Fig. 4: αCD40 therapy resulted in suppressed cytotoxic T cell responses in preclinical glioma models.

a Quantification of CD3+ T cells in the tumor area of αCD40-treated GL261 glioma-bearing brains that were positive (red square) or negative (red circle) for TLS. n(TLS−) = 7 mice, n(TLS+) = 10 mice. Two-tailed t-test. b, c Kaplan–Meier survival curve of GL261 (n = 19 mice for rIgG2a, n = 20 mice for αCD40 group) and CT-2A (n = 10 mice/group) tumor-bearing mice treated with αCD40 or rIgG2a on days 10, 13, 16, and 19 (as indicated by arrows). Black line: rIgG2a; Red line: αCD40. Log-rank test. d–g HSNE analysis of multicolor flow cytometry data showing spatial clustering of tumor-infiltrating CD3+ cells in GL261 and CT-2A tumor-bearing mice treated with rIgG2a or αCD40. n(rIgG2a) = 5 mice, n(αCD40) = 7 mice. d HSNE analysis identified six meta-clusters (MC) of CD3+ T cells, expressing different levels of CD4 and CD8 (e). f Spatial distribution of CD3+ T cells from GL261 and CT-2A tumor-bearing brains, in rIgG2a-treated vs. αCD40-treated mice. g Frequency distribution of CD3+ T cells from each model and treatment group in each MC. h–n Data was obtained from GL261 and CT-2A tumor-bearing mice treated with rIgG2a or αCD40. h CD8+ T cells as a percentage of CD45+ cells. n(GL261) = 8 mice/group, n(CT-2A) = 5 mice/group. p(GL261) < 0.0001, p(CT-2A) < 0.0001. i–n show data on CD8+ T cells. i CD44+CD62L− cells, n(GL261-rIgG2a)&(CT-2A) = 5 mice/group; n(GL261-αCD40) = 7 mice. p(GL261) = 0.0031, p(CT-2A) < 0.0001. j CD127-KLRG1+ cells, n(GL261-rIgG2a)&(CT-2A) = 5 mice/group, n(GL261 αCD40) = 7 mice. p(GL261) = 0.003, p(CT-2A) = 0.0004. l CD69+ cells, n(GL261 rIgG2a) = 7 mice, n(GL261 αCD40) = 8 mice, n(CT-2A) = 5 mice/group. p(GL261) = 0.0024, p(CT-2A) = 0.0005. m CD107a+ cells, n(GL261) = 6 mice/group; n(CT-2A) = 5 mice/group. p(GL261) = 0.01, p(CT-2A) < 0.0001. n PD1+TIM3+LAG3+ cells n(GL261 rIgG2a) = 5 mice, n(GL261 αCD40) = 7 mice, n(CT-2A) = 5 mice/group. p(GL261) = 0.0462. h–j and l–m Two-tailed t-test. k Heat map showing the mean fluorescence intensity (MFI) of proliferation, exhaustion and memory markers on CD8+ T cells. n(GL261 rIgG2a) = 5 mice, n(GL261 αCD40) = 7 mice, n(CT-2A) = 5 mice/group. o Experimental layout used to obtain data shown in panels (p–r). In brief, GL261 glioma-bearing mice were treated with rIgG2a or αCD40 on days 10, 13, 16, and 19 post-tumor implantation. On day 22 splenocytes were isolated, stained with cell trace violet (CTV) and re-stimulated in vitro with concanavalin A (ConA) for 24 h. p Percentage of CD8+ splenocytes in different generations (G), where cells in G0 did not proliferate and cells in G6 underwent six cycles of proliferation. An example of how generations were defined is shown in Supplementary Fig. 6a. p(G0) = 0.0464, p(G6) = 0.0069. Multiple t-test with Sidak-Bonferroni correction. q Mean fluorescence intensity (MFI) of CD69 on CD8+ splenocytes. p = 0.0248. r CD107a+ T cells as a percentage of CD8+ splenocytes. p = 0.0007. q, r Two-tailed t-test. p–r n = 5 mice/group. For all graphs: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Bars: mean ± SEM. In all graphs except (a): black circle indicates rat IgG2a (rIgG2a), red square indicates agonistic CD40 antibodies (αCD40). Source data are provided as a Source Data file.