Fig 2: Autologous MHC-Dependent Killing of Bladder Tumors by CD4+ and CD8+ TIL, Related to Figure 5Analysis of the increase in the number of dead cells over time from the same killing assay for CD4eff TIL (ie cultures with Tregs sorted out during expansion) at 30:1 effector:target ratio (A), CD4eff TIL at 30:1 effector:target ratio with a pan-anti-MHCII antibody (B), CD8+ TIL at 30:1 effector:target ratio (C), or CD8+ TIL at 30:1 effector:target ratio with a pan-anti-MHCI antibody (D), are shown. Control traces from separate wells with tumor only are included. All traces were normalized to the number of dead cells per mm2 at time point 0. Experiments were done using Cytotox Red. The observation of autologous tumor killing by CD4+ and CD8+ TIL above the background level of spontaneous death is representative of 2 independent experiments involving distinct aliquots from the same patient.

Fig 3: Targeting TAM with anti-CSF-1R blockade improves anti-PD-1 immunotherapy response in GBM-on-a-Chip.(A) A schematic outlining a dual inhibition therapeutic strategy for targeting both PD-1/PD-L1 and TAM CSF/CSF-1R signaling to inhibit the systemic immunosuppression among GBM, TAM and CD8+ T-cell. (B) Quantified M2-like marker CD163 expression on TAM in response to CSF-1R inhibitor BLZ945 in different GBM subtypes (GBML20, GBML08, and GBML91), showing the limited expression of CD163 in all GBM subtypes. (C) qPCR experiment confirming the inhibition of CD163 expression in TAM with the administration of CSF-1R inhibitor BLZ945. (D) Quantified results showing more infiltrated allogeneic CD8+ T-cells in all GBM subtypes (GBML20, GBML08, and GBML91) with PD-1 and CSF-1R dual inhibition therapy as compared to Nivolumab and BLZ945 monotherapy. (E) Quantified results showing an increased influx of activated CD154+CD8+ T-cells in PD-1 and CSF-1R dual inhibition therapy as compared to Nivolumab monotherapy. (F) Quantified cytokine levels showing significantly increased expression of pro-inflammatory cytokine (TNF-a) and decreased expression of immunosuppressive cytokine (TGF-ß1) in most GBM subtypes with dual inhibition therapy. Fold changes were calculated relative to control. Note the patient-specific responses with different pharmacological treatment. (G) Representative apoptosis images showing more apoptotic (green nuclei) GBM cells with co-blockade of PD-1 and CSF-1R relative to control in all GBM subtypes (GBML20, GBML08, and GBML91). Live GBM cells were stained with CellTracker Red (red color). (H) A therapeutic response summary of the heterogeneous and systemic immunosuppression in nine lines of GBM patients’ derived cells using GBM-on-a-Chip for relative percentages of GBM cell apotosis, CD154+CD8+ (%) and CD163+CD68+ (%) cell populations. 100% stacked bar chart was used to show the relative difference among distinct drug treatments. CSF-1R inhibitor BLZ945 (0.1 µg/ml) and PD-1 blockade nivolumab (1 µg/ml) were used in all the monotherapy or dual inhibition treatments. All control groups were treated with fresh cell culture media supplemented with DMSO (0.01%) and human IgG4 isotype control antibody (1 µg/mL, BioLegend). Error bars represent ± s.e.m. from three independent experiments. n > 80 in (B), (D), (E), and (H). P-values were calculated using the Student’s paired sample t-test or one-way ANOVA. *, p<0.05.

Fig 4: Influence of culture conditions on tumor-infiltrating lymphocytes (TILs) and T cell receptor (TCR) clusters.(a) Combined TCRβ clusters for fresh-frozen tumor (FFT), CD8+ double-positive (DP), CD8+ non-DP, and TILs cultured in four different conditions from metastatic tumor sample of patient mp26. Left: colors indicate VDJdb-defined clonotype specificities. Right: colors indicate the sample of origin. (b–e) TCRβ repertoire analysis for patient mp26 TILs cultured in four distinct conditions. Panels show (b) normalized counts of cluster-related TCRβ clonotypes, (c) proportion of the repertoire occupied by Melan-A-specific TCRβ clusters, (d) count and subset matching of clonotypes preferentially expanded compared to the IL-2high condition, and (e) the count of CD8+ DP matched clonotypes among the top 1500 clonotypes. N = number of separate tumor fragments. (f) Fraction of regulatory T cells (Tregs) (CD25+CD127-FoxP3+) among CD4+ T cells in cultured TILs from patient mp41 in each culture condition. (g) TIL counts relative to IL-2high culture conditions; n = 4 patients. (h) CD8+/CD4+ ratio for TILs from patient mp41 in each culture condition. Data in b ,c, f, g were analyzed with one-way ANOVA, Bonferroni multiple comparisons correction, with each condition compared to the IL-21/IL-2low/anti-PD-1 condition. Data in the panel (e) was analyzed with Kruskal-Wallis test, Dunn’s multiple comparison test, with each group compared to IL-21/IL-2low/anti-PD-1.

Fig 5: T cell receptor (TCR) clusters in CD39+PD1+ tumor-infiltrating lymphocytes (TILs).(a) The experimental workflow. (b–g) TCRβ repertoire analysis for CD8+ (b, d, f) and CD4+ (c, e, g) double-positive (DP) and non-DP TIL subsets sorted from metastatic lymph nodes of eight melanoma patients. Panels show repertoire clonality calculated as [1 − Normalized Shannon-Wiener index] (b, c), normalized counts (d, e), and cumulative frequency of cluster-related clonotypes, that is, total weight of all clusters as a proportion of TCRβ repertoire (f, g). Paired t-test. (h) TCRβ clusters identified in repertoires obtained from fresh-frozen tumor (FFT) samples, and sorted CD8+ DP and non-DP TILs for HLA-A*02 patient mp26. VDJdb-matched TAA-specific clusters are colored in red. ( i, j) Cumulative frequency of (i) VDJdb-matched TAA-specific clonotypes and (j) VDJdb-matched TAA-specific cluster-related clonotypes within CD8+ DP, CD8+ non-DP, and FFT TCRβ repertoires of patient mp26. One-way ANOVA, Bonferroni multiple comparisons correction. (k, l) Proportion of CD137+ cells among CD8+ T cells (k) and proportion of VDJdb-matched TAA-specific clonotypes in sorted CD137+CD8+ T cells (l) in DP and non-DP TILs from patient mp26 that were cultured and re-stimulated with TAA-loaded or control dendritic cells.