Generation of HLA KO primary human T cells with reduced allogeneic recognition. (A) Flow-cytometric analysis of β2m^– T cells (gated on living lymphocytes) from B2M gRNA edited (left) and unedited (right) PBMCs of a healthy donor. (B) Discordance plot as generated by ICE analysis of a β2m-edited sample (blue) in comparison with mock-edited cells (gray, control). Discordance refers to the extent of disagreement between the wild type and edited sample at each base within a defined inference window (orange line). (C) Percentage of successful HLA class I KO as detected by flow cytometry (FACS) and in ICE analysis after sequencing (DNA) (n = 6 technical replicates). (D) Percentage of CD69⁺CD137⁺ CD8⁺ effector cells after 48 hours of coculture with indicated target cells (n = 3 technical replicates). Effector cells were peripheral mononuclear blood cells from donor A, cocultured for 7 days together with PBMCs from donor B (allo priming). Target cells were autologous unedited cells from donor A (HLA I match), allogeneic B2M KO cells (no HLA I), and allogeneic unedited cells (HLA I mismatch) from donor B. Target cells were sorted for CD3⁺ and successful KO, if applicable. Statistical testing by ordinary two-way ANOVA and Tukey's multiple comparison test, mean with SD; ns, not significant; ∗∗∗∗ p < .0001.