Journal: Frontiers in Immunology
Article Title: Translational 3D in vitro models for immunotherapy testing: from reconstituted organoid co-culture assays to autologous ex vivo patient tissues
doi: 10.3389/fimmu.2026.1721016
Figure Lengend Snippet: ICI testing in Ex Vivo Patient Tissue assay in native TME environment. (a) Left panel shows cropped images from sample OLS-016 shown as a projection from an image stack, part of a single well. Raw images of nuclei stained with Hoechst33342 and actin cytoskeleton with rhodamine-Phalloidin. Object segmentation output after Ominer processing, separated for large object and single cells. Middle and right panels show high resolution images of samples OLP-028 and OLS-147 immuno-stained with EpCAM and CD3, and CD45 antibodies, confirming presence of both multicellular and single tumor cells as well as immune infiltrated cells in the resection samples present after 6 days culture. (b) Quantified HCI readout of untreated negative control (medium), immuno-stimulatory controls CD3/CD28 Dynabeads and SEA, immune checkpoint inhibiting (ICI) Pembrolizumab and Ipilimumab, and toxicity control Staurosporin in NSCLC samples OLS-016, OLS-113, OLS-090. Normalized, as percentage of control (poc), tumoroid size (median) shown in a violin plot, object count (median) shown in a box plot, single cell count (mean) shown in a bar chart. n=8 replicates per condition (c) Cytokine analyses in pooled supernatants per treatment condition collected from OLS-016 assay plate for analytes IFN-γ, Granzyme B, and IL-2. Averages of two pooled sets of quadruplicates with standard deviation shown, no statistics shown. (d) Single DAB staining and Multiplex IF (mIF) on FFPE embedded tumor fragments from patient-derived NSCLC samples. Single staining for epithelial marker Pan-CK, TME and immune cell and biomarkers α-SMA, CD45, CD3, CD4, CD8, CD68, CD80, CD163, FoxP3, PD-L1, PD1, CTLA-4. mIF for epithelial marker EpCAM, and TME markers α-SMA and CD45. Quantification on the stacked bar shows the representative presence of each subpopulation in the mIF sample. (e) Flow cytometry analyses of the immune populations (CD45 + ) of sample OLS-090 identifies most cells as basic T cells (CD3 + ) next to myeloid cells (CD11b + ) presented in a donut plot. (f) A more extensive flow cytometry panel tracks subsets of immune cells at baseline and timepoints 24h and 144h represented in donut plots. Overall, relative proportions of cells remain stable in the first day, whereas after 144h fewer cells could be recovered from the sample, of which the CD8 + population was progressively less present. (g) Heatmaps of n=49 NSCLC, n=6 Ovarian cancer, n=10 Breast cancer, and n=5 Bladder cancer ex vivo samples’ responses to controls and immuno-stimulatory and ICI compounds in EVPT platform. Total tumoroid area was normalized as % of medium control for every sample. Shown is the median value from 8 technical replicate wells. Statistical analysis used for 4b is Ordinary One-way ANOVA, followed by post-hoc multiple comparison test using Tukey’s methodology. Scale bar in 4a segmentations (left) represents 200 µm in IF 100 µm (middle, right), in 4d 100 µm.
Article Snippet: Test treatments included bispecific antibodies targeting EGFR and CD3 (0.2 ng/mL-2 μg/ml) (Creative Biolabs- #CBL-BilE003-11) or BSA and CD3 control (2 μg/ml) (Creative Biolabs, Cat #BSFV-C001), Claudin 18.2 targeting BiTE AMG910 (1 ng/mL-10 μg/mL) (MCE, Cat #HY-P99350) or Ultra-LEAFTM Purified Human IgG1 Isotype Ctrl (10 μg/mL) (Biolegend, Cat #403501), EGFRscFv-CD28-CD3ζ CAR T-cells and Non transduced T cells (both ProMab Biotechnologies).
Techniques: Ex Vivo, Staining, Negative Control, Control, Single Cell, Standard Deviation, Multiplex Assay, Derivative Assay, Marker, Flow Cytometry, Comparison