Extended Data Fig. 3: Anti-CTLA-4 m2a treatment drives rapid and broad rearrangement of the adaptive and innate immune TME.

(a) UMAP of scRNA-seq data from day 1 post-treatment tumor infiltrating immune cells. Color code for cell type assignment is indicated in the legend. (b) UMAP projections of relative cell density per treatment, or relative CTLA-4 expression. (c) Comparison of the fractions of the indicated lymphoid cell types in individual mice in each treatment group. n = 3, 6 and 7 biologically independent samples for Ctrl, m1 and m2a groups respectively were used for statistical analysis. (d) Comparison of the fractions of the indicated myeloid cell types in individual mice in each treatment group. n = 2, 6 and 7 biologically independent samples for Ctrl, m1 and m2a groups respectively were used for statistical analysis. (e) Consensus hierarchical clustering (n clusters = 2, Methods) of cellular makeup across samples with Spearman correlation as distance metric. Color coded for P co-clustering after 500 iterations. (f) Histogram of CD4⁺ draining lymph node-derived cells frequency per FOXP3 protein expression, as measured by flow cytometry. n = 5 per group. Average percentage to the right of the dashed line is stated. Day 4 post-treatment. (g) As in (f) for day 1 post-treatment. (h) Growth curves of MCA-205 tumors in mice untreated, or treated with 50 µg/mouse anti-CTLA-4 m2a at days 6, 9, 12, with or without 100 µg/mouse anti-Gr1 at days 5 and 13. n = 5 mice per arm. In (c) and (d), bars represent mean ± SE of mice per treatment arm. Two-way ANOVA (fraction ~ treatment * batch) was used.