Extended Data Fig. 6: Blocking cancer cell IFN-I signaling increases the frequency of tumor-infiltrating CD8 T cells and improves ICB response.

A. Survival of mice with Res 237 WT or Oas1 KO (KO) tumors either non-treated (NT) or treated with anti-CTLA4 (aCTLA4) (NT, n = 10 mice per condition; aCTLA4, n = 20 mice per condition). Two independent KO clones are shown. P-values determined by two-sided log-rank test. B. Percent of tumor infiltrating CD8 T cells in total CD45⁺ cells from mice with Res 237 WT or Res 237 Oas1 KO (KO) tumors. P-values determined by two-sided t-test. C. MHC-I surface expression at baseline and after either IFNB or IFNG treatment of the indicated cell lines to assess knockout of B2m and/or Ifnar1 in Res 499 cancer cells. D. Tumor volume growth of mice injected with Res 499 tumors and treated with anti-CTLA4 plus anti-PD1 (dICB), delayed administration of the JAK inhibitor ruxolitinib (JAKi), or both (NT and JAKi, n = 5 mice per condition; dICB and JAKi + dICB, n = 10 mice per condition). P-values determined by a mix-effect regression model. E. Percent of PRF1⁺ CD8 T cells relative to total CD8 T cells in Res 499 WT or Res 499 Ifnar1 KO (KO) tumors from mice treated with or without anti-PD1 (aPD1) (n = 5 mice per group). P-values determined by two-sided t-test. F-G. Survival of mice with B16 WT or Ifnar1 KO (KO) tumors (F), or CT26 WT or Ifnar1 KO (KO) tumors (G), either non-treated (NT) or treated with anti-PD1 (NT, n = 5 mice per condition; aPD1, n = 10 mice per condition). P-values determined by two-sided log-rank test. H. Representative density plot of peripheral CD45⁺ cells in control mice or mice treated with anti-CD8 or anti-NK1.1 depleting antibody. I. Data shown in Fig. 5a replotted to emphasize effect of host immune cell status. Boxplots represent the 25^th percentile, median, 75^th percentile, and 1.5x IQR (whiskers).

Source data