Stromal Regions Research Articles

Spatial transcriptomics depict ligand-receptor crosstalk heterogeneity at the tumor-stroma interface in long-term ovarian cancer survivors.

Advanced high-grade serous ovarian cancer (HGSC) is an aggressive disease that accounts for 70% of all ovarian cancer deaths. Nevertheless, 15% of patients diagnosed with advanced HGSC survive more than 10 years. The elucidation of predictive markers of these long-term survivors (LTS) could help identify therapeutic targets for the disease, and thus improve patient survival rates. To investigate the stromal heterogeneity of the tumor microenvironment (TME) in ovarian cancer, we used spatial transcriptomics to generate spatially resolved transcript profiles in treatment naïve advanced HGSC from LTS and short-term survivors (STS) and determined the association between cancer-associated fibroblasts (CAF) heterogeneity and survival in patients with advanced HGSC. Spatial transcriptomics and single-cell RNA sequencing data were integrated to distinguish tumor and stroma regions, and a computational method was developed to investigate spatially resolved ligand-receptor interactions between various tumor and CAF subtypes in the TME. A specific subtype of CAFs and its spatial location relative to a particular ovarian cancer cell subtype in the TME correlated with long-term survival in advanced HGSC patients. Also, increased APOE-LRP5 crosstalk occurred at the stroma-tumor interface in tumor tissues from STS compared to LTS. These findings were validated using multiplex immunohistochemistry. Overall, this spatial transcriptomics analysis revealed spatially resolved CAF-tumor crosstalk signaling networks in the ovarian TME that are associated with long-term survival of HGSC patients. Further studies to confirm whether such crosstalk plays a role in modulating the malignant phenotype of HGSC and could serve as a predictive biomarker of patient survival are warranted.

Spatial multi-omics revealed the impact of tumor ecosystem heterogeneity on immunotherapy efficacy in patients with advanced non-small cell lung cancer treated with bispecific antibody

BackgroundImmunotherapy for malignant tumors has made great progress, but many patients do not benefit from it. The complex intratumoral heterogeneity (ITH) hindered the in-depth exploration of immunotherapy. Conventional bulk sequencing...

Abstract C111: Spatial transcriptome profiling of tumor and tumor microenvironment of triple negative breast cancer in patients of African American and European American descent

Abstract Triple-negative breast cancer (TNBC) is an aggressive breast cancer subtype that disproportionately affects women of African origin. AA women with TNBC have worse clinical outcomes than women of European descent as it presents with a more aggressive and progressive disease. Understanding the underlying biology of the tumor and its microenvironment is critical to address the disparity associated with racial differences in breast cancer. Bulk RNA-sequencing is the most common and useful method to analyze whole transcriptome. However, this method cannot deconvolute the individual components of the tumor and its associated stroma. The NanoString GeoMx digital spatial profiling is a multiplexed platform that spatially resolves and quantifies the abundance of genes present in tumor and stroma. In this study, we have utilized spatial RNA-seq technology to determine the differential gene expression in tumor epithelial cells and its associated stroma in AA and EA TNBC patients. Gene expression from spatially defined regions in tumor and stroma in each patient and in between races identified distinct molecular pathways. The tumor epithelial region of AA has high expression of fatty acid biosynthesis, nucleotide excision repair, RNA helicase like proteins and exhaustion signaling while EA tumor epithelial region had high expression of neuropilin 2, interferon signaling and TGF beta signaling. The pathways overexpressed in the AA stromal region included tryptophan metabolism, chemokine signaling pathway, interferon signaling and exhaustion signature while in the EA stromal region, the pathways pertinent to mTOR signaling, AURORA kinase, cysteine endopeptidases were more prominently expressed. AA tumors were significantly more inflamed than EA and exhibited high infiltration of immune cells. The stromal region of AA tumors has high population of macrophages, CD8 T memory, Treg’s and neutrophils more than EA tumors. The tumor epithelial region of AA has high infiltration of plasmacytoid dendritic cells (pDC), B memory, monocytes, natural killer cells (NK), CD4 memory, CD4 naïve, Treg’s and plasma cells. However, the exhaustion signature proteins, PDCD1 and LAG3 were also higher in tumor-infiltrated and stromal immune cells in AA tumor. The EA tumors were scarce in immune cells in the stroma as well as inside the tumor epithelial, though interferon signaling was substantially higher in EA tumor epithelial cells as compared to AA. Thus, using NanoString GeoMx digital spatial profiling platform, we have spatial profiled AA and EA TNBC and assessed the inter and intra-tumor heterogeneity within and in between races. This is one of the first study that describes not only expression variation of genes between races but also between stroma and tumor epithelial cells within the tissue of same patient and between different patients. These findings address the biological determinant of racial disparity by patient-centered investigation of factors associated with tumor and tumor microenvironment. Citation Format: Deepa Bedi, Balasubramanyam Karanam, Isra Elhussin, Chitra R. Nayak, Clayton Yates. Spatial transcriptome profiling of tumor and tumor microenvironment of triple negative breast cancer in patients of African American and European American descent [abstract]. In: Proceedings of the 15th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2022 Sep 16-19; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2022;31(1 Suppl):Abstract nr C111.

Accumulation of senescent cells in the stroma of aged mouse ovary.

Senescent cells play a detrimental role in age-associated pathogenesis by producing factors involved in senescence-associated secretory phenotype (SASP). The present study was conducted to examine the possibility that senescent cells are present in aged ovaries and, if so, to determine the tissue region where senescent cells accumulate using a mouse model. Female mice at 2-4 and 8-10 months were used as reproductively young and aged models, respectively; the latter included mice with and without reproductive experience. Cells positive for senescence-associated β-galactosidase (SA-β-Gal) staining, one of the markers of cellular senescence, were detected in the stromal region of aged, but not young, ovaries regardless of reproductive experience. Likewise, the localization of cells expressing CDKN2A (cyclin dependent kinase inhibitor 2A), another senescence marker, in the stromal region of aged ovaries was detected with immunohistochemistry. CDKN2A expression detected by western blotting was significantly higher in the ovaries of aged mice with reproductive experience than in those without the experience. Moreover, cells positive for both γH2AX (a senescence marker) and fluorescent SA-β-Gal staining were present in those isolated from aged ovaries. In addition, the transcript levels of several SASP factors were significantly increased in aged ovaries. These results suggest that senescent cells accumulate in the ovarian stroma and may affect ovarian function in aged mice. Additionally, reproductive experience may promote accumulation.

Arachnoid granulations are lymphatic conduits that communicate with bone marrow and dura-arachnoid stroma.

Arachnoid granulations (AG) are poorly investigated. Historical reports suggest that they regulate brain volume by passively transporting cerebrospinal fluid (CSF) into dural venous sinuses. Here, we studied the microstructure of cerebral AG in humans with the aim of understanding their roles in physiology. We discovered marked variations in AG size, lobation, location, content, and degree of surface encapsulation. High-resolution microscopy shows that AG consist of outer capsule and inner stromal core regions. The fine and porous framework suggests uncharacterized functions of AG in mechanical CSF filtration. Moreover, internal cytokine and immune cell enrichment imply unexplored neuroimmune properties of these structures that localize to the brain-meningeal lymphatic interface. Dramatic age-associated changes in AG structure are additionally identified. This study depicts for the first time microscopic networks of internal channels that communicate with perisinus spaces, suggesting that AG subserve important functions as transarachnoidal flow passageways. These data raise new theories regarding glymphatic-lymphatic coupling and mechanisms of CSF antigen clearance, homeostasis, and diseases.

Identification of tumour-infiltrating myeloid subsets associated with overall survival in lung squamous cell carcinoma.

Lung squamous cell carcinoma (LUSC) is a primary subtype of lung cancer with limited therapeutic options and poor prognosis, and tumour-infiltrating myeloid cells (TIMs) are key regulators of LUSC. However, the correlation between the abundance of TIM subtypes and clinical outcomes of LUSC remains unexplored. This study aimed to develop and validate a prognostic model for low- and high-risk patients with LUSC based on myeloid cell microenvironments. TIM markers in the tumoural (T) and stromal (S) regions were quantified using immunohistochemistry for 502 LUSC patients. L1-penalized Cox regression was used to develop a myeloid survival score (MSS) model based on the training cohort, followed by validation in distinct cohorts from multiple centres. RNA sequencing and immunostaining were used to examine the mechanisms of myeloid cells in LUSC progression and predict potential drug targets and therapeutic agents. Of the 12 myeloid markers, CD163T, CD163S, and S100A12T were highly associated with overall survival (OS) in LUSC patients. The MSS of the three myeloid signatures accurately categorized LUSC patients into risk categories, with an observable difference in OS between the training and validation cohorts. Tumours with high MSS were associated with enhanced antioxidative ability and hedgehog signalling and a shift to a more pro-tumorigenic microenvironment, accompanied by a reduced tumour cell immunogenicity and increased CD8+ T cell exhaustion patterns. Additionally, in high-risk patients, potential drug targets and compounds regulating hedgehog signalling were identified. Our study provides the first prognostic myeloid signature for LUSC, which may help advance precision medicine. © 2022 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Abstract B010: Investigating endoplasmic reticulum stress responses in pancreatic ductal adenocarcinoma under lipid imbalance

Abstract Pancreatic ductal adenocarcinoma (PDAC) is the 4th leading cause of cancer-related death with a five-year survival rate of 10%. The complex genetic makeup and heterogeneity of its microenvironment make PDAC difficult to treat. Desmoplasia, a major feature of PDAC’s TME, are dense fibrotic and inflammatory stromal regions. Desmoplastic reaction results in poor intratumoral perfusion and elevated interstitial fluid pressure, creating regions with severe oxygen and nutrient deprivation. One result is a deficiency in unsaturated fatty acids (uFAs) because the de novo synthesis of uFAs requires oxygen. Our results indicate that PDAC cells experience endoplasmic reticulum (ER) stress and trigger subsequent ER stress responses under uFA limited conditions in vitro. The ER stress responses are governed by three major stress sensors: Inositol-requiring enzyme 1𝛼 (IRE1𝛼), PKR-like ER kinase (PERK) and activating transcription factor 6𝛼 (ATF6𝛼). We hypothesize that ER stress responses are tumor promoting in PDAC and used as a survival strategy under uFA deprivation. In addition, previous studies showed that MYC-driven cancer cells are highly reliant on the ER stress response pathways. Therefore, we hypothesize that the two major PDAC transcriptional subtypes “Classical” and “Basal-like” have distinct dependencies on ER stress response pathways, with the basal-like subtype being more dependent due to an enrichment of MYC gene signatures. Our results suggest that opposite to cancer cell promoting, ER stress responses mediated by IRE1𝛼, PERK and ATF6𝛼 are cytotoxic in vitro under uFA deprived conditions. By analyzing the published bulk RNA-seq from patient samples, we found that the basal-like PDAC subtype is enriched with an ER stress gene signature. To avoid contamination from other cell populations, we are currently analyzing a published scRNA-seq dataset from 15 independent patient samples. We will investigate whether the ER stress signature is also enhanced in basal-like cells. Basal-like PDAC cells are usually more mesenchymal; we therefore created isogenic “epithelial” and “mesenchymal” lines to mimic PDAC subtypes. However, we did not see signaling preferences between the lines. One caveat is that the basal-like subtype have many more features than being mesenchymal, and our isogenic system might not be a faithful modeling of the subtypes. We are studying whether hypoxic PDAC cells experience ER stress in vivo due to a deprivation in uFAs using multiplex fluorescent imaging, stimulated Raman microscopy and/or mass spectrometry. We are evaluating the role of ER stress responses in vivo and see whether inhibiting ER stress response alone or in combination with other reagents gives rise to potential therapeutic opportunities. This project will provide a better understanding of the role ER stress responses play under uFA deprivation and will also enable us to further understand PDAC subtypes in terms of how they utilize ER stress responses. Citation Format: Yanqing (Christine) Jiang, Xu Han, Mai Wang, Raymond Ng, Pamela Burgess-Jones, John Tobias, Nathan Coffey, Andrew Hu, Brian Keith, Celeste Simon. Investigating endoplasmic reticulum stress responses in pancreatic ductal adenocarcinoma under lipid imbalance [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer; 2022 Sep 13-16; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2022;82(22 Suppl):Abstract nr B010.

Polarization clustering of biological structures with Mueller matrix parameters.

Mueller matrix imaging polarimetry (MMIP) is a promising technique for the characterization of biological tissues, including the classification of microstructures in pathological diagnosis. To expand the parameter space of Mueller matrix parameters, we propose new vector parameters (VPs) according to the Mueller matrix polar decomposition method. We measure invasive bladder cancer (IBC) with extensive necrosis and high-grade ductal carcinoma in situ (DCIS) with MMIP, and the regions of cancer cells and fibrotic stroma are classified with the VPs. Then the proposed and existing VPs are mapped on the Poincaré sphere with 3D visualization, and an indicator of spatial feature is defined based on the minimum enclosing sphere to evaluate the classification capability of the VPs. For both IBC and DCIS, the results show that the proposed VPs exhibit evident contrast between the regions of cancer cells and fibrotic stroma. This study broadens the fundamental Mueller matrix parameters and helps to improve the characterization ability of the MMIP technique.

Pregnancy preparation: redistribution of CCR7-positive cells in the rat uterus.

Changes in the endometrium prior to implantation may be critical in predicting pregnancy outcomes. This study shows that the endocrine system directs positional changes in CCR7+ cells before implantation, which may be critical for developing maternal tolerance. Suppression of the maternal immune system is vital for the implantation of the semi-allogeneic embryo. Although progress in understanding the dialogue between mother and embryo has been made, key interactions between maternal immune cells, hormones, and chemokines remain elusive. Uterine expression of the C-C chemokine receptor type 7 (CCR7) could recruit T regulatory cells and facilitate localized immune suppression. To test this concept, Ccr7 mRNA and protein were assessed in uterine tissue. Ccr7 mRNA expression peaked at day 4 in pregnant rat uteri and then declined at days 5 and 6. CCR7 protein showed similar quantitative changes. To test if female sex steroids affected the spatial distribution of CCR7-expressing cells, uteri from ovariectomized rats, progesterone-pretreated rats (2 mg daily), and progesterone-pretreated rats injected with estradiol (0.2 µg) were analyzed. Progesterone increased CCR7-positive (+) cells in the antimesometrial stroma. Progesterone and estradiol increased CCR7+ cells in the mesometrial stroma. Estradiol increased the density of cluster of differentiation 4 (CD4) positive cells in the mesometrial stromal region over progesterone alone. The density of cells expressing the T regulatory cell marker, forkhead box protein 3 (FOXP3), increased in the antimesometrial stroma in response to progesterone alone. Progesterone and estradiol increased FOXP3+ cells in the antimesometrial region of the stroma. Co-localization of CCR7, CD4, and FOXP3 in the stroma suggests CCR7+ cells are T regulatory cells. Polarization of CCR7+ cells in the endometrial stroma was an intrinsic response regulated by sex steroids and did not require the presence of an embryo.

Integrating microarray-based spatial transcriptomics and single-cell RNA-sequencing reveals tissue architecture in esophageal squamous cell carcinoma.

SummaryBackgroundThe tumor microenvironment (TME) serves as an important factor in tumorigenesis and metastasis. Although distinct cell subsets can be identified via single-cell RNA sequencing (scRNA-seq), the spatial composition of cells within the TME is difficult to characterise.MethodsTissue samples were collected from three patients with esophageal squamous cell carcinoma (ESCC), and scRNA-seq was performed to identify distinct cell subsets. In addition, a microarray-based spatial transcriptomics (ST) method was used to characterise the spatial landscape of expression data via an array of spots. Using multimodal intersection analysis (MIA) to integrate scRNA-seq and ST, the exact cellular components of the tumor and stromal regions were annotated.FindingsThe subpopulations of seven stromal cells were identified within the TME of ESCC, and the architecture of scRNA-seq-determined subsets was mapped in cancer and stromal regions. The distribution of various stromal cells and their subpopulations was heterogeneous. Compared with immune cells, non-immune stromal cells were significantly enriched in the TME. Most subsets of epithelial cells were enriched in the cancer regions, whereas inflammatory cancer-associated fibroblasts were correlated with the stromal regions. Furthermore, TME features were different between metastatic and non-metastatic samples and between the primary and metastatic sites of the metastatic sample.InterpretationThis study revealed the spatial landscape of various cell subsets within the TME and the potential cross-talk among diverse cells, which provides novel insights into cancer intervention.FundingA full list of funding bodies that contributed to this study can be found in the Acknowledgements section.

Neoadjuvant Chemotherapy Is Associated with Altered Immune Cell Infiltration and an Anti-Tumorigenic Microenvironment in Resected Pancreatic Cancer.

Neoadjuvant chemotherapy is increasingly administered to patients with resectable or borderline resectable pancreatic ductal adenocarcinoma (PDAC), yet its impact on the tumor immune microenvironment is incompletely understood. We employed quantitative, spatially resolved multiplex immunofluorescence and digital image analysis to identify T-cell subpopulations, macrophage polarization states, and myeloid cell subpopulations in a multi-institution cohort of up-front resected primary tumors (n = 299) and in a comparative set of resected tumors after FOLFIRINOX-based neoadjuvant therapy (n = 36) or up-front surgery (n = 30). Multivariable-adjusted Cox proportional hazards models were used to evaluate associations between the immune microenvironment and patient outcomes. In the multi-institutional resection cohort, immune cells exhibited substantial heterogeneity across patient tumors and were located predominantly in stromal regions. Unsupervised clustering using immune cell densities identified four main patterns of immune cell infiltration. One pattern, seen in 20% of tumors and characterized by abundant T cells (T cell-rich) and a paucity of immunosuppressive granulocytes and macrophages, was associated with improved patient survival. Neoadjuvant chemotherapy was associated with a higher CD8:CD4 ratio, greater M1:M2-polarized macrophage ratio, and reduced CD15+ARG1+ immunosuppressive granulocyte density. Within neoadjuvant-treated tumors, 72% showed a T cell-rich pattern with low immunosuppressive granulocytes and macrophages. M1-polarized macrophages were located closer to tumor cells after neoadjuvant chemotherapy, and colocalization of M1-polarized macrophages and tumor cells was associated with greater tumor pathologic response and improved patient survival. Neoadjuvant chemotherapy with FOLFIRINOX shifts the PDAC immune microenvironment toward an anti-tumorigenic state associated with improved patient survival.

Computational tumor stroma reaction evaluation led to novel prognosis-associated fibrosis and molecular signature discoveries in high-grade serous ovarian carcinoma.

BackgroundAs one of the key criteria to differentiate benign vs. malignant tumors in ovarian and other solid cancers, tumor-stroma reaction (TSR) is long observed by pathologists and has been found correlated with patient prognosis. However, paucity of study aims to overcome subjective bias or automate TSR evaluation for enabling association analysis to a large cohort.Materials and methodsServing as positive and negative sets of TSR studies, H&E slides of primary tumors of high-grade serous ovarian carcinoma (HGSOC) (n = 291) and serous borderline ovarian tumor (SBOT) (n = 15) were digitally scanned. Three pathologist-defined quantification criteria were used to characterize the extents of TSR. Scores for each criterion were annotated (0/1/2 as none-low/intermediate/high) in the training set consisting of 18,265 H&E patches. Serial of deep learning (DL) models were trained to identify tumor vs. stroma regions and predict TSR scores. After cross-validation and independent validations, the trained models were generalized to the entire HGSOC cohort and correlated with clinical characteristics. In a subset of cases tumor transcriptomes were available, gene- and pathway-level association studies were conducted with TSR scores.ResultsThe trained models accurately identified the tumor stroma tissue regions and predicted TSR scores. Within tumor stroma interface region, TSR fibrosis scores were strongly associated with patient prognosis. Cancer signaling aberrations associated 14 KEGG pathways were also found positively correlated with TSR-fibrosis score.ConclusionWith the aid of DL, TSR evaluation could be generalized to large cohort to enable prognostic association analysis and facilitate discovering novel gene and pathways associated with disease progress.

Prognostic significance of immunoscore related markers in bladder cancer

BackgroundThe significance of total and specific subpopulations of tumor-infiltrating lymphocytes (TILs) in cancer is now well-documented. In the present study, we investigated the relevance of CD3+, CD8 +, CD45RO +, and FOXP3 + TILs to the prognosis and survival of patients with bladder cancer and the disease's clinical-pathological parameters.MethodsInfiltration of each subset was immunohistochemically evaluated in both stromal and intratumoral regions of tumor tissues from 85 patients with urothelial cell carcinoma of the bladder, with known survival.ResultsOur results indicated that intratumoral CD45RO+ lymphocytes were significantly higher in high-grade tumors than in low-grade ones (P = 0.028). The frequencies of intratumoral CD3+ (P = 0.002), CD8 + (P = 0.008), intratumoral (P = 0.002), and stromal (P = 0.017) CD45RO+ lymphocytes were also higher in patients with muscular invasion than those without invasion. The frequencies of intratumoral CD3+ (P = 0.043), CD8+ (P = 0.003), CD45RO+ (P = 0.023), and total CD45RO+ (P = 0.015), showed variation in patients with different T-stage, as well; mostly increased in T2 versus Ta and T1. Comparing patients in different stages revealed an increase in the frequencies of total CD3+ (P = 0.011), intratumoral CD3+ (P = 0.006), total CD8+ (P = 0.012), intratumoral CD8+ (P = 0.009) and stromal CD8+ (P = 0.034), as well as total and stromal CD45RO+ lymphocytes (P = 0.01 and P = 0.034, respectively) in stage II comparing to stage I, while the frequencies of stromal CD3+ (P = 0.077) and CD8+ (P = 0.053) cells tended to be decreased in stage III compared to stage II.ConclusionsWe collectively observed that the frequency of immune cells, especially CD45RO+, CD3+, and CD8+ lymphocytes, were significantly higher in early-progressed tumors. This observation could be explained by continuous and prolonged stimulation of immune cells with tumor antigens during tumor progression or an increase in the recruiting factors, especially in the early stages, to eliminate tumor cells. However, with tumor progression to the late stages, the inhibitory microenvironment provided by tumor cells suppresses or changes the functionality of the effector and memory immune cells to help tumor growth. However, more functional studies with larger sample sizes are needed to reveal the real status of the immune system in patients with bladder cancer.

A Mathematical Model of Aqueous Humor Production and Composition.

PurposeWe develop a mathematical model that predicts aqueous humor (AH) production rate by the ciliary processes and aqueous composition in the posterior chamber (PC), with the aim of estimating how the aqueous production rate depends on the controlling parameters and how it can be manipulated.MethodsWe propose a compartmental mathematical model that considers the stromal region, ciliary epithelium, and PC. All domains contain an aqueous solution with different chemical species. We impose the concentration of all species on the stromal side and exploit the various ion channels present on the cell membrane to compute the water flux produced by osmosis, the solute concentrations in the AH and the transepithelial potential difference.ResultsWith a feasible set of parameters, the model predictions of water flux from the stroma to the PC and of the solute concentrations in the AH are in good agreement with measurements. Key parameters which impact the aqueous production rate are identified. A relevant role is predicted to be played by cell membrane permeability to and , by the level of transport due to the Na+-H+ exchanger and to the co-transporter of Na+/K+/2Cl−; and by carbonic anhydrase.ConclusionsThe mathematical model predicts the formation and composition of AH, based on the structure of the ciliary epithelium. The model provides insight into the physical processes underlying the functioning of drugs that are adopted to regulate the aqueous production. It also suggests ion channels and cell membrane properties that may be targeted to manipulate the aqueous production rate.

Abstract 2032: Comprehensive spatial transcriptomic analysis of an integrated, diverse cohort reveals distinct molecular topographic patterns in triple negative breast cancer

Abstract Background: Triple negative breast cancer (TNBC) is an aggressive subtype of breast cancer that disproportionately affects African-American and black (AAB) women, who have higher rates of incidence and mortality. To understand this disparity, we must apply advanced molecular profiling to racially diverse cohorts. We propose the use of spatial transcriptomics to map whole transcriptome data to several thousand coordinates within tissue sections, for comprehensive and unbiased characterization of TNBC tumors and their microenvironment. Methods: We applied the Visium Spatial Gene Expression method (10x Genomics) to 28 fresh frozen TNBC tissue sections obtained from 14 patients (7 AAB and 7 Non-Hispanic White). Sections were stained with H&E, and regions of tumor and stroma were annotated by a pathologist. The assay yielded whole transcriptome data from over 36,000 spatially-defined features, averaging 1,380 features per section. Expression data from each sample was normalized and subject to dimensionality reduction and clustering analysis. All sections were also integrated into a single dataset, from which integrated clusters (IC) were defined. Several bioinformatics tools were used to annotate the data at the feature, cluster, and IC level. Join count statistical analysis was employed to quantify patterns of spatial aggregation or dispersion of ICs across samples. Results: Clustering analysis revealed that all samples exhibited a great deal of heterogeneity; most contained several transcriptionally distinct regions of both tumor and stroma. Application of the ESTIMATE gene sets at the feature level accurately defined tumor and non-malignant regions when compared to histopathological annotation. Gene set enrichment analysis at the cluster level allowed further classification of biological processes within each sample. Following integration of all 28 samples, 9 ICs were defined from transcriptional data. Although all ICs were present in all samples, IC5 - characterized by a strong hypoxic signature - was overrepresented in AAB samples. ICs were then mapped back to individual samples and subject to join count statistics. This revealed strong spatial autocorrelation within each IC, as well as significant spatial pairings of ICs. The fibrotic clusters IC3 and IC7 were strongly contiguous in all samples (z-score > 5), as were tumor clusters IC1 and IC4. Conversely, we found spatial exclusion between IC3 and tumor clusters IC5 and IC2. Conclusion: Our study provides novel evidence of spatially related populations across diverse TNBC samples. Our findings suggest that, while TNBC tumors are highly heterogeneous, they exhibit elements of a common spatio-transcriptional architecture. Moreover, this provides a new framework in which to evaluate transcriptional differences between racial groups within a broader spatial context. Citation Format: Rania Bassiouni, Michael Idowu, Lee D. Gibbs, Pamela J. Grizzard, Michelle G. Webb, Ashley Noriega, Valentina Robila, David W. Craig, John D. Carpten. Comprehensive spatial transcriptomic analysis of an integrated, diverse cohort reveals distinct molecular topographic patterns in triple negative breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2032.

Abstract 1202: A repository of PDX histology images for exploring spatial heterogeneity and cancer dynamics

Abstract Patient-derived xenografts (PDXs) recapitulate intratumoral spatial heterogeneity and simulate a tumor microenvironment in which human immune and stromal cells in the PDX are replaced over passages by murine cells partially lacking immune function. Histological imaging enables exploring the spatial heterogeneity and dynamics of cancer, stromal, and immune cell interactions as correlates of tumor stage and therapeutic response over passages. We created a repository of curated, haematoxylin and eosin (H&E) images as a community resource for addressing these questions. Images were generated at five sites within the NCI’s PDX Development and Trial Centers Research Network (PDXNet) and the NCI Patient-Derived Models Repository. Over 900 images, including 739 from PDXs and 190 from paired patients, are hosted on the Seven Bridges Genomics Cancer Genomics Cloud. They represent 42 cancer subtypes, including breast cancer (n=134), colon adenocarcinoma (COAD; n=94), pancreatic cancer (n=87), lung adenocarcinoma (LUAD; n=80), melanoma (n=71), and squamous cell lung cancer (LUSC; n=65). Paired human/PDX images are available for each of these cancers. Human and/or PDX images generated following patient treatment are available for 37 of the subtypes. Most images are from early passages (P0: 158; P1: 292; P2: 152; P3: 69; >P3: 55). Annotations include sex, age, race, ethnicity, and, for most images, pathological assessment of tissue-level percent cancer, stromal, and necrotic cell content (n=639) and tumor stage (n=650). RNA and exome sequencing data are available for 99 and 228 images, respectively, matched at the patient or sample level. Quality control was performed using HistoQC. Cells were segmented and labeled as neoplastic, necrotic, immune, stromal, or other using Hover-Net and predictions of total neoplastic cell area correlated with whole-slide pathological assessment of cancer cell percentage (COAD: r=0.51; LUSC: r=0.59). HD-Staining, another classification approach, was applied to a subset of images and our clinical annotations will facilitate validation of this and related methods. Features of 512 x 512 pixel tiles were computed using the Inception V3 convolutional neural network pre-trained on ImageNet. Unsupervised clustering of these features demonstrate inter-patient heterogeneity within pathologist-annotated tumor regions. A classifier developed using pathologist-annotated cancer, stromal, and necrotic regions and trained on the features in LUSC images (n=10 images) achieved a cross-validation accuracy of 96% for cancer tiles across (n=5) LUAD images. Accuracy was lower for stromal classification (90%), likely reflecting current limitations of our small, but growing, labeled training set. Our repository of clinically-annotated PDX H&E images should aid the community in studying spatial heterogeneity and in training deep learning-based image analysis methods. Citation Format: Brian S. White, Xingyi Woo, Soner Koc, Todd Sheridan, Steven B. Neuhauser, Akshat M. Savaliya, Lacey E. Dobrolecki, John D. Landua, Matthew H. Bailey, Maihi Fujita, Kurt W. Evans, Bingliang Fang, Junya Fujimoto, Maria Gabriela Raso, Shidan Wang, Guanghua Xiao, Yang Xie, Sherri R. Davies, Ryan C. Fields, R Jay Mashl, Jacqueline L. Mudd, Yeqing Chen, Min Xiao, Xiaowei Xu, Melinda G. Hollingshead, Shahanawaz Jiwani, PDXNet Consortium, Yvonne A. Evrard, Tiffany A. Wallace, Jeffrey A. Moscow, James H. Doroshow, Nicholas Mitsiades, Salma Kaochar, Chong-xian Pan, Moon S. Chen, Luis G. Carvajal-Carmona, Alana L. Welm, Bryan E. Welm, Michael T. Lewis, Ramaswamy Govindan, Li Ding, Shunqiang Li, Meenhard Herlyn, Michael A. Davies, Jack A. Roth, Funda Meric-Bernstam, Carol J. Bult, Brandi Davis-Dusenbery, Dennis A. Dean, Jeffrey H. Chuang. A repository of PDX histology images for exploring spatial heterogeneity and cancer dynamics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1202.

Abstract 2031: High plex proteomic prognostic marker discovery for patients with pancreatic adenocarcinoma using digital spatial profiling

Abstract Pancreatic Ductal Adenocarcinoma (PDAC) long-term outcomes following surgical resection remain poor, with only 20% of patients surviving 5 years post pancreatectomy. With its immune-privileged nature, starting from the early pre-neoplastic state, it appears to easily evade the antitumor immune response. Although there is rationale for targeting PDAC immune pathways, little benefit has been observed to date. The aim of the study was to interrogate the immune landscape of PDAC utilising the Nanostring GeoMx Digital Spatial Profiler (DSP), enabling high-plex proteomic characterisation whilst maintaining tumor microenvironment (TME) topographical features. We assessed tumor samples from 28 treatment-naive PDAC cases represented in a multiregional FFPE tissue microarray (TMA) for which extensive IHC, molecular, genomic characterisation and clinicopathological follow-up data is available. Multiplex immunofluorescent staining for nuclei, panCK, aSMA and CD3 was used to guide region selection using the GeoMx DSP system in multiple TMA cores. 60 immune markers were quantified simultaneously in different tissue compartments defined by immunofluorescence co-localization including tumor (panCK+), immune stromal (PanCK-) regions. The spatially informed assessment by DSP was validated by both regression and variable prognostication compared with IHC for stromal CD3, CD8 and CD68 in near serial TMA PDAC sections. Unsupervised analysis of proteome data (DSP) in the panCK-negative regions identified an immune poor group associated with shorter OS (13.0 versus 31.1mths, P=0.005). When transcriptomic subtype was considered, the checkpoint inhibitor B7-H3 was significantly upregulated in the squamous subtype tumours versus the classical group (Log2: 1.63, P = 0.001). Patients with high B7-H3 expression, using a median expression cut-off, were associated with shorter OS on multivariate analysis (HR: 4.16, P=0.01) a finding that was validated in an external cohort at the bulk transcriptome level. This pilot scale discovery study shows the potential of the Nanostring DSP technology to identify spatially-informed biomarkers with prognostic relevance in biopsy sized samples from treatment-naive PDAC. A number of relevant candidate immune predictors were identified in a spatial context, which are currently undergoing validation in a larger independent cohort and within the neoadjuvant setting. Future studies will apply this technology to pre and post treatment biopsy samples. Citation Format: Assya Legrini, Holly Leslie, Stephan Dreyer, Joanne Edwards, Andrew Biankin, David Chang, Nigel Jamieson. High plex proteomic prognostic marker discovery for patients with pancreatic adenocarcinoma using digital spatial profiling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2031.

Abstract 2034: Spatial transcriptomics identifies unique pharmacodynamic effects of checkpoint inhibitor treatment on the tumor microenvironment in NSCLC

Abstract Background: Immune checkpoint inhibitor (ICI) therapy has improved outcomes in non-small cell lung cancer (NSCLC), particularly in patients with high tumor infiltrating lymphocytes (TILs). Identifying the pharmacodynamic (PD) impact of ICI on immune cells in the tumor microenvironment (TME) can inform therapeutic development for patients with progressive disease. We sought to identify PD changes in the TME from patients before and after ICI therapy using a spatial transcriptomics platform that allows highly multiplexed profiling of 1,800 genes (Nanostring Digital Spatial Profiler [DSP]). Methods: Formalin fixed, paraffin embedded (FFPE) tumor tissue from 22 patients was sourced commercially. Patients were treated with neoadjuvant chemotherapy, then underwent a surgical tumor resection. After surgery an adjuvant chemotherapy was administered until progression; patients then received monotherapy ICI (nivolumab or pembrolizumab). Once progressed on ICI, another resection was performed. Patients were then treated with chemotherapy and followed until progression and/or death. The DSP technology was used to independently profile RNA from PanCK+ (tumor) and PanCK- (stroma) regions in the tissue based on fluorescence staining. Six circular regions of 600μm in diameter were selected for analysis using the GeoMx instrument; each area contained CD3+ cells. Additionally, immunohistochemistry for PDL1 and CD3 was performed, images were scored by a pathologist and analyzed with digital pathology algorithms. Results: Spatial transcriptomic analysis of pre- vs post-ICI treatment in stroma revealed significant increases in T cell genes (CD3E, TCRB, NKG7), T-cell activation markers (CD69, IFNG, OX40, GZMB, ICOS), costimulatory signaling (CD28, CD80, CD86), and immune checkpoints (CTLA4, TIGIT). Ayers et al., JCI 2017 identified 28 genes predictive of ICI response, 12 were significantly upregulated in the stroma post-ICI (26 were present in the DSP panel). We identified genes in the pre-ICI stromal microenvironment that were highly expressed in patients with partial response to ICI, the most significant genes were involved in immune regulation (IFIT1) and extracellular matrix remodeling (MMP3). In contrast, stromal genes highly expressed in patients with progressive disease were associated with T-cell maintenance (ETS1, IL7R, CCL19). Conclusions: In this study we used spatial transcriptomics to profile tissue regions where T-cells were in close proximity to tumor cells (microns). This focused, local PD analysis confirmed activated T-cells are present post-ICI. Ongoing studies in a larger cohort will be used to identify tumoral mechanisms of resistance and immune dysfunction to inform future therapeutics and combinations. Citation Format: Li Fan, Omar Jabado, Nora Pencheva, Patricia Coutinho de Souza, Brandon Higgs, Angelo Harris, Patrick Franken, Anantharaman Muthuswamy, Maria Jure-Kunkel, Suzana Couto, Kate Sasser, Mark Fereshteh. Spatial transcriptomics identifies unique pharmacodynamic effects of checkpoint inhibitor treatment on the tumor microenvironment in NSCLC [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2034.

Abstract 637: Spatial transcriptomics for FFPE characterizes the molecular and cellular architecture of malignant changes in pancreatic pre-malignant lesions

Abstract We have optimized an experimental and computational pipeline to adapt spatial transcriptomics (ST) approaches based upon the Visium (10x Genomics) technology to infer cellular composition and intercellular interactions of FFPE clinical specimens. We apply this technology to deliver an approach to examine pancreatic intraepithelial neoplasia (PanIN) to identify intrinsic and extrinsic mechanisms that are associated with the progression of these pre-malignant lesions to invasive carcinoma. Currently, most pancreatic cancers are diagnosed at an advanced stage that reflects in dismal survival rates and a better understanding of PanINs biology will provide valuable insights for early therapeutic interventions. Thus, we used PanINs as our model system to implement the FFPE ST workflow. Our workflow for FFPE ST analysis facilitates sectioning of small regions (5mm in diameter) from a paraffin block that are stained and imaged with H&E and concurrently measured for genome-wide transcriptional profiling. Subsequently, the image is used for automated cell annotation using an algorithm, CODA, trained to identify normal and neoplastic pancreatic cell types. CODA identified the normal pancreatic histological regions (ducts, acini, islets of Langerhans, stroma), as well as the neoplastic cells. This automated analysis enables isolation of specific spots for differential expression analysis to pinpoint the transcriptional changes that occur within neoplastic cells along ducts in PanIn and their changes between high-grade and low-grade lesions. The spatial gene expression analysis identified clusters that mapped to the cell types annotated by CODA and the marker genes of each cluster matched known markers for the correspondent cell type. Although PanINs are very small in size (< 1mm), we found specific clusters accurately mapped to these lesions in each sample. Overall, the spatial sequencing data presented enough depth and complexity to allow differential expression and pathway analysis. We observed a significant number of deregulated genes in PanINs compared to normal ducts. Some deregulated genes are known PanIN markers, but potential new markers were also identified. Moreover, the integration of CODA with gene expression changes enables us to verify that unique stromal regions annotated with CODA and associated with PanIns are in fact heterogeneous and formed by distinct cell subtypes. Altogether, our workflow combining automated cell annotation with STA from the same section provides a methodology to precisely examine the sample architecture while measuring heterogeneity at the transcriptional level. This combined approach can be applied to different FFPE tumor types to leverage the use of large bioarchives of samples not previously accessible to genome-wide spatial methods. Citation Format: Alexander T. Bell, Kohei Fujikura, Jacob Stern, Rena Chan, James Chell, Stephen Williams, Ashley Kiemen, Elizabeth M. Jaffee, Denis Wirtz, Laura D. Wood, Elana J. Fertig, Luciane T. Kagohara. Spatial transcriptomics for FFPE characterizes the molecular and cellular architecture of malignant changes in pancreatic pre-malignant lesions [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 637.

A machine learning model for separating epithelial and stromal regions in oral cavity squamous cell carcinomas using H&E-stained histology images: A multi-center, retrospective study

ObjectiveTissue slides from Oral cavity squamous cell carcinoma (OC-SCC), particularly the epithelial regions, hold morphologic features that are both diagnostic and prognostic. Yet, previously developed approaches for automated epithelium segmentation in OC-SCC have not been independently tested in a multi-center setting. In this study, we aimed to investigate the effectiveness and applicability of a convolutional neural network (CNN) model to perform epithelial segmentation using digitized H&E-stained diagnostic slides from OC-SCC patients in a multi-center setting. MethodsA CNN model was developed to segment the epithelial regions of digitized slides (n = 810), retrospectively collected from five different centers. Deep learning models were trained and validated using well-annotated tissue microarray (TMA) images (n = 212) at various magnifications. The best performing model was locked down and used for independent testing with a total of 478 whole-slide images (WSIs). Manually annotated epithelial regions were used as the reference standard for evaluation. We also compared the model generated results with IHC-stained epithelium (n = 120) as the reference. ResultsThe locked-down CNN model trained on the TMA image training cohorts with 10x magnification achieved the best segmentation performance. The locked-down model performed consistently and yielded Pixel Accuracy, Recall Rate, Precision Rate, and Dice Coefficient that ranged from 95.8% to 96.6%, 79.1% to 93.8%, 85.7% to 89.3%, and 82.3% to 89.0%, respectively for the three independent testing WSI cohorts. ConclusionThe automated model achieved a consistently accurate performance for automated epithelial region segmentation compared to manual annotations. This model could be integrated into a computer-aided diagnosis or prognosis system.