Extended Data Figure 2: Characterization of and examples for CD4⁺CD8⁺ and wild-type cross-reactive T cell responses induced by neo-epitope vaccination.

a, Pre- and post-vaccination T cell frequencies against neo-epitope-encoding RNAs (nine patients). In vitro stimulation of T cells as in Fig. 1f, except autologous DCs for read-out were loaded with RNA encoding single neo-epitopes instead of OLPs. 88% (110 out of 125) concordance rate of both measurements, with slightly higher sensitivity of the OLP read-out. b, c, Pre- and post-vaccination CD4⁺ and CD8⁺ T-cell-enriched cultures of patient P19 stimulated with the patient’s pentatope RNAs read-out against autologous DCs loaded with OLPs representing mutated neo-epitopes in the ST5 and UTP6 proteins. b, Example for an exclusive de novo CD4⁺ T cell response. c, Example for a de novo response recognized by both CD4⁺ and CD8⁺ T cells at different portions of the mutated sequence. d, CD4⁺ and CD8⁺ T cell cultures were quality controlled for purity after stimulation by flow cytometry to exclude cross contamination of the separated in vitro stimulation cultures. e, Wild-type epitope cross-reactivity of post-vaccine CD4⁺ (top left) and CD8⁺ (top right) T cells (tested for 55 neo-epitopes in total). Autologous DCs loaded with RNA (not marked), with different OLPs (P1, P2) of the OLP pool encoding the mutated or the wild-type sequence used as read out (marked with ‘P’). Examples of reactivity with mutated and wild-type epitopes analysed by ELISpot is shown for P04 CDC37L1(P186L). f–h, Three examples of wild-type epitope cross-reactive responses. For all three responses immune recognition of autologous DCs that naturally express the wild-type genes was not detected. g, Sanger sequencing of P05 FAM135B in control DCs used in ELISpot (f). Control RNA, luciferase. h, Left, example for wild-type reactivity observed with encoding RNA only but not with OLPs. Right, differential cross-reactivity depended on the assessed OLP in a single T cell response.