cohort after the initial recruitment carried out in the days immediately following the hospital discharge or at follow-up visit. We did not have the possibility to retrieve data on viral diversity and many subjects also refused to undergo booster vaccination, so it was not possible for us to collect data statistically robust enough to draw a meaningful conclusion on the durability and efficacy of memory.

We are pleased that our work inspired a stimulating scientific discussion, and we would like to thank Hinpetch Daungsupawong and Viroj Wiwanitkit for giving us the opportunity to give more details of our work. We hope that our observation could pave the way for other groups to deepen all the aspects related to the T cell responses and to the generation and maintenance of the T cell memory compartment following SARS-CoV-2 infection.