Fig. 3

α-TGFβ and α-PD-1 synergize in eliciting tumor rejection in a CD8a + cell-dependent manner. a Growth curves of each of the six SCC lines, as indicated, after treatment with control IgG or α-TGFβ/α-PD-1 combination therapy on days 0, 4 and 8 (n = 8–10 mice per arm). Tumors were measured every other day after tumor implantation and therapy (see Fig. 1b). Summary of tumor responses for CCK168 (b) and CCK169 (d), and overall survival for CCK168 (c) following treatment on days 0, 4 and 8 with control IgG, α-PD-1, α-TGFβ or combination therapy. In (b) and (d), individual tumors were classified as complete responders (CR), partial responders (PR), or progressive disease (PD), according to their growth characteristics (CR, tumor eradicated with no regrowth; PR, tumor shrinkage ≥30%; PD, no effect of drugs compared to control IgG). For each drug arm, the percentages of total tumors within each response group are shown. n = 10 mice per arm per experiment, n = 3 independent experiments. c Kaplan Meier survival plot for mice bearing CCK168 tumors, after treatment on days 0, 4 and 8 with the indicated drugs. 2000 mm3 tumor size was used as the cut off for survival. e Average CCK168 tumor growth curves following treatment with α-PD-1/α-TGFβ combination therapy versus isotype control IgGs in mice with or without CD8+ T cell depletion. Animals were treated as in Fig. 1b, except that 24 h prior to the first therapeutic drug dose, mice received ip injection of CD8a + cell-depleting antibody or isotype matched control IgG. f) CCK168 tumor-bearing mice were administered one of six IgG combinations, either human α-pan-TGFβ IgG2 (XPA.42.068), human α-TGFβ1/β2 IgG2 (XPA.42.089) or human α-keyhole limpet hemocyanin as control IgG2, together with rat α-PD-1 IgG2 or rat IgG2 control, on days 0, 4 and 8. Average tumor volume (mm³) +/− SD. * = p < 0.05; ** = p < 0.01; *** = p < 0.001: Fisher’s exact test (b), Gehan-Breslow-Wilcoxon test (c), Student’s test (e, f)