10X Genomics Research Articles

Abstract 2026: Single-cell RNA-sequencing analysis of tumor endothelial cells reveals suppression of specific endothelial subtypes by the combination treatment of Wnt/β-catenin signaling modulator E7386 plus lenvatinib

Abstract Background: Aberrant activation of Wnt/β-catenin signaling plays an important role in both carcinogenesis and modulation of the tumor microenvironment. E7386, a novel orally active antitumor agent, inhibits the interaction between β-catenin and CREB-binding protein (CBP) and thereby modulates Wnt/β-catenin signaling. We previously reported that E7386 in combination with lenvatinib (LEN), a multiple receptor tyrosine kinase inhibitor mainly targeting VEGFRs and FGFRs, showed a greater antitumor activity and reduction of tumor microvessels than each treatment alone in preclinical tumor models. Multiple clinical studies are in progress including the combination of E7386 plus LEN (NCT04008797). Here, we further tried to elucidate the mechanism underlying the antitumor and antiangiogenic effect of the combination of E7386 plus LEN by using single-cell RNA sequencing (scRNA-seq) of tumor endothelial cells (ECs). Methods: Mice bearing subcutaneous RAG (mouse renal cell carcinoma cell line) tumors were treated with E7386 at 25 mg/kg and/or with LEN at 10 mg/kg by daily oral gavage for 8 days. Tumor ECs (CD31+/CD45−) were isolated by magnetic cell separation and flow cytometry and were subjected to scRNA-seq using 10x Genomics pipeline. Bioinformatics data analysis was conducted by Clarivate Analytics. After quality control and filtering, ECs were gated by their expression of pan-EC markers and were further classified into 9 subtypes (tip, capillary, arterial, venous, proliferating, immature, interferon, activated artery, and lymphatic) according to previously reported markers for each subtype. Results: Among classified subtypes, proliferating and immature subtypes showed clear reduction in the percentage of total ECs with combination of E7386 plus LEN versus LEN monotherapy. In both proliferating and immature subtypes, a subset of Wnt/β-catenin downstream genes showed increased expression with LEN monotreatment. Conversely, these Wnt/β-catenin downstream genes were downregulated with the combination of E7386 plus LEN, suggesting that Wnt/β-catenin signaling may be associated with a drug resistance mechanism against LEN in these subtypes. Pathway analysis indicated that genes involved in cell proliferation and survival were downregulated by the E7386 plus LEN combination versus LEN monotreatment in both proliferating and immature subtypes. Conclusion: Taken together, these results suggest a role for E7386 in suppressing the activation of Wnt/β-catenin signaling induced by LEN monotreatment in specific EC subtypes; this activity leads to enhanced antiangiogenic and antitumor activity with E7386 plus LEN combination therapy in a preclinical tumor model. Citation Format: Masahiko Kume, Ryoga Ishida, Yoichi Ozawa, Jialing Shen, Hiroto Konishi, Megumi Kuronishi, Takayuki Kimura, Yusuke Adachi, Yu Kato. Single-cell RNA-sequencing analysis of tumor endothelial cells reveals suppression of specific endothelial subtypes by the combination treatment of Wnt/β-catenin signaling modulator E7386 plus lenvatinib [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2026.

Abstract 1502: Deciphering the spatial and temporal dynamics of premetastatic niche in lung mediated by Treg and myeloid cell interactions

Abstract Introduction: The lung is a frequent site for metastases, where pre-metastatic niches (PMN) form in anticipation of tumor cell arrival. A deeper understanding of the molecular and immune features underlying the PMN formation may facilitate developing innovative strategies to prevent and treat metastasis more effectively in its early stages. Methods: In this study, we developed mouse models by subcutaneously implanting the 344SQ cell line, derived from the metastasis of a genetically engineered mouse model (GEMM) harboring KrasG12D/+; p53R172H/+ mutations, which reliably metastasizes to the lungs. We collected paired samples of the subcutaneous tumors, peripheral blood mononuclear cells (PBMCs), and lung tissue at pre-metastatic, early-metastatic, and late-metastatic phases. These samples were subjected to single-cell RNA sequencing (scRNA-seq), single-cell TCR/BCR sequencing (scTCR/BCR-seq), and spatial gene expression profiling with 10X Genomics Visium technology. Results: ScRNA analysis revealed large-scale comprehensive longitudinal changes in cell components throughout the invasion and metastasis cascade, which were validated by spatial gene expression. Furthermore, analysis of Visium data deduces the spatial information of these dynamic changes.We identified a subset of tumor-associated Tregs (TA_Tregs, enriched in the subcutaneous tumor tissues) that existed in small amounts in the lung tissue in the pre-metastatic phase when there are no tumor cells detectable in the lung. These cells were characterized by high expression of Treg-associated suppressive genes (Tnfrsf9, Tnfrsf4, Tnfrsf18, Traf1, Nfkbia), inhibitory molecules (Ikzf2, Il2ra, Tight, Pdcd1), cytokines and chemokines (Ccr5, Ccr8, Ccl5, Cxcr6) and other Treg effector molecules (Id2, Hif1a, Rora, Il1r2, Itgav, Stat1, Icos, Areg). The scTCR-seq results and trajectory analysis revealed the TA_Tregs might derive from minor clones of lung resident Tregs (LR_Tregs) and play an active role in remodeling an immunosuppressive microenvironment by closely interacting with other cell components in the TME. We also observed a significant alteration in the ratio of macrophage subpopulations between Macro_Spic and Macro_Vcan. This ratio exhibited a continuous decrease in lung tissues, which could be an early indication of the initiation of pre-metastatic niche (PMN) conditioning. Conclusion: By simultaneously characterizing cellular gene expression, immune repertoires, and their corresponding spatial information, we revealed the dynamic changes of the lung tissue microenvironment (LTME) during from a normal state to pro-metastasis state at single-cell resolution. The LTME of metastatic organs may be primed before the establishment of metastasis. Citation Format: Yanhua Tian, Jian-Rong Li, Bo Zhu, Jared Fradette, Hong Chen, Chenyang Li, Leticia Rodriguez, Daniel Kraushaar, Luisa Solis, Alexandre Reuben, Ignacio Wistuba, Don L. Gibbons, Chao Cheng, Jianjun Zhang. Deciphering the spatial and temporal dynamics of premetastatic niche in lung mediated by Treg and myeloid cell interactions [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1502.

Abstract 579: Assessing the longitudinal efficacy of anti-PD-1 therapy in a murine non-small cell lung cancer model using single-cell and spatial transcriptomics

Abstract Lung cancer has a high incidence rate and is the main cause of cancer-related deaths with 127,070 fatalities expected in the U.S. this year. Non-small cell lung cancer (NSCLC) accounts for an estimated 85% of all lung cancer cases and is typically diagnosed at an advanced disease stage, where outcomes are often poor and treatment options are limited. Targeted therapies are only available to a minority of NSCLC patients with identifiable tumor-driving mutations. Anti-programmed cell death protein 1 (anti-PD-1) immunotherapy has become a first-line treatment strategy for patients without tumor-driving gene mutations. However, only 15-20% of NSCLC patients exhibit a positive response to anti-PD-1 therapy. Moreover, patients who initially respond to anti-PD-1 therapy still face the risk of primary or acquired resistance, potentially leading to cancer progression. This study leverages a murine lung cancer cell line representative of NSCLC isolated from a genetically engineered mouse model expressing KrasG12D and P53−/flox oncogenic mutations in club cells, employing a cell type-specific promoter (Scgb1a1). After successfully establishing the oncogenic cell line, the cells were subcutaneously injected into immunocompetent C57BL/6 mice and subjected to anti-PD-1 therapy for 3 weeks post-injection. Tumor growth curves revealed a significant size reduction in anti-PD-1-treated tumors when compared to the IgG isotype control group, as confirmed by a paired t-test analysis (p < 0.045). To evaluate the longitudinal dynamics of the tumor microenvironment, single-cell sequencing, and spatial transcriptomics were performed throughout effective anti-PD-1 treatment and following tumor relapse. Samples were collected at 2 time points during anti-PD-1 treatment, specifically at 1 and 3 weeks, as well as 3 weeks after the completion of the anti-PD-1 treatment. Subsequently, these samples underwent single-cell RNA sequencing and spatial transcriptomics using the 10x Genomics platform. Single-cell sequencing revealed modulations in immune recruitment throughout the treatment while relapsed tumors display elevated levels of myeloid gene expression and a shift to a highly fibrotic tumor phenotype post-anti-PD-1 therapy. These findings suggest that the established oncogenic cell line exhibits initial responsiveness to anti-PD-1 therapy in vivo before undergoing major changes within the tumor microenvironment. This study highlights the relevance and translational potential of our oncogenic cell line in advancing our understanding of NSCLC and its response to anti-PD-1 therapy. *This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. LLNL-ABS-857319. Citation Format: Nicole F. Leon, Oscar A. Davalos, David Baliu-Rodriguez, Deepa K. Murugesh, Aimy Sebastian, Nicholas R. Hum. Assessing the longitudinal efficacy of anti-PD-1 therapy in a murine non-small cell lung cancer model using single-cell and spatial transcriptomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 579.

Abstract 6925: Integrative single-cell tracking of genome evolution and tumor cell plasticity in small cell lung cancer (SCLC)

Abstract SCLC is the deadliest type of lung cancer, which is mostly diagnosed at late stages and accounts for 15% of pulmonary tumors. Even though chemotherapy is initially effective, patients with SCLC relapse quickly and develop resistant tumors. Genome studies revealed a universal loss of TP53 and RB1; recent transcriptional studies categorized SCLC based on the expression of 4 lineage transcription factors (TFs). We aimed to dissect how genomic cues impact transcriptional phenotypes, tumor cell plasticity and the dynamics of tumor and immune cell interactions, to thus decipher molecular mechanisms of phenotypic divergence and therapy resistance in SCLC. We performed genome sequencing and single cell transcriptome profiling of 55 tumors from patients with SCLC, including primary tumors, local and distant metastases and of paired relapsing tumors acquired after chemotherapy. Single cell sequencing was performed with the 10x Genomics platform and yielded > 200.000 cells. Transcriptome profiles at single cell levels revealed co-expression of at least two of the lineage TFs ASCL1, NEUROD1 or POU2F3 in each tumor. Copy number (cn) estimation for transcriptome data pointed to genomic events impacting transcriptional heterogeneity within the tumor. Furthermore, our data pointed to a marked intra- and interpatient, as well as to inter-metastatic heterogeneity. To specifically investigate how underlying genetic alterations affect molecular phenotypes, integrative studies across all patients were performed. This revealed shared transcriptional programs in tumors with amplifications of MYC family members; trajectory inference for cases with low and high-level copy gains of MYCN pointed to distinct transcriptional states. We performed multi-regional and longitudinal studies of matched patient cases to reconstruct the genomic patterns of clonal evolution, and projected genomic subsets to transcriptional profiles at single cell level, thus analyzing the effect of genome diversity on the transcriptional landscape. Additionally, tumor intrinsic profiles of heterogeneity were mapped to single cell data of patient tumors to elucidate the effects on tumor-immune cell interactions. All together our data points to a concerted regulation of a multitude of lineage factors and transcriptional programs in each SCLC tumor and provides a first comprehensive framework for the study of underlying genetic alterations that shape the transcriptional landscape and phenotypic plasticity in SCLC. Citation Format: Marcel Schmiel, Laura Kaiser, Maria Cartolano, Martin Peifer, Roman K. Thomas, Julie George. Integrative single-cell tracking of genome evolution and tumor cell plasticity in small cell lung cancer (SCLC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6925.

Abstract 1150: Unveiling the spatial landscape of tumor and stroma heterogeneity in colorectal cancer with spatial transcriptomics

Abstract Heterogeneity in colorectal cancer (CRC) manifests in the diversity and interactions among tumor cells, support stroma, and immune components, critically impacting patient prognosis and therapeutic efficacy. In this study, spatial transcriptomics (ST, 10x Genomics Visium Spatial Gene Expression) was performed on 40 retrospectively collected CRC samples, along with detailed clinicopathological parameters and long-term clinical outcomes, with the goal of unraveling the molecular topography of the tumor and surrounding stromal compartments, and associating their spatial interactions to clinical outcomes. Five primary RNA groups were identified through unsupervised clustering on ST data, correlating with tissue classes detected in H&E-stained images via a pre-trained deep learning model. Markers and pathways specific to these tissues were identified. Furthermore, immune cell infiltration in these regions was deconvoluted using reference single-cell RNA sequencing (scRNA-seq) data. Sub-clustering within the tumor groups revealed five major tumor subtypes, each characterized by a distinct set of differentially expressed marker genes. Enrichment analysis across these subtypes highlighted distinct activation and suppression in pathways including KRAS, p53, MAPK, and Integrin. Tumors exhibited intra- and inter-tumoral heterogeneity. Location-specific tumor subtypes were observed, with those in right-sided CRC displaying increased immune cell infiltration, linked to enriched pathways such as Type II Interferon Signaling and Cytokine Signaling. Conversely, left-sided CRC subtypes had reduced immune cell presence, and were characterized by enhanced pathways related to vascular development, Hypoxia, and p53 Signaling. Inferred clonal lineages from copy number variation (CNV) using ST suggested a spatially structured phylogenetic organization corresponding to transcriptional clusters. Additionally, the ST data unveiled distinct spatial distribution patterns across various cancer-associated fibroblast (CAF) phenotypes, such as matrix, inflammatory, vessel-associated, tumor-promoting, and antigen-presenting CAFs. Transcriptional patterns uncovered through scRNA-seq, including epithelial-mesenchymal transition (EMT) signatures and consensus molecular subtypes (CMS), displayed unique spatial layouts within the tumor microenvironment. To our knowledge, this is the most comprehensive study to date on spatial transcriptomic heterogeneity within CRC patients, revealing distinct tumor, stroma and immune compositions and spatial structures. This spatially detailed map provides novel insights into CRC heterogeneity, critical for advancing personalized treatment. Citation Format: Jie Zhou, Todd B. Sheridan, Sergii Domanskyi, Stephanie L. Cowles, Sunghee Park, Ilham Putra, Olga Anczukow, Jeffrey H. Chuang, Jill C. Rubinstein. Unveiling the spatial landscape of tumor and stroma heterogeneity in colorectal cancer with spatial transcriptomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1150.

Abstract 5006: Discovery of immunogenic neo-peptides from actionable RNA fusions for developing cancer vaccines

Abstract The global surge in cancer cases, with a staggering 20 million new instances reported in 2023 alone, underscores an urgent demand for advanced and reliable cancer therapeutics. Immunotherapies centered around neoantigens have emerged as a promising avenue for enhancing treatment efficacy and have become a focal point in cancer research. We hypothesize that the tumor cells present a rich neoantigenic repertoire and the immunogenic neoantigens arising from the junctions of proteins translated from chimeric RNAs present viable targets for peptide and mRNA vaccines, offering a promising approach to impede tumor progression. To validate our hypothesis, we acquired a dataset of RNA fusions detected through RNAseq of tissue extracted from patient-derived xenograft (PDX) models established by XenoSTART, a global non clinical oncology contract research organization. This comprehensive dataset encompassed 985 PDX models across 27 cancer types, revealing three fusions associated with the neuregulin-1 (NRG1) gene. NRG1 gene fusions have been identified as a clinically actionable target for multiple solid tumors as they result in ErbB-mediated pathway activation. CD74-NRG1 has emerged as the most frequently reported NRG1 fusion across many cancers such as pancreatic ductal adenocarcinoma, triple negative breast cancer and ovarian cancer. The CD74-NRG1 fusion was identified in two PDX models harboring lung cancer (ST2891 and ST3204) and in an ovarian cancer model (ST088). The lung cancer PDXs displayed a notable abundance of fusion junction-crossing reads. We generated the CD74 (Exon 1-6) -NRG1 (Exon 6-12) fusion transcript model, where the first 6 exons of CD74 fused to exon 6 of NRG1 leading to an in-frame fusion. Validation of this fusion in PDX tissue was accomplished through PCR and Sanger sequencing of the PCR product. We assessed the affinity of the 9-mer neo peptide sequences formed by the translated junction of CD74-NRG1 to Major Histocompatibility Complex (MHC) Class I molecules using the in-silico prediction pipelines MHCNuggets and MixMHCPred-2 and found HLA-A*68:23 to be the MHC allele with highest binding affinity. HLA-Arena was used to investigate the binding affinity between the peptides and the MHC molecules via molecular docking, which revealed 4 peptides to be strong binders to HLA-A*68:23.The immunogenicity of these neoantigenic peptides to HLA-A*68:23 matched Peripheral Blood Mononuclear Cells (PBMCs) will be assessed by the IFN-γ Enzyme Linked Immunosorbent Spot (ELISpot) assay. To elucidate the specific T-cell clonotypes responding to the neo peptides, the 10X Genomics single-cell 5’ Gene Expression Assay, coupled with T Cell Receptor mapping, will be employed. The objective of the study is to establish a robust combinatorial therapeutic strategy leveraging peptide and mRNA vaccine technology to support an effective framework for personalized cancer treatment. Citation Format: Sakuni Rankothgedera, Micah Castillo, Shiyanth Thevasagayampillai, Cole Woody, Aaranyah Kandasamy, Armando Diaz, Michael Wick, Preethi Gunaratne. Discovery of immunogenic neo-peptides from actionable RNA fusions for developing cancer vaccines [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5006.

Abstract 780: Normal breast tissues harbor rare populations of aneuploid epithelial cells

Abstract Although aneuploidy is a hallmark of invasive breast cancers, the presence of aneuploid cells in breast tissues of disease-free women is poorly understood. Furthermore, the expansion of aneuploid cells and the specific copy number events that they harbor, as well as their relationship to invasive breast cancers has not been characterized in large cohorts of women. In this study, we obtained viable cell suspensions from 31 disease-free women that underwent reduction mammoplasty surgeries and fresh tissue collection in the HBCA project along with the collection of clinical metadata. The viable cell suspensions from 31 women were enriched for epithelial cells and used to perform high-thoughput nanowell scDNA-seq. In total, 55,321 cells were analyzed. The normal breast tissue samples were also embedded into OCT and frozen to perform Spatial Transcriptomics (Visium 10X Genomics) profiling and embedded into FFPE to perform histopathological analysis. Our data shows that all women harbored rare populations of aneuploid epithelial cells (mean 4.3%) that increased in frequency and their number of copy number aberrations with age. Many aneuploid epithelial cells (mean 79%) in normal breast tissues underwent clonal expansions of their genomes and harbored chromosomal aberrations detected in invasive breast cancers (gain of 1q and loss of 7q, 10q, 16q). Spatial mapping showed that aneuploid cells are localized in focal regions of ductal and lobular epithelial structures with normal histopathological features. Collectively, these data show that aneuploid epithelial cells exist in the normal breast tissues of disease-free women and increase with age. In many women, the aneuploid cells expand and harbor CNA events associated with invasive breast cancer. Citation Format: Yiyun Lin, Junke Wang, Kaile Wang, Shanshan Bai, Aatish Thennavan, Runmin Wei, Yun Yan, Jianzhuo Li, Heba Elgamal, Emi Sei, Anna Casasent, Mitchell Rao, Chenling Tang, Jessica Montalvan, Chandandeep Nagi, Sebastrian Winocour, Bora Lim, Alastair Thompson, Nicholas Navin. Normal breast tissues harbor rare populations of aneuploid epithelial cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 780.

Abstract 5067: Exploring the genomic landscape of de novo neuroendocrine prostate cancer: Insights for innovative therapies from spatial gene expression analysis

Abstract Introduction & Objectives Neuroendocrine prostate cancer (NEPC) accounts for less than 1% of all prostate cancers but has a worse prognosis than androgen receptor-positive prostate adenocarcinoma (ARPC). Rare cases exist where both de novo NEPC and ARPC are found in the same tissue sample. Our study examines one such unique case: a 78-year-old male treated at Ehime University Hospital who was diagnosed with both ARPC and adjacent de novo metastatic NEPC. This case provides an excellent opportunity for studying gene expression in these two types of prostate cancer. Materials & Methods We used Visium CytAssist Spatial Gene Expression technology from 10x Genomics to analyze formalin-fixed, paraffin-embedded (FFPE) tissue samples from this unique patient. Results We observed elevated expression levels of specific neuroendocrine markers like PEG10, DLL3, NCAM1, SYP, and CHGA in de novo NEPC regions. In contrast, androgen receptor markers such as KLK3, ACPP, TMPRSS2, and NKX3.1 were more prevalent in ARPC areas. Furthermore, genes like Chek1, BRCA1/2, TOP2A, and FANCA were also elevated in NEPC, suggesting that PARP and TOP2 inhibitors could be effective. Expression levels of Rbfox3 and SFRTM2 were lower in NEPC areas, implying possible roles in epithelial-mesenchymal transition (EMT) and Wnt signaling pathways. Additionally, fibrotic markers like HGF, HMOX1, ELN, and GREM1 were upregulated, suggesting that de novo NEPC tumors might be 'immune cold,' thus requiring treatments to boost the effectiveness of immune checkpoint inhibitors. Conclusion The data from additional cases could open the door for new treatment options for NEPC and improve survival rates for castration-resistant prostate cancer patients. Our research serves as a pioneering application of Visium CytAssist on FFPE samples and is the first to examine gene expression in both ARPC and de novo NEPC. A subset of these findings was published in the International Journal of Molecular Science (IJMS) in May 2023. Citation Format: Ryuta Watanabe, Noriyoshi Miura, Mie Kurata, Riko Kitazawa, Tadahiko Kikugawa, Takashi Saika. Exploring the genomic landscape of de novo neuroendocrine prostate cancer: Insights for innovative therapies from spatial gene expression analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5067.

Abstract 1273: Tumor progression and tumor microenvironment of pancreas cancer ascites revealed by scRNA-seq and spatial transcriptomics

Abstract Peritoneal metastasis of pancreas cancer is one of the ultimate modes of spread in the cancer. Little is known about how cancer cells spread into abdominal cavity and what is different from hematogenous metastasis. It is also not well characterized how tumor microenvironment prevent cancer cells from spreading in abdominal cavity. We performed single-cell RNA-seq analysis and special transcriptomic analysis to clarify the progression and environment of the cancer. We collected and processed ascites samples of 20 PDAC samples from 19 patients. Sixteen ascites samples were obtained from abdominal paracentesis and two cases in the postmortem period. For one patient ascites was collected both pre and postmortem. All but one patient received chemotherapy prior to ascites collection. Routine clinical cytology evaluation of ascites samples from 17 patients indicated it was malignant in 11 patients and nonmalignant in six patients. All ascites samples were purified by gradient centrifuge using Ficoll-Paque Plus to remove erythrocytes and necrotic debris. For the platform, we adopted the 10x Genomics Chromium Single Cell Gene Expression platform for single-cell RNA-seq library preparation. Spatial transcriptomics was performed for eight FFPE samples from the two autopsy patients with paired cancer ascites using 10x Visium. The CytAssist was used to transfer transcripts of pre-sectioned and pre-stained tissues with 11 × 11 mm Visium Capture Area slides. The sample locations were metastatic tumors to bowel wall, rib bone, mesentery, iliac nerve, diaphragm, peritoneum, pelvis, and primary pancreas tumor. A total of 248,263 cells were analyzed in single-cell RNA-seq. Cell population in pancreas cancer ascites consisted of macrophages, lymphocytes, dendritic cells, plasma cells, plasmacytoid dendritic cells, mesothelial cells, and cancer cells. Cancer cells were detected in twelve samples whereas no cancer cells were detected in eight samples. Gene ontology and gene set enrichment analyses showed macrophages in ascites with cancer cells involved neutrophil activation while those in ascites without cancer cells retained antigen presentation capacity and inducing interferon gamma pathway. Mesothelial cells in ascites with cancer cells showed more apoptosis and lymphocyte proliferation while those in ascites without cancers cells triggered fibrosis and mitosis. These data suggest that macrophage and mesothelial cells could be categorized into anti-tumor cells and tumor associated cells. Ongoing investigations are exploring ligand-receptor relationships. We are going to compare the expression profile of tumor cells in ascites fluid and in tissue from various organs by integrating single cell RNA-seq data and special transcriptomics data. Citation Format: Shigeaki Umeda, Elias-Ramzey Karnoub, Fiyinfolu Balogun, Alejandro Jiménez-Sánchez, Christine A. Iacobuzio-Donahue. Tumor progression and tumor microenvironment of pancreas cancer ascites revealed by scRNA-seq and spatial transcriptomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1273.

Abstract 873: Elucidating immune-related gene transcriptional programs via factorization of large-scale RNA profiles

Abstract Background: Despite the recent success in Immune checkpoint blockade (ICB) therapy, limited patients benefit due to immune-related adverse events or treatment resistance, especially in solid tumors. To overcome these hurdles, understanding the molecular mechanisms behind treatment response is essential. However, the current lack of immune gene transcriptional programs (GTPs) hampers data-driven immunological discovery. Constructing immunity-specific knowledgebases with rigorously curated gene sets (irGSs) and rich immunological language can unlock new insights, enabling interpretation of high-throughput immune microenvironment profiling studies and fostering personalized, effective cancer treatments. Methods: We collected 83 BulkRNAseq datasets from the ImmuneSigDB. These datasets contain 1826 samples challenged with infections, cytokines or immunological perturbations of different kinds and magnitude, possessing yet-to-be discovered immune functions that lie beneath the transcriptomic profiles. Using non-negative matrix factorization (NMF), we identified gene sets with coordinated expression. We used CITE-seq and selected scRNAseq to annotate the immunological functions of each gene set and validated the clinical utilities of these gene sets from different aspects using data from Cancer Genome Atlas Program (TCGA) pan-cancer, ICB cohorts and a 10X Genomics Visium FFPE Human Breast Cancer spatial slide. Results: We presented 19 lymphoid-data derived and 9 myeloid-data derived gene sets (irGSs), encompassing a diverse array of immune functions. Through extensive validation using BulkRNAseq, scRNAseq and spatial transcriptomics data, we revealed six irGSs-defined pan-cancer microenvironment subtypes in TCGA with significantly distinct survival patterns regardless of cancer types. Moreover, irGSs are associated with T cell exhaustion phenotypes, which in turn well predicted (classification accuracy of 72%) Nivolumab response using 104 treatment-naïve melanoma patients. Lastly, irGSs simultaneously isolated out tumor regions and highlighted immune infiltration with high specificity in breast cancer spatial transcriptomics H&E image. Conclusions: These pioneering gene sets, originally derived from non-cancerous experiments, hold tremendous promise for cancer research across diverse contexts. Similarities between cancerous and sepsis immune microenvironments underscore their wide applicability. By studying gene set activities, immunologists can gain profound insights into cancer survival drivers, unravel the intricacies of ICB treatment mechanisms, and potentially conquer therapeutic resistance. These translational utilities herald a transformative era in cancer immunotherapy and open new frontiers in the fight against cancer. Citation Format: Shan He, Vakul Mohanty, Matthew Gubin, Hind Rafei, Rafet Basar, Merve Dede, Xianli Jiang, Yukun Tan, Maura Gillison, Katayoun Rezvani, Weiyi Peng, Ken Chen. Elucidating immune-related gene transcriptional programs via factorization of large-scale RNA profiles [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 873.

Abstract 6805: CD81 and CD82 in tumor infiltrating lymphocytes within the NSCLC tumor microenvironment play a crucial role in T cell activation and cytokine production

Abstract Background: The role of tetraspanins, particularly CD81 and CD82, in T cell activation, shedding light on their contribution to anti-tumor effects in the tumor microenvironment. The findings provide insights for potential applications of these tetraspanins in immunotherapy. Method: The study employed 10x Genomics for single-cell RNA sequencing and processed the data using Seurat. It focused on the tumor microenvironment, categorizing cell types, especially TIL subsets. T cells were sorted based on CD81 and CD82 expression and activated, and their activation markers and cytokine production were analyzed. Bulk RNA sequencing was performed on both sorted T cells and tumor samples, and the TCR repertoire was examined. Cytotoxicity of CD81 and CD82 transduced cells was measured using the xCELLigence RTCA HT system. Result: Tetraspanins have been linked to T cell co-stimulation and activation, with CD81 and CD82 playing a significant role in TCR signaling and activation. Single-cell and bulk RNA sequencing data from lung cancer patients reveal high expression of CD81 and CD82 in the tumor microenvironment, especially in T cells, correlating with T cell activation and cytokine production. In vitro experiments confirm the activation and effector cytokines, especially IFN-γ, TNF-α, and IL-2, producing potential of T cells expressing CD81 and CD82, suggesting their importance in anti-cancer immunity. Overexpression of CD81 and CD82 enhances T cell cytotoxicity against cancer cells, further supporting their role in T cell function. Despite some limitations, these findings shed light on the significant impact of tetraspanins in modulating immune responses, particularly in the context of cancer immunotherapy. Conclusion: We report that CD81 and CD82 serve as specific markers for activated T cells within the TME. CD81 and CD82 are correlated with the activation and regulation of T cells, and they may also mediate the regulation of relevant cytokines. Overall, our findings provide better insights into the molecular mechanisms regulating TILs present in the TME of non-small cell carcinoma and propose CD81 and CD82 as associated markers. These discoveries could serve as a foundation for the development of new immunotherapies targeting the TME. Citation Format: Kwangmin Na, Seul Lee, Dong Kwon Kim, Young Seob Kim, Joon Yeon Hwang, Seong-san Kang, Sujeong Baek, Chai Young Lee, Seung Min Yang, Yu Jin Han, Mi hyun Kim, Heekyung Han, Youngtaek Kim, Jae Hwan Kim, Seunghyun Jeon, Youngseon Byeon, Jii Bum Lee, Sun Min Lim, Min Hee Hong, Kyoung-Ho Pyo, Byoung Chul Cho. CD81 and CD82 in tumor infiltrating lymphocytes within the NSCLC tumor microenvironment play a crucial role in T cell activation and cytokine production [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6805.

Abstract 2150: Spatial transcriptomic analysis of histologically matched regions adjacent to and within prostate cancer reveals RNA expression differences between African Americans and European Americans

Abstract Introduction. Prostate cancer (PCa) of American men of African ancestry (AA) is diagnosed at an earlier median age and more advanced stage with a poorer prognosis and significantly higher mortality than in Americans of primarily Northern European ancestry (EA). The mechanism of how ancestry contributes to aggressive PCa in AA patients is not well understood. Given the inherent heterogeneity and multifocality of prostate cancer, Spatial transcriptomics is of particular interest because it allows distinct cell populations within the tumor microenvironment to be characterized, and their possible effects on local tumors to be identified. Here we compare transcription between four AA and four EA patients in regions of peritumoral stroma, inflammation, Benign Prostatic Hyperplasia (BPH), and Gleason score patterns 3 (G3) and 4 (G4). Methods: Formalin fixed, paraffin embedded (FFPE) prostatectomy tissue blocks were obtained from 8 patients (4 AA and 4 EA) prior to treatment of the patients and matched for clinical variables (i.e., biochemical relapse, age, Gleason score, and tumor stage). 10um tissue sections were mounted onto Visium Spatial Gene Expression Slides and processed following 10x Genomics protocol. Sample counts were processed through Space Ranger. 18000 unique genes, spatial barcodes, and clinician guided annotations were imported to the R program Seurat. Data was processed using SCTransform. Significance of each transcript differences between each tissue were calculated. Results: Digital histologic slides of prostate cancer specimens from AA (n=4) and EA (n=4) patients were annotated by a urologic pathologist to identify regions of peritumoral stroma, inflammation, BPH, G3, and G4. Each spot was interrogated for the expression of 18000 transcripts, with an average of 17796 being detected per spot. Among the largest differences observed between AA and EA, G3 and G4 had distinct transcriptome profiles in the four AA as compared to G3 and G4 in four EA. Also, when differential expression (padj < 0.05) in peritumoral stroma was compared to G4 in AA, we identified 290 differentially expressed genes, whereas the same comparison in EA cohorts identified 600 differentially expressed genes. Conclusion: Spatial transcriptomic analysis revealed differences in gene expression across multiple tissue types in prostate cancer patients of different ancestries. Citation Format: Vinay Kumar, Dominic Castaneda, James A. Nguyen, Giovanna A. Giannico, Michael M. Ittmann, Patricia Castro, Omid Yazdanpanah, Arash Rezazadeh, Tara S. Jennings, Michael Mcclelland, Farah Rahmatpanah. Spatial transcriptomic analysis of histologically matched regions adjacent to and within prostate cancer reveals RNA expression differences between African Americans and European Americans [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2150.

Abstract 6944: Transcriptional programs underpinning reoccurring cancer cell states, plasticity and tumour phylogenies in clear cell kidney cancer

Abstract Intratumor heterogeneity (ITH) is a substrate for tumor evolution and confers amenability to stressors and therapies. Sources of ITH include subclonal mutations, microenvironmental factors, and plasticity whereby a cell has the ability to adopt different cell states. Clear cell kidney cancer (ccRCC) has a well-characterized genetic initiating event - VHL inactivation - resulting in activation of HIF-1 and HIF-2 transcription factors. Although VHL inactivation is consistently clonal in nature, ccRCC nonetheless exhibits extensive genetic and morphological ITH. This makes ccRCC an excellent model in which to study ITH. We performed single cell RNA-seq on multiregional samples from kidney tumors and macroscopically normal kidney tissue. We obtained 55 samples from 16 tumors and used the 10x Genomics platform to obtain gene expression data from ~150,000 single cells, including ~80,000 cancer and normal epithelial cells. We identified three cancer cell states, co-existing in all clear cell kidney cancer tumors analyzed and marked by different gene expression programs: 1) differentiated-like cells marked by proximal tubule cell genes characteristic of the cancer cell of origin; 2) epithelial-mesenchymal transition (EMT)-like cells and 3) injured-like cells. Differentiated markers and EMT markers were associated with good and poor prognosis, respectively, indicating that these cell subpopulations are clinically important. Notably, these cell states are also present in the normal proximal tubule although in different proportions to the tumor samples, which contain more cells in the EMT-like and injured-like states. This suggests that the core features of ITH are inherent to the cell of origin of the cancer, but that the cell state trajectories are dysregulated in the tumor. Furthermore, in tumors these states exist as a continuum, whereby transitional cells in between states appear to exist, suggesting that cancer cells are continually and dynamically transitioning between states. Lastly, we used chromosomal copy number variation (CNV) patterns to study genetic ITH. We examined the effect of 14q loss, which is associated with ccRCC metastasis, on gene expression. We found that 14q loss is associated with downregulation of HIF-1 transcriptional activity together with a reciprocal increase in HIF-2 activity, MYC activity and oxidative phosphorylation, highlighting potential mechanisms by which 14q loss may driver ccRCC metastasis. Citation Format: Olivia Lombardi, Ran Li, Lisa Browning, Faiz Jabbar, Hannah Evans, Peter J. Ratcliffe, David R. Mole. Transcriptional programs underpinning reoccurring cancer cell states, plasticity and tumour phylogenies in clear cell kidney cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6944.

Abstract 313: Benchmarking immune repertoire sequencing technologies

Abstract In the past decade, single cell gene expression technologies have transformed basic and translational research and are more frequently found in the toolbox of immunologists alongside the standard bulk B-cell receptor (BCR) sequencing and transcriptome profiling assays. While conventional bulk BCR assays capture the frequency of single chains, single-cell immune profiling reveals the frequency of paired heavy and light chains along with the cells transcriptional profile. Here we present the results of a single-blind benchmarking experiment conducted in collaboration with the U.S. Food and Drug Administration (FDA) that compares the performance of standard bulk DNA and RNA BCR repertoire sequencing assays, with the 10x Genomics Single Cell Immune Profiling solution. The FDA selected nine unambiguously monoclonal BCR-expressing cell lines to generate a reference mixture that was shared with each study participant, without revealing the input proportions. This reference cell line mixture was also diluted to 10% and 1% with B-cells sorted from peripheral blood mononuclear cells (PBMC) to further test the limits of clonotype detection. For each experiment, we targeted ~10,000 cells and in aggregate generated a paired gene expression and V(D)J dataset from ~100,000 cells. BCR clonotypes were detected in 78% of the captured cells across all samples profiled. Each of the 9 top unique clonotypes observed in our dataset corresponded to a cell line in the reference mixture, allowing us to decode the mixing percentages of each BCR-expression cell line. In addition, we were able to identify rare cell line clonotypes to as low as 0.01% in the dilution experiments. Furthermore, the multimodal nature of the single cell data also resolved the transcriptional profile of each cell line into its own cluster thus enabling independent verification of clonotype proportions. In summary, we have generated a resource for the immunology community to evaluate and compare data generated with different repertoire sequencing technologies. Citation Format: Shamoni Maheshwari, Carolyn Morrison, Sarah Taylor, BCR-SEQC consortium. Benchmarking immune repertoire sequencing technologies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 313.

Abstract 2022: E7386, a Wnt/β-catenin signaling modulator, suppresses the differentiation of regulatory T cells in combination with lenvatinib plus anti-PD-1 antibody

Abstract Background: E7386, a selective inhibitor of the interaction between β-catenin and CREB binding protein (CBP), showed antitumor activity in preclinical tumor models with aberrant activation of the Wnt/β-catenin signaling pathway. Previously, we reported the enhanced antitumor activity of a triple combination (E7386 in combination with Lenvatinib [LEN, a multiple receptor tyrosine kinase inhibitor mainly targeting VEGFRs and FGFRs] plus anti-PD-1 antibody) versus a 2-drug combination (LEN plus anti-PD-1 antibody). In this study, we further investigated the immunomodulatory functions of E7386 as a triple combination by single-cell RNA-sequencing (scRNA-Seq) analysis and subsequent functional assays. Methods: Mice bearing subcutaneous RAG (mouse renal cell carcinoma) tumors were randomized into 4 groups (n=6 in each group), including ⑴ Nontreatment, ⑵ E7386 at 25 mg/kg (PO, QD), ⑶ LEN at 10 mg/kg (PO, QD) plus anti-PD-1 antibody at 10 mg/kg (IP, BIW), and ⑷ triple combination. Tumor tissues were collected after a 2-week treatment period. CD45+ immune cells were isolated from the tumors by magnetic cell separation (MACS) and subjected to scRNA-Seq using 10x Genomics pipelines. Bioinformatics data analysis was conducted with Clarivate Analytics. To evaluate the effects of E7386 on Treg differentiation, an in vitro-induced regulatory T cells (Tregs) differentiation assay was conducted. Naïve CD4+ T cells were isolated by MACS and plated under iTreg differentiation conditions (stimulation of TGF-β and IL-2) with or without treatment of E7386. Induced Tregs and conditional culture medium were collected for evaluation of Treg differentiation, including gene expression and protein analysis. Results: CD45+ immune cells were clustered into 14 cell-type groups. Increased proportion of monocytes, NK cells, Tregs, CD4+ T cells, CD8+ T cells, eosinophils, and neutrophils were observed after LEN plus anti-PD-1 treatment. Among these subpopulations, Treg were most significantly suppressed by E7386 plus LEN plus anti-PD-1. Gene expression analysis revealed that the triple combination suppressed the activated gene expression of the Stat5/Foxp3 axis in CD4+ T cells, particularly Tregs, induced by LEN plus anti-PD-1 treatment. Consistently, the iTreg differentiation assay confirmed the suppressive function of E7386 on the Treg differentiation rate. Multiple subsequent in vitro assays also demonstrated the inhibitory effects of E7386 on TGF-β secretion and the expression levels of TGF-β downstream target genes. Conclusion: In summary, these data suggest a suppressive role of E7386 on Treg differentiation, which leads to antitumor activity of E7386 in combination with lenvatinib and anti-PD-1 antibody in a preclinical tumor model. Citation Format: Jialing Shen, Ryoga Ishida, Masahiko Kume, Takayuki Kimura, Yusuke Adachi, Yu Kato. E7386, a Wnt/β-catenin signaling modulator, suppresses the differentiation of regulatory T cells in combination with lenvatinib plus anti-PD-1 antibody [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2022.

Abstract 3708: Characterization of EpCAM-positive cells for specific detection of disseminated cancer cells in bone marrow of early breast cancer patients

Abstract Obviously, breast cancer cells that disseminated from the primary tumor before surgical resection are responsible for metastasis. However, detection of disseminated cancer cells (DCC) in breast cancer patients during minimal residual disease is still difficult, impeding the study of early systemic cancer and the extraction of clinically relevant diagnostic or therapeutic information. Others and we have shown previously that EpCAM is a suitable DCC detection marker, however not entirely specific in bone marrow. To improve DCC detection, we here compared different methods to enrich DCC exploiting information from our ongoing single cell RNA-seq study. Based on our ongoing scRNA-seq study (see poster of Elia Raya), we generated a qPCR assay to classify candidate DCC based on an expression signature of four genes. This assay categorizes plasma cell-like EpCAM+ cells from healthy donors and candidate DCC from breast cancer patients with high fidelity. With this assay at hand, we compared two enrichment strategies: depletion of CD11b, CD319 and Glycophorin A positive cells from density gradient enriched mononuclear cells (MNC) vs. positive selection of EpCAM+ cells and subsequent addition of markers that separate out DCC among EpCAM+ cells. For this we applied 10x genomics profiling of EpCAM+ cells from bone marrow. We depleted and screened the bone marrow of 221 patients suffering from non-metastatic early breast cancer for EpCAM-positive cells. From 87 patient samples we could isolate EpCAM-positive cells. We performed single cell RNA-sequencing on 64 EpCAM-positive cells from 54 patients of which 84% were classified as cancer cells and 16% of bone marrow origin. Applying the selected gene signature, 76% of isolated cells were classified as DCC, 16% were undefined and 8% as non-cancer bone marrow cell (NCC). Upon positive selection of EpCAM+ cells implementing additional markers, we isolated 53 cells from 14 patients. Of these, 68% showed a DCC or DCC-like-signature, 15 % a NCC and 17% an undefined profile. To get a better understanding of these EpCAM-positive confounder populations we currently perform a scRNA-seq profiling using 10X on the enriched EpCAM+ cells. Our data shows that EpCAM is a promising marker to identify DCCs in the bone marrow of early cancer patients, but it also indicates that we find several groups of confounder cells. To differentiate between these cell types during staining and in our qPCR assay we are currently testing additional markers. Citation Format: Jonas Roth, Elia Raya, Sarah Hücker, Huiqin Koerkel-Qu, Stephan Seitz, Christoph Klein. Characterization of EpCAM-positive cells for specific detection of disseminated cancer cells in bone marrow of early breast cancer patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3708.

Abstract 6108: Investigation of breast tumor biology and microenvironment in women of African descent using a single cell multiomic approach

Abstract Women of African descent are at an increased risk of developing and dying from aggressive subtypes of breast cancer. A connection between aggressive disease and Western Sub-Saharan African ancestry has been postulated, but it remains largely unknown to what extent breast cancer in Africa is reminiscent of breast cancer in U.S. African American (AA) women who experience disproportionately high mortality rates. We performed ATAC- and RNA-sequencing on 9 human triple-negative breast cancer cell lines of U.S. origin and discovered that African ancestry influences the chromatin landscape, leading to disparate transcription factor (TF) activity and downstream gene expression patterns indicative of an aggressive tumor biology. Here, we describe an ambitious study that employs single-nucleus (sn) ATAC- and RNA-sequencing (snMultiome) of frozen breast tumors to characterize chromatin accessibility and gene expression patterns with single-cell resolution in AA (n=33), Kenyan (n=25), and European American (EA, n=24) women in relation to genetic ancestry, risk factor exposures, clinical characteristics, and 5-year survival. To achieve this, we successfully isolated intact, high-quality single nuclei from archival frozen breast tumor tissue through an optimized combination of enzymatic digestion and automated tissue homogenization. We performed snMultiome sequencing of 82 tumors using the 10x Genomics platform. Following filtering, normalization (SCT for snRNA; LSI for snATAC), peak calling (MACS2), and integration (Harmony), our dataset includes a total of 296,557 nuclei. Cancerous (163,419 nuclei) and non-cancerous (133,138 nuclei) cells were distinguished based on DNA copy number (CopyKat). Within the microenvironment, 11 major immune, epithelial, and stromal cell types were successfully annotated, exhibiting distinct patterns by population group (e.g. AA tumors showed markedly increased abundance of myeloid and T-cells, while Kenyan tumors showed increased abundance of pericytes and fibroblasts). A large number of enriched TFs within each cell type varied significantly by population group, suggesting distinct chromatin accessibility patterns related to genetic ancestry. Within cancerous cells, striking intra- and inter-tumoral heterogeneity was observed by genetic ancestry even within molecular subtype groups. Current efforts focus on in-depth molecular characterization of ancestry- and risk factor-related differences in the tumor epithelium and microenvironment and distinct signatures present in lethal disease. This project holds the potential to yield crucial insights into how ancestry or other factors may influence the etiology of different breast cancer subtypes, as well as produce clinically actionable biomarkers and therapeutic targets to enhance precision medicine within patient populations at high risk for aggressive disease. Citation Format: Alexandra R. Harris, Huaitian Liu, Brittany Jenkins-Lord, Tiffany H. Dorsey, Francis Makokha, Shahin Sayed, Gretchen Gierach, Stefan Ambs. Investigation of breast tumor biology and microenvironment in women of African descent using a single cell multiomic approach [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6108.

Abstract 338: Unraveling layers of proteogenomic complexity in cancer through multiomic exploration of archived FFPE tissue

Abstract Formalin Fixed Paraffin Embedded (FFPE) tissues are a staple in clinical diagnostics associated with solid tumor. The preserved tumor tissue can be interrogated using classical histopathology and complemented with molecular approaches to elucidate cancerous SNPs, indels, structural variants, repeat expansion, copy number expansions, and mutational burden. Increasingly, approaches are being deployed to interrogate tissue not only through individual datasets but through integrated approaches that provide spatial information that defines the tumor microenvironment. Here, we illustrate using FFPE blocks from various tissues a multiomic workflow that allows for deep exploration of the molecular underpinnings of cancerous tissue. FFPE blocks were serially sliced into various FFPE slides with a single slide H&E stained. Individual slides were then utilized to explore the genome, epigenome, single cell RNA-seq, and digital spatial profiling. Genome information was captured using hybrid capture based approaches (whole exome sequencing and targeted cancer panel) followed by deep NGS on an Illumina platform and analyzed for a variety of variants and tumor mutational burden. DNA methylation was detailed using target capture probes targeting DNA methylation sites. Single cell approaches were applied to explore the transcriptome using 10X Genomics scRNAseq Flex kit. Digital spatial profiling was done using the NanoString for both transcriptomics and proteomics, with the GeoMx Whole Transcriptome Atlas and Human Core and Pan-Tumor protein panels. Aggregating these data illustrates the expanding landscape of information that can be extracted from FFPE derived tissue and the potential for novel discovery and diagnostic power integrating these complex data types holds. Citation Format: Andrea J. O'Hara, Yang Han, Ilaria DeVito, Laure Turner, Haythem Latif. Unraveling layers of proteogenomic complexity in cancer through multiomic exploration of archived FFPE tissue [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 338.

Abstract 5498: Multi-modal examination of spatial heterogeneity in the astrocytoma microenvironment

Abstract Astrocytoma is a diffuse tumor that infiltrates the brain. Despite post-operative therapy including radiation and chemotherapy, 52% of patients’ tumors recur within five-years of surgery. These tumors are characterized by mutations in the IDH1 or IDH2 gene that confer metabolic alterations, most notably the production of the onco-metabolite 2-hydroxyglutarate. To reveal the metabolic alterations and their upstream regulation across the tumor microenvironment, we applied spatial metabolomics and spatial transcriptomics to sequential sections from resected tumor samples of patients with low grade astrocytoma. This necessitated extensive optimization of atmospheric-pressure MALDI-TOF, including the establishment of a custom automated computational pipeline with abilities to identify metabolic niches. Additionally, using sequential sections of the same tumor, we also compared all commercially available imaging-based spatial transcriptomic technologies (10x Genomics Xenium, Nanostring CosMx, Vizgen MERSCOPE) to evaluate their suitability with regards to building a multi-modal model of metabolic activity and regulation across the tumor microenvironment. We found that 10x Genomics Xenium was most suited for this task as image registration across sequential sections was comparatively straight-forward. Combining spatial transcriptomics generated with 10x Genomics Xenium and spatial metabolomics across 8 matched sequential slices of the same tumor, we identified metabolic niches associated with distinct cell type composition and transcriptional regulation. Citation Format: Joel Moffet, Jurgen Kriel, Tianyao Lu, Lutz Freytag, James Whittle, Sarah Best, Saskia Freytag. Multi-modal examination of spatial heterogeneity in the astrocytoma microenvironment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5498.

Abstract 1139: Unravelling tumor heterogeneity in patients with HER2-low hormone receptor-positive breast cancer using spatial transcriptomics

Abstract Background: The current binary distinction between HER2-positive and HER2-negative breast cancer (BC) has recently been challenged by the emergence of the HER2-low entity. Antibody-drug conjugates targeting HER2 have shown efficacy in both HER2-positive and HER2-low BC, while some preliminary evidence showed efficacy also in patients with HER2-zero BC. Here, by combining spatial transcriptomics (ST) and morphological annotation, we aimed to investigate the inter- and intra-tumor heterogeneity of ERBB2 expression in relation to breast cancer IHC subtypes. Methods: We performed ST (Visium 10X Genomics) on frozen tumor samples from patients with hormone receptor positive (HR+) ductal BC with HER2 IHC staining and FISH. Hematoxylin/eosin slides were morphologically annotated integrating manual and machine learning-based approaches reaching single-cell resolution (QuPath software). The relative histomorphological categories composition of each spot across the ST slide was computed as percentage of pixels. Spots were defined as tumoral if the percentile of pixels in that spot surpassed the 80th percentile of all the tumor spots in the cohort. Results: Our cohort consisted of 56 HER2-zero (IHC 0), 30 HER2-low (IHC 1+) and 20 HER2-positive (IHC 2+/FISH positive or IHC 3+) samples. HER2-zero samples contained a significantly higher percentage of acellular stroma spots compared to HER2-low (pvalue=0.04), while no other morphological annotation showed a difference, including tumor spots. Only tumor spots were retained for downstream analysis. A difference in the ERBB2 gene expression was found at the spot level, with a higher expression in HER2-low samples compared to HER2-zero samples (median expression 0.4, 0.7 and 3.1 in HER2-zero, HER2-low and HER2-positive, respectively). Inter-sample heterogeneity was observed, with 27% of the HER2-zero samples having equal or higher mean ERBB2 expression that the mean ERBB2 expression of the HER2-low cohort. Variability within individual samples was observed, revealing two peaks in ERBB2 expression, one characterized by low levels and the other by medium-high levels. This pattern was frequently observed in samples classified as HER2-zero and HER2-low, indicating the existence of tumor spots with diverse ERBB2 expression. Conclusions: Our results revealed inter- and intra-tumor heterogeneity of ERBB2 expression in HER2-zero and HER2-low BC. The use of spatial transcriptomics can provide more details of the spatial distribution of ERBB2. Further validation is warranted. Citation Format: Danai Fimereli, Matteo Serra, Mattia Rediti, Frederic Lifrange, Nicola Occelli, Laetitia Collet, Delphine Vincent, Ghizlane Rouas, Ligia Craciun, Denis Larsimont, David Venet, Miikka Vikkula, Francois P. Duhoux, Françoise Rothé, Christos Sotiriou. Unravelling tumor heterogeneity in patients with HER2-low hormone receptor-positive breast cancer using spatial transcriptomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1139.