Fig. 6: COX-2 inhibition is essential for tumor control during chemotherapy and immunotherapy combination treatment.

a Mice were inoculated subcutaneously with 4T1 tumor cells and treatment began on day 5-6 post-injection when tumor volumes were 94.0 ± 4.6 mm³ (mean ± SEM). b Waterfall plot showing percent change in tumor size two weeks post-treatment start, each bar represents one mouse (n = 6 (ICB + CXB), 7 (CTX + CXB), 16 (CTX + ICB), 18 (control and CTX) or 27 (CTX + ICB + CXB) per group, pool of four independent experiments). c Growth profiles of tumors in mice treated with control (n = 10), CTX + ICB (n = 10) and CTX + ICB + CXB (n = 12), pool of two independent experiments. Arrow indicates treatment start, mice received CXB or vehicle treatment bidaily. d Analysis of circulating leukocytes in peripheral blood two weeks on-treatment in mice treated with control (n = 5), CTX + ICB (n = 5) and CTX + ICB + CXB (n = 7), representative plot shown of two independent experiments. Statistical significance for neutrophils, monocytes, CD8⁺ and CD4⁺ T cells is shown. e–j Analysis of tumor-infiltrating leukocytes three weeks on-treatment in mice treated with control (n = 10), CTX + ICB (n = 10) and CTX + ICB + CXB (n = 12), pool of two independent experiments. Frequency and absolute number normalized per gram of tumor for neutrophils (e), monocytes (f), CD8⁺ (g) and CD4⁺ T cells (h). Frequency of IFNγ⁺ (i) and CD44⁺ (j) of CD8⁺ and CD4⁺ T cells. Data in c–j are represented as mean ± SEM, ns = not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 as determined by one-way ANOVA with Tukey’s multiple comparisons test or Kruskal–Wallis test with Dunn’s multiple comparisons test for non-parametric data (b, e-j) or two-way ANOVA (c). Source data and exact p values are provided as a Source Data file.