1/ I'm going to try to contain myself, but very important/elegant article out tonight in Cell with potentially critical implications for immunity as well as a possible explanation for differential severity in presentation of COVID-19 @AaronRichterman cell.com/cell/fulltext/…
2/ Multi-institution group used a bioinformatic approach to predict SARS-CoV-2 peptides that would be likely T-cell targets. This approach is well established and validated and predicts 50% of T-cell response in prior models
3/ They recruited 20 patients who recovered from COVID-19 (all IgG positive) and collected blood samples 20-35 days after symptom onset when all patients were asymptomatic. 
4/ They stimulated PBMCs in 10 of these recovered patients & identified SARS-CoV-2 reactive T-cells, with around 50% of cells reactive to Spike protein and 50% reactive to other epitopes. 100% of recovered patients had CD4 reactive T-cells and 70% had CD8 reactive T cells
5/ Next they found that spike-specific reactive T-cell responses correlated with the titer of anti-Spike IgG (which makes sense since most protective antibody responses are dependent on CD4 T cell help)
6/ They then looked at responses in 11 COVID-negative donors (whose samples had been collected from 2015-2018 (excluding the possibility of COVID). They found around 50% had reactive T-cells to SARS-CoV-2 proteins (more likely to be non-Spike epitopes)
7/ Next they determined the relative frequency of specific CD4+ T-cell responses for various antigens and they found that the majority of T-cell responses were against the most highly expressed viral proteins: more antigen, more presentation, more reactive T cells.
8/ For instance they note: "spike, M, and N. On average, these antigens accounted for 27%, 21% and 11% of the total CD4+ T cell response, respectively."
9/ So eloquent: "In sum, these results, fully scanning the SARS2 orfeome, demonstrate a pattern of robust and diverse SARS-CoV-2−specific CD4+ T cell reactivity in convalescing COVID-19 cases that correlated largely with predicted viral protein abundance in infected cells"
10/ 6/11 unexposed donors had CD4+ reactive T-cells to SARS-CoV-2 proteins with a different hierarchy compared with those recovering from COVID-19. 23% had some reactivity to S protein, but there was minimal reactivity to M or N proteins, all were IgG+ to common coronaviruses
11/ In their discussion they note that a COVID-19 vaccine consisting of Spike protein would likely be able to elicit spike CD4+ T-cell responses but that adding other antigens such as M and N would better mimic responses in convalescing patients
12/ They note that Spike was not a dominant target of CD8+ T cell response and they suggest some additional epitopes that might help elicit a stronger CD8+ T cell response as seen in their recovered patients
13/ They note that it is not clear if the cross immunity they measured from other coronavirus infections leads to milder COVID-19 courses, but there is mouse data that showed that cross reactive CD-4 T cells were protective in a SARS model
14/ They do note that their sample focused on mild/moderate non-hospitalized patients and so there are some limitations to generalizing their data
15/ But this is a major study characterizing T-cell responses after infection and corroborating what others have reported in terms of some baseline cross reactive T-cells. Protection from cross immunity might be the explanation for the wide illness severity spectrum
16/ And it seems to me this is great news in terms of providing the theoretical basis for a vaccine. How cool is science, @AaronRichterman ??
17/ We tweeted about this preprint a few weeks ago which showed similar results @AaronRichterman :
