Fig 2: The disruptive targeted insertion of HLA-E at the B2M locus does not affect the antitumor activity of ?TRACCAR?B2MHLAE.a, b Antitumor activity of engineered T cells in vitro. T cells edited as indicated and expressing a CAR specific for either CD22 (a) or CD123 (b) were co-cultivated in vitro, with nano-luciferase-expressing RAJI or MOLM13 leukemia cell lines, respectively, using variable effector to target (E/T) ratios. The nano-luciferase signal obtained after co-culture was normalized to the one obtained for tumor cells fully depleted by Zwittergent® to determine the frequency of tumor cells lysis. On each box plot, the central mark indicates the median, the bottom and top edges of the box indicate the interquartile range (IQR), and the whiskers represent the maximum and minimum data point. Each point represents one experiment performed with a given T-cell donor (n = 3 biological replicate for CAR T-cell specific for CD22 and CD123, n = 3 and n = 1 technical replicates for CAR T-cell-specific CD22 and CD123, respectively). p-values are documented in the source file for clarity. c Schematic showing the experimental design to investigate the antitumor activity of ?TRACCAR123?B2MHLAE in the xenograft mouse model. Immunodeficient NSG mice were adoptively transferred with MOLM13-Luc-GFP leukemia cells (2.5 × 105 cells/mouse in 100 µL of PBS via i.v.). On day 7, leukemia-bearing mice were adoptively transferred (i.v.) to randomized mice with mock-transduced T cells (n = 7), ?TRACCAR123T cells (n = 7), ?TRACCAR123?B2M T cells (n = 9), ?TRACCAR123?B2MHLAE T cells (n = 9) or no T cells (n = 8) (7 × 106 viable CAR+ cells/mouse in 100 µL of PBS via i.v. injection). Mice were monitored and followed for overall survival. d Kaplan–Meier plot obtained for different mouse cohorts. Log-rank (Mantel–Cox) tests were used for statistical analysis. P values are indicated on the figures. Source data are provided as a Source Data file.

Fig 3: Targeted insertion of HLA-E at the B2M locus efficiently prevents NK-cell-mediated depletion of ?TRACCAR?B2MHLAE in vitro.a Schematic showing the experimental design to investigate the susceptibility of ?TRACCAR?B2MHLAE cells, engineered under suboptimal conditions, to NK-cell-dependent depletion in vitro. b Representative flow-cytometry plots showing the frequency of HLA-ABC(+)/(-) and HLA-E(+)/(-) subpopulations of ?TRACCAR22?B2M and ?TRACCAR22?B2MHLAE T cells obtained after co-culture. c Box plots representing the ratio of HLA-E(+) to HLA-E(-) computed from the remaining HLA-ABC(-) subpopulation of ?TRACCAR?B2MHLAE T cells cultured with or without NK cells; ?TRACCAR22?B2MHLAE T cells (left panel, T-cell donor (n = 2) and NK-cell donor (n = 1)); ?TRACCAR123?B2MHLAE T cells (right panel, T-cell donor (n = 2) and NK-cell donor (n = 4)). Each point in the box plot represents one experiment performed independently with a given T-cell donor and NK-cell donor. d Schematic showing the experimental design to investigate the susceptibility of ?TRACCAR?B2MHLAE cells, engineered using optimal conditions, to NK-cell-dependent depletion in vitro. e Representative flow-cytometry plots showing the frequency of HLA-ABC(+)/(-) and HLA-E(+)/(-) subpopulations of ?TRACCAR22?B2M and ?TRACCAR22?B2MHLAE T cells obtained after co-culture. f Box plots representing the fraction of HLA-ABC(-) cells of ?TRACCAR22?B2M and ?TRACCAR22?B2MHLAE T cells remaining after co-culture (n = 5 NK donors, n = 1 T-cell donor with two technical duplicates averaged). g Box plot representing the frequencies of NKG2A(+), NKG2C(+) cells detected by flow cytometry in healthy NK-cell donors (n = 17). h Box plot representing the ratio of NKG2A(+)/NKG2C(+) cell frequencies detected by flow cytometry in healthy donors (n = 17). NKG2A>C and NKG2C>A donors are shown in the light-blue and dark-blue fields, respectively. In each box plot, the central mark indicates the median, the bottom and top edges of the box indicate the interquartile range (IQR), and the whiskers represent the maximum and minimum data point. Two-tailed paired t test with a confidence interval of 95% was used to compute statistics for the dataset illustrated in (c, f, g). P values are indicated in the figures. Source data are provided as a Source Data file.

Fig 4: Targeted insertion of HLA-E at the B2M locus enables the efficient and prolonged antitumor activity of ?TRACCAR?B2MHLAE in the presence of activated NK cells in vitro.a Schematic showing the serial killing assay designed to investigate the long-term antitumor activity of ?TRACCAR22?B2MHLAE cells toward RAJI-Luc cells in the presence or absence of activated NK cells. The three different serial killing assay scenarios investigated are illustrated (no NK, NK day 0, and NK day 1). Black, blue, and red arrows indicate addition of RAJI cells, CAR T cells (?TRACCAR22, ?TRACCAR22?B2M and ?TRACCAR22?B2MHLAE, n = 1 donor) and NK cells (n = 1 donor), respectively. Luminescence and flow-cytometry analysis of cell populations are indicated at the different measurement time points. b Frequency of RAJI-Luc cells killing observed by luminescence with the three different serial killing scenarios performed with a CAR T cell to RAJI ratio of 2.5:1 and NK to CAR T-cell ratio of 1:1. Each point represents one experiment performed with a given T-cell donor and a given NK donor. c Representative flow-cytometry plots showing the different cell populations remaining at the end of the serial killing assay in absence of NK cell (left panel) and in the presence of NK cells added at day 0 and at day 1 (middle and right panel, respectively). The gating strategy is indicated by blue boxes. d Absolute CAR T cells and RAJI cells counts obtained by flow-cytometry analysis and cell counts in the same condition as in (b). Source data are provided as a Source Data file.

Fig 5: TALEN-mediated multiplex editing enables efficient CAR and HLA-E expression in TCRaß/B2M double knockout T cells.a Schematic showing the editing strategy at the TRAC and B2M loci to generate ?TRACCAR?B2MHLAE T cells from wild-type T cells. b Experimental design for multiplex editing of TRAC and B2M loci using TALEN and adeno-associated viral (AAV6) particles and analysis of the resulting engineered T cells. c Representative flow-cytometry analysis of mock-transfected T cells and T cells engineered with different combinations of TALEN and AAV6 particles (CAR123m and HLA-Em) c, top panel, flow-cytometry plots showing the frequency of TCRaß and HLA-ABC expression in viable engineered T cells. c bottom panel, flow-cytometry plots showing the frequency of CAR and HLA-E expression detected among TCRaß(-)/HLA-ABC(-) viable engineered T cells. The black bold box depicts the parent population. d Box plot showing the frequency of different subpopulations detected in engineered ?TRACCAR22?B2MHLAE T cells (left) and ?TRACCAR123?B2MHLAE T cells (right). The gating strategy used to obtain the frequency of each subpopulation is illustrated by the black, red, green, and blue bold boxes in (c), right flow-cytometry panel). In each box plot, the central mark indicates the median, the bottom and top edges of the box indicate the interquartile range (IQR), and the whiskers represent the maximum and minimum data point. Each point represents one experiment performed with a given donor (n = 6 for ?TRACCAR22?B2MHLAE and n = 7 for ?TRACCAR123?B2MHLAE). Source data are provided as a Source Data file.