Fig. 3: Spp1hi-TAMs have a critical role in promoting immunotherapeutic resistance by inducing exhaustion in CD8+ T cells in vivo.
From: Evolution of myeloid-mediated immunotherapy resistance in prostate cancer
a,b, UMAP (a) and bar plots (b) showing immunosuppression scores among myeloid cells in mouse prostate cancer (n = 6,397; P < 0.001 for comparisons of Spp1hi-TAM and other subsets). Boxes represent IQR and bars indicate 25% − 1.5 × IQR and 75% + 1.5 × IQR, with outliers beyond 1.5 × IQR. The red dashed line shows the median score for Spp1hi-TAMs. c, Flow-cytometry plots showing reduced proliferation of activated splenic CD8+ T cells 3 days after co-culturing with Spp1hi-TAMs from CRPC. d,e, Quantification of proliferating (P = 0.02, P = 0.04 and P = 0.14 for effector:target (E:T) ratios of 1:1, 1:5 and 1:10, respectively (d) and polyfunctional (IFN-γ+TNF-α+; P = 0.01) CD8+ T cells with and without Spp1hi-TAMs at various ratios (e). Results are normalized to activated T cells alone; mean + s.e.m. from n = 4 experiments, with different colours for each and symbols for averages of 2–3 replicate wells. Red dashed lines indicate the normalized mean frequency of activated CD8+ T cells. f, Dosing schedule for ICIs (anti-CTLA-4 + anti-PD-1) or isotype-matched controls after adoptive transfer of Spp1hi-TAMs or PBS into CRPC. g, CRPC growth curves for ICI or isotype treatments after Spp1hi-TAM or PBS transfer from n = 3 experiments (P = 0.002, P = 0.02 and P = 0.59 for PBS+isotype versus PBS + ICIs, PBS + ICIs versus Spp1hi-TAM + ICIs and PBS + isotype versus Spp1hi-TAM + ICIs, respectively); PBS + isotype, n = 6; PBS + ICIs, n = 7; Spp1hi-TAM + ICIs, n = 7. Symbols represent mean ± s.e.m. h, Survival curves from the same experiment as g (P = 0.023, P = 0.013 and P = 0.755). i, Exhausted (CD38+PD-1+) CD8+ T cell frequencies in CRPC after Spp1hi-TAMs or PBS transfer with or without ICIs, assessed 1 day after the final injection (P = 0.02, P = 0.02, P > 0.99). Bars show mean + s.e.m. from n = 3 experiments; symbols represent individual mice. Statistical significance was determined by Kruskal–Wallis tests with Dunn’s correction (b,i), two-sided one-sample t-tests (d,e), ordinary one-way ANOVA with Sidak correction (g) and log-rank tests (h); *P < 0.05, **P < 0.01, ***P < 0.001; NS, not significant.
