Transcription Profile And Pathway Analysis Of The Endocannabinoid Receptor Inverse Agonist AM630 In The Core And Infiltrative Boundary Of Human Glioblastoma Cells

Background: We have previously reported that the endocannabinoid receptor inverse agonist AM630 is a potent inhibitor of isocitrade dehydrogenase-1 wild-type glioblastoma (GBM) core tumor cell proliferation. To uncover the mechanism behind the anti-tumour effects we have performed a transcriptional analysis of AM630 activity both in the tumour core cells (U87) and the invasive margin cells (GIN-8), the latter representing a better proxy of post-surgical residual disease. Results: The core and invasive margin cells exhibited markedly different gene expression profiles and only the core cells had high expression of a potential AM630 target, the CB1 receptor. Both cell types had moderate expression of the HTR2B serotonin receptor, a reported AM630 target. We found that the AM630 driven transcriptional response was substantially higher in the central cells than in the invasive margin cells, with the former driving the up regulation of immune response and the down regulation of cell cycle and metastatic pathways and correlating with transcriptional responses driven by established anti-neoplastics as well as serotonin receptor antagonists. Conclusion: Our results highlight the different responsiveness of the core and invasive margin cells. Taken together, whilst our findings identify AM630 as an anti-neoplastic drug, showing a high correlation with known anti-proliferative drugs, we find distinct drug sensitivies of the infiltrative margin relative to contrast-enhanced core regions of GBM upon which failed molecular targeted therapies to date have been predicated.

Features of the Lymphoid Microenvironment in Invasive Ductal Breast Cancer

The purpose of the study was to assess the distribution of tumor-infiltrating T-, B-lymphocytes and natural killer cells in various molecular subtypes of invasive ductal breast cancer and to establish their relationship with the degree of tumor differentiation. Materials and methods. The basis of the scientific work was a complex morphological, including immunohistochemical study of 193 cases of invasive ductal breast cancer. General histological processing of the samples was carried out in accordance with the standard technique. Immunohistochemical studies for CD3, CD20, CD56, ER, PR, c-erbB2, Ki-67 were performed according to the manufacturer's protocol with the control of samples. The grade of malignancy was determined according to the modified scheme of P. Scarff, H. Bloom and W. Richardson. The presence, localization, and expression intensity of diagnostic and prognostic biomarkers CD3, CD20, and CD56-positive cells (T-, B-, and natural killer cells, respectively) were determined using the new bioimage analysis software QuPath. Results and discussion. The differences obtained in our study varied depending on the subpopulations of immune cells and their location in the tumor tissue. The density of T and B lymphocytes was higher within the tumor and in the invasive margin in the non-luminal phenotypes compared to the luminal A and B phenotypes. Compared to the density of T-lymphocytes, the B-cell infiltrate was significantly (p <0.01) less pronounced in tumors of both luminal and non-luminal phenotypes. The lowest density of natural killer cells among tumor-infiltrating lymphocytes was found within the tumor of luminal subtypes. The percentage of infiltrates of T and B cells both within the tumor focus and in the invasive margin, as well as natural killer cells in the invasive margin, was significantly lower with G1 and G2 than with the degree of differentiation G3 (p <0.01). As for intratumoral natural killer cells, differences were found only between the degree of differentiation of G2 and G3, and the differences between G1 and G2 were not statistically significant (p> 0.05). Qualitative and quantitative assessment of the intensity of expression of T-, B- and natural killer cells also demonstrated a different degree of expression of lymphocytes, depending on the molecular subtype and location of the infiltrate. The luminal subtype A was characterized by a significant predominance of mild expression (2+) CD3 both within the tumor and in the invasive margin of the tumor (p <0.05), while in the luminal B and Her2 + phenotypes, the intensity of CD3 expression prevailed at the 1+ level. Triple negative tumors were characterized by strong expression of (3+) CD3 within the tumor and in the invasive margin of the tumor in all cases studied. Conclusion. Thus, tumor-infiltrating lymphocytes in different molecular subtypes of invasive ductal breast cancer can be considered a prognostic biomarker. Our results indicate a relationship between lymphoid infiltrate and the degree of differentiation in invasive ductal breast cancer, especially in less favorable molecular subtypes

EPCO-15. SPATIAL-TRANSCRIPTOMICS REVEALS UNIQUE MOLECULAR FEATURES OF FLUORESCENCE-SORTED 5-AMINOLEVULINIC ACID+ INFILTRATIVE TUMOR CELLS ASSOCIATED WITH RECURRENCE AND POOR SURVIVAL IN GLIOBLASTOMA

Spatiotemporal heterogeneity of glioblastoma (GBM) originating from genomic and transcriptional variation in spatially distinct sites, may contribute to subtype switching in GBM prior to and upon recurrence. Fluorescence-guided neurosurgical resection utilizing 5-aminolevulinic acid (5ALA) has enabled the isolation of infiltrative margin tumor cells (5ALA+ cells) from a background of non-neoplastic cells. To falsify the hypothesis that the 5ALA+ subpopulation(s) is defined by a unique cellular state, we have explored the spatial-transcriptomic (ST) landscape to interrogate molecular signatures unique to infiltrating 5ALA+ cells in comparison to GBM core, rim, and invasive margin non-neoplastic cells. ST analysis reveals that GBM molecular subtype plasticity is not restricted to recurrence, but manifests regionally in a cell-type-specific manner. Whilst GBM core and rim are highly enriched with Classical and Proneural subtypes, 5ALA+ cells are uniquely enriched with the Mesenchymal subtype (MES). Upregulation of the wound response pathway in 5ALA+ cells implicate hijacking of the wound healing pathway of neural cells to promote tumor growth. Exon-intron split analysis revealed an upregulation of exonic counts for MES and wound-response genes in 5ALA+ cells, implying that these genes are under active post-transcriptional control. Network analysis suggests that wound response genes, including chemokine CCL2 that recruits regulatory T-cells and monocytic myeloid-derived suppressor cells, are controlled by an IRF8-mediated transcriptional program in 5ALA+ cells. A higher stemness signature both in 5ALA+ cells and non-neoplastic cells of the invasive margin emphasizes the role of this microenvironment in stemness acquisition and defines 5ALA+ cells as a rare sub-population of GBM stem cells. Finally, we establish a link between the unique molecular-signatures of 5ALA+ cells, and poor survival and GBM recurrence. Characterization of the 5ALA+ infiltrative sub-population offers an opportunity to develop more effective GBM treatments and urges focus away from the GBM proliferative core, upon which failed targeted therapies have been predicated.

420 PROSTVAC in combination with nivolumab enhanced immune cell infiltration in prostate cancer

BackgroundProstate cancer (PC) is the most common non-cutaneous diagnosed cancer among men in USA.1 Although clinical outcomes are favorable for patients with localized disease, 20–30% of patients will develop metastatic prostate cancer (mPC) and have poor prognosis. Immunotherapy, as a single agent, provides benefit to a small subset of PC patients, which is thought to be partially due to its known cold tumor immune microenvironment (TIME). Combination studies are needed to enhance benefit.2 Prostvac is a therapeutic cancer vaccine engineered to activate an immune response against prostate-specific Antigen (PSA).3 Prostvac alone could induce systemic immune response by increasing immune-cell infiltrates in and around the tumor.4 In this study, we are exploring the effect of Prostvac in combination with nivolumab in TIME in prostate cancer.MethodsWe treated locally advanced prostate cancer patients (n=6) undergoing radical prostatectomy (RP) with neoadjuvant Prostvac in combination with nivolumab, an immune checkpoint PD-1 inhibitor. Dynamic changes in TIME before and after treatment were studied using multiplex immunofluorescence (Opal Method). Formalin fixed paraffin-embedded sections from matched pre-treated prostate biopsies and post-treated RP samples were stained with a validated T cell panel (DAPI, CD4, CD8, FOXP3, Ki67, Pan CK and PD-L1). To analyze the data, TIME was segmented into 3 compartments: intratumoral, invasive margin and benign.ResultsCombination immunotherapy significantly increased CD4+ T cell density in the invasive margin (mean 211.5 cells/mm2 vs 592.2 cells/mm2, p<0.05), with similar trend in the intratumoral and the benign compartments. CD8+ T cell density increased after treatment in the invasive margin (mean 47.25 cells/mm2 vs 157cells/mm2) and the benign compartment. 5/6 and 4/6 patients showed more than 2-fold increase of CD4 and CD8 T cells in the TIME, respectively, in at least one of the three compartments. Increased proliferative indices in CD4+ and CD8+ T cells were also seen after treatment. Tregs were present in low frequencies in TIME (maximum of 12 cells/mm2) with no significant changes. Moreover, a significant drop in tumor cell Ki67 after treatment (mean 252.8 cells/mm2 vs 100.5 cells/332, p<0.05) suggests that the combination may control tumor growth.ConclusionsThe combination of Neoadjuvant Prostvac and nivolumab was associated with increased immune cell infiltration in a cohort of early prostate cancer patients. A broader examination of the TIME and the role immune cells undertake to control tumor growth is on-going.Trial RegistrationNCT02933255ReferencesSiegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin (Internet) 2020;70:7–3Zhao SG, Lehrer J, Chang SL, et al. The immune landscape of prostate cancer and nomination of PD-L2 as a potential therapeutic target. J Natl Cancer Inst 2018;111:301–10.Madan RA, Arlen PM, Mohebtash M, et al. Prostvac-VF: a vectorbased vaccine targeting PSA in prostate cancer. Expert Opin Investig Drugs 2009;18:1001–11Abdul Sater H, Marté JL, Donahue RN, et al. Neoadjuvant PROSTVAC prior to radical prostatectomy enhances T-cell infiltration into the tumor immune microenvironment in men with prostate cancer. J Immunother Cancer 2020;8(1):655–64Ethics ApprovalThis study was performed in compliance with ethical standard and was approved by the NIH IRB, 17C-0007. All patients participating in this study gave an informed consent before taking part.

Intratumour heterogeneity in microRNAs expression regulates glioblastoma metabolism

While specific microRNA (miRNA) signatures have been identified in glioblastoma (GBM), the intratumour heterogeneity in miRNA expression has not yet been characterised. In this study, we reveal significant alterations in miRNA expression across three GBM tumour regions: the core, rim, and invasive margin. Our miRNA profiling analysis showed that miR-330-5p and miR-215-5p were upregulated in the invasive margin relative to the core and the rim regions, while miR-619-5p, miR-4440 and miR-4793-3p were downregulated. Functional analysis of newly identified miRNAs suggests their involvement in regulating lipid metabolic pathways. Subsequent liquid chromatography–mass spectrometry (LC–MS) and tandem mass spectroscopy (LC–MS/MS) profiling of the intracellular metabolome and the lipidome of GBM cells with dysregulated miRNA expression confirmed the alteration in the metabolite levels associated with lipid metabolism. The identification of regional miRNA expression signatures may underlie the metabolic heterogeneity within the GBM tumour and understanding this relationship may open new avenues for the GBM treatment.

Spatial-transcriptomics reveals unique defining molecular features of fluorescence-sorted 5-aminolevulinic acid+ infiltrative tumor cells associated with glioblastoma recurrence and poor survival

Spatiotemporal-heterogeneity of glioblastoma (GBM) originating from the genomic and transcriptional variation in spatially distinct intra-tumor sites, may contribute to subtype switching in GBM prior to and upon recurrence. Fluorescence-guided neurosurgical resection utilizing 5-aminolevulinic acid (5ALA) has enabled the isolation of infiltrative margin tumor cells (5ALA+ cells) from a background of non-neoplastic cells. We have explored the spatial-transcriptomic (ST) landscape to interrogate molecular signatures unique to infiltrating 5ALA+ cells in comparison to GBM core, rim, and invasive margin non-neoplastic cells. ST analysis reveals that GBM molecular subtype plasticity is not restricted to recurrence, but manifests regionally in a cell-type-specific manner. Whilst GBM core and rim are highly enriched with Classical and Proneural subtypes, the unique enrichment of the Mesenchymal subtype (MES) in 5ALA+ cells supports the hypothesis that MES 5ALA+ cells may drive GBM recurrence. Upregulation of the wound response pathway in 5ALA+ cells signifies the possibility of hijacking the wound healing pathway of neural cells to promote tumor growth. Exon-intron split analysis revealed an upregulation of exonic counts for MES and wound-response genes in 5ALA+ cells, implying that these genes are under active post-transcriptional control. Network analysis suggests that wound response genes, including chemokine CCL2 that recruits regulatory T-cells and monocytic myeloid-derived suppressor cells, are controlled by an IRF8-mediated transcriptional program in 5ALA+ cells. A higher stemness signature both in 5ALA+ cells and non-neoplastic cells of the invasive margin emphasizes the role of this microenvironment in stemness acquisition and defines 5ALA+ cells as a rare sub-population of GBM stem cells. Finally, we establish a link between the unique molecular signatures of 5ALA+ cells and poor survival and GBM recurrence. Characterization of the 5ALA+ infiltrative sub-population offers an opportunity to develop more effective GBM treatments and urges focus away from the GBM proliferative core, upon which failed targeted therapies have been predicated.

Spatial analysis of tumor infiltrating lymphocytes based on deep learning using histopathology image to predict progression-free survival in colorectal cancer

PurposeThis study aimed to explore the prognostic impact of spatial distribution of tumor infiltrating lymphocytes (TILs) quantified by deep learning (DL) approaches based on digitalized whole slide images stained with hematoxylin and eosin in patients with colorectal cancer (CRC).MethodsThe prognostic impact of spatial distributions of TILs in patients with CRC was explored in the Yonsei cohort (n=180) and validated in the TCGA cohort (n=268). Concurrently, two experienced pathologists manually measured TILs at the most invasive margin as 0-3 by the Klintrup-Mäkinen (KM) grading method and compared to DL approaches. Interobserver agreement for TILs was measured using Cohen’s kappa coefficient.ResultsOn multivariate analysis of spatial TILs features derived by DL approaches and clinicopathological variables including tumor stage, Microsatellite instability, and KRAS mutations, TILs densities within 200 μm of the invasive margin (f_im200) was remained as the most significant prognostic factor for progression-free survival (PFS) (HR 0.004 [95% CI, 0.0001-0.1502], p=.002) in the Yonsei cohort. On multivariate analysis using the TCGA dataset, f_im200 retained prognostic significance for PFS (HR 0.031, [95% CI 0.001-0.645], p=.024). Interobserver agreement of manual KM grading based on Cohen’s kappa coefficient was insignificant in the Yonsei (κ=.109) and the TCGA (κ=.121), respectively. The survival analysis based on KM grading showed statistically significant different PFS from the TCGA cohort, but not the Yonsei cohort.ConclusionsAutomatic quantification of TILs at the invasive margin based on DL approaches showed a prognostic utility to predict PFS, and could provide robust and reproducible TILs density measurement in patients with CRC.Data and Code AvailabilitySource code and data used for this study is available at the following link: https://github.com/hwanglab/TILs_Analysis

A template to quantify the location and density of CD3 + and CD8 + tumor-infiltrating lymphocytes in colon cancer by digital pathology on whole slides for an objective, standardized immune score assessment

Background In colon cancer, the location and density of tumor-infiltrating lymphocytes (TILs) can classify patients into low and high-risk groups for prognostication. While a commercially available ‘Immunoscore®’ exists, the incurred expenses and copyrights may prevent universal use. The aim of this study was to develop a robust and objective quantification method of TILs in colon cancer. Methods A consecutive, unselected series of specimens from patients with colon cancer were available for immunohistochemistry and assessment of TILs by automated digital pathology. CD3 + and CD8 + cells at the invasive margin and in tumor center were assessed on consecutive sections using automated digital pathology and image analysis software (Visiopharm®). An algorithm template for whole slide assessment, generated cell counts per square millimeters (cells/mm2), from which the immune score was calculated using distribution volumes. Furthermore, immune score was compared with clinical and histopathological characteristics to confirm its relevance. Results Based on the quantified TILs numbers by digital image analyses, patients were classified into low (n = 83, 69.7%), intermediate (n = 14, 11.8%) and high (n = 22, 18.5%) immune score groups. High immune score was associated with stage I–II tumors (p = 0.017) and a higher prevalence of microsatellite instable (MSI) tumors (p = 0.030). MSI tumors had a significantly higher numbers of CD3 + TILs in the invasive margin and CD8 + TILs in both tumor center and invasive margin, compared to microsatellite stable (MSS) tumors. Conclusion A digital template to quantify an easy-to-use immune score corresponds with clinicopathological features and MSI in colon cancer.

Heterogeneity and Geographical Patterns of PD-L1 Positive Tumor Infiltrating Immune Cells in Colonic Adenocarcinoma

Introduction/Objective Programmed death-ligand 1 (PD-L1) status is an important prognostic and predictive biomarker. PD-L1 status is difficult to accurately assess in tumor cells (TCs) due to tumor heterogeneity and lack of standardized histologic evaluation. PD-L1 expression by tumor infiltrating immune cells (TIICs) is an independent prognostic marker and may provide fewer diagnostic challenges, as TIICs appear less heterogeneous. We examined the geographical patterns of PD-L1 expression in TCs and TIICs in order to evaluate heterogeneity and ease of histological evaluation. Methods 197 blocks of colonic adenocarcinoma from 33 patients were retrospectively reviewed and stained for PD- L1. PD-L1 was interpreted as positive or negative in the TCs and TIICs. Geographical patterns of expression were recorded for the TCs and the TIICs at the surface, center of tumor, and invasive margin. Geographical patterns were compared using a chi-square test. Results 56% of blocks had PD-L1 positive TCs. 42.7% showed positive TCs at the invasive margin. 38.2% had positive TCs at the invasive margin and surface. No blocks had positive TCs at the center of the tumor. 79% of blocks had PD-L1 positive TIICs. 20% showed positive TIICs at the invasive margin. 37.4% had positive TIICs at the surface and invasive margin. 15.5% of TIICs expressed PD-L1 throughout the tumor. There was a statistically significant difference in PD-L1 expression on TCs versus TIICs in 4 of the 7 geographic patterns. Conclusion PD-L1 expression on TCs of colonic adenocarcinoma is more heterogeneous than PD-L1 expression by TIICs. TIICs express PD-L1 more consistently throughout the tumor and are easier to identify and report. PD-L1 expression by TIICs is an independent prognostic indicator and may be more useful than TCs when evaluating the tumor microenvironment and determining PD-L1 status.