Spatial metabolomics and transcriptomics reveal cell type-specific dynamics in the placenta of patients with late-onset preeclampsia

Invasive ductal carcinoma (IDC) is the most common and aggressive form of breast cancer, characterized by tissue invasion and metastasis, contributing to significant morbidity and mortality. The tumor microenvironment (TME), composed of stromal and immune cells, plays a critical role in IDC progression and therapy resistance. However, the spatial organization and molecular heterogeneity within the IDC TME remain underexplored. Spatial transcriptomics enables the preservation of tissue architecture while revealing molecular signatures at single-cell resolution, offering new insights into cancer biology. We utilized the CosMx® Spatial Molecular Imager (SMI) platform with the CosMx Whole Transcriptome panel (approximately 19, 000 genes) to analyze a formalin-fixed, paraffin-embedded (FFPE) IDC tumor sample. The platform provides high-resolution, multiplexed RNA detection within its native tissue context. Hematoxylin and eosin (H&E) staining of the same slide provided a histological reference for spatial transcriptomics. Transcriptomic data were analyzed using Uniform Manifold Approximation and Projection (UMAP) for dimensionality reduction and Leiden clustering to identify cell clusters and spatial patterns. Marker genes for each cluster were used to annotate cell types and assess their roles within the TME. Analysis identified two distinct tumor regions, each subdivided into subregions with unique and shared gene expression profiles. These tumor regions aligned with histological features observed in H&E staining, validating the spatial transcriptomic findings. We observed region-specific stromal compositions, with fibroblast populations exhibiting distinct gene expression patterns. For example, one region was enriched for PIGR, CLU, and LTF, markers linked to immune response, tumor progression, and tumor-suppressive properties. These results highlight the molecular and spatial heterogeneity of IDC and the complex interactions within the TME. This study demonstrates the potential of spatial transcriptomics to uncover the molecular and spatial complexity of IDC. Integrating transcriptomic data with histological validation revealed novel insights into the TME, highlighting key gene signatures and stromal dynamics. The CosMx platform is a powerful tool for advancing precision oncology and guiding future therapeutic strategies in IDC. -Bruker Confidential- Citation Format: Liang Zhang, Shanshan He, Michael Patrick, Stefan Rogers, Haiyan Zhai, John Lyssand, Michael Rhodes, Joseph Beechem. Single-cell spatial whole transcriptome reveals tumor heterogeneity and stromal dynamics in invasive ductal carcinoma of the breast [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2067.

Nasopharyngeal carcinoma (NPC) is a heterogeneous cancer with variable therapeutic responses, highlighting the need to better understand the molecular factors influencing treatment outcomes. This study aims to explore spatially metabolic and gene expression alterations in NPC patients with different therapeutic responses and PD-1 expression levels. Methods: This study employs spatial metabolomics (SM) and spatial transcriptomics (ST) to investigate significant alterations in metabolic pathways and metabolites in NPC patients exhibiting therapeutic sensitivity or elevated programmed death 1 (PD-1) expression. The spatial distribution of various cell types within the TME and their complex interactions were also investigated. Identified prognostic targets were validated using public datasets from TCGA, and further substantiated by in vitro functional analyses. Results: SM analysis revealed substantial reprogramming in lipid metabolism, branched-chain amino acid (BCAA) metabolism, and glutamine metabolism, which were closely associated with therapeutic response and PD-1 expression. ST analysis highlighted the critical role of interactions between precursor T cells and malignant epithelial cells in modulating therapeutic response in NPC. Notably, six key genes involved in BCAA metabolism (IL4I1, OXCT1, BCAT2, DLD, ALDH1B1, HADH) were identified in distinguishing patients with therapeutic sensitivity from those with therapeutic resistance. Functional validation of DLD and IL4I1 revealed that gene silencing significantly inhibited NPC cell proliferation, colony formation, wound healing, and invasion. Silencing DLD or IL4I1 induced cell cycle arrest. Reduction in α-Ketomethylvaleric acid (KMV) levels was demonstrated upon IL4I1 silencing. Immunohistochemical analysis further confirmed that high expression of these six genes was significantly associated with poor prognosis in NPC patients, a trend corroborated by data from the TCGA head and neck cancer cohort. Conclusions: This study highlights the pivotal roles of key molecular players in therapeutic response in NPC, providing compelling evidence for their potential application as prognostic biomarkers and therapeutic targets, thereby contributing to precision oncology strategies aimed at improving patient outcomes.

In South and Southeast Asia, the habit of chewing betel nuts is prevalent, which leads to oral submucous fibrosis (OSF). OSF is a well-established precancerous lesion, and a portion of OSF cases eventually progress to oral squamous cell carcinoma (OSCC). However, the specific molecular mechanisms underlying the malignant transformation of OSCC from OSF are poorly understood. In this study, the leading-edge techniques of Spatial Transcriptomics (ST) and Spatial Metabolomics (SM) are integrated to obtain spatial location information of cancer cells, fibroblasts, and immune cells, as well as the transcriptomic and metabolomic landscapes in OSF-derived OSCC tissues. This work reveals for the first time that some OSF-derived OSCC cells undergo partial epithelial-mesenchymal transition (pEMT) within the in situ carcinoma (ISC) region, eventually acquiring fibroblast-like phenotypes and participating in collagen deposition. Complex interactions among epithelial cells, fibroblasts, and immune cells in the tumor microenvironment are demonstrated. Most importantly, significant metabolic reprogramming in OSF-derived OSCC, including abnormal polyamine metabolism, potentially playing a pivotal role in promoting tumorigenesis and immune evasion is discovered. The ST and SM data in this study shed new light on deciphering the mechanisms of OSF-derived OSCC. The work also offers invaluable clues for the prevention and treatment of OSCC.

Background: Invasive lobular carcinoma (ILC) is the second most common histological breast cancer subtype; however, little is known about its tumor microenvironment (TME). Here, we aimed to identify and to characterise ILC subgroups based on TME heterogeneity by combining spatial transcriptomics (ST) and single cell RNA-sequencing (scRNA-seq). Methods: We performed ST (Visium 10x Genomics) on frozen tumor samples from 41 primary hormone receptor positive (HR+), HER2-negative (HER2-) ILCs. Information coming from the morphological annotation of ST slides and the gene expression-based clustering of ST spots was integrated and used as input for a patient level classification (using intNMF algorithm). Subgroups of patients were annotated using morphology, gene set enrichment analysis and ST deconvolution based on scRNA-seq data (CARD software). SCAN-B bulk RNA-seq dataset was used as validation set (HR+, HER2- ILC samples, n = 853). Results: Four subgroups of patients were identified: proliferative (P, n = 12, enriched in tumor cells and proliferation-related pathways), normal-stroma enriched (NSE, n = 10, enriched in fibroblasts, stroma-related pathways and showing an higher level of colocalization between invasive and in situ carcinoma), metabolic (M, n = 9, enriched in metabolic related-pathways) and metabolic-immune enriched (MIE, n = 10, enriched in adipocytes, metabolic and immune-related pathways). Deconvolution of ST spots with single cell data revealed matching results with morphology (in terms of cell types). We observed an enrichment of myofibroblasts in NSE, an enrichment of cancer epithelial in P and, interestingly, an enrichment of endothelial cells in M subgroup (fdr < 0.25). MIE was characterized by perivascular like-endothelial cells and myeloid cells (monocytes and macrophages, fdr < 0.25). Of note, at the level of the ST slide, myeloid cells in MIE subgroup were more present in the adipose tissue compartment (fdr < 0.25) compared to the rest of the slide. NSE subgroup showed a trend for association with better prognosis. Gene signature-based assignment identified the same four subgroups in SCAN-B: GSEA analysis and comparisons among the four subgroups showed matching results between our dataset and SCAN-B. Of note, in SCAN-B, we observed differences in prognosis among the subgroups, with NSE showing better, M and MIE intermediate and P worse prognosis for overall survival and relapse free interval (p < 0.01). Conclusions: Four subgroups of ILC based on TME heterogeneity and showing differences in prognosis were identified and validated in external cohort. Of note, two of these groups were related to metabolism, highlighting the value of such process in ILC biology. Importantly, myeloid cells were the predominant immune cell types in ILC, and they showed a specific compartmentalization inside the TME. Further validation is needed. Citation Format: Matteo Serra, Andreas Papagiannis, Mattia Rediti, Frederic Lifrange, Nicola Occelli, Laetitia Collet, Delphine Vincent, Ghizlane Rouas, Danai Fimereli, Ligia Craciun, Denis Larsimont, David Venet, Miikka Vikkula, Francois P. Duhoux, Françoise Rothé, Christos Sotiriou. Integrating spatial and single cell transcriptomics to identify and characterise biologically driven subgroups in invasive lobular carcinoma [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 5066.

Dissecting the brain with spatially resolved multi-omics

Single-cell sequencing studies have characterized the transcriptomic signature of cell types within the kidney. However, the spatial distribution of acute kidney injury (AKI) is regional and affects cells heterogeneously. We first optimized coordination of spatial transcriptomics and single-nuclear sequencing data sets, mapping 30 dominant cell types to a human nephrectomy. The predicted cell-type spots corresponded with the underlying histopathology. To study the implications of AKI on transcript expression, we then characterized the spatial transcriptomic signature of 2 murine AKI models: ischemia/reperfusion injury (IRI) and cecal ligation puncture (CLP). Localized regions of reduced overall expression were associated with injury pathways. Using single-cell sequencing, we deconvoluted the signature of each spatial transcriptomic spot, identifying patterns of colocalization between immune and epithelial cells. Neutrophils infiltrated the renal medulla in the ischemia model. Atf3 was identified as a chemotactic factor in S3 proximal tubules. In the CLP model, infiltrating macrophages dominated the outer cortical signature, and Mdk was identified as a corresponding chemotactic factor. The regional distribution of these immune cells was validated with multiplexed CO-Detection by indEXing (CODEX) immunofluorescence. Spatial transcriptomic sequencing complemented single-cell sequencing by uncovering mechanisms driving immune cell infiltration and detection of relevant cell subpopulations.

Innovations in spatial omics technologies applied to human tissues have led to breakthrough discoveries in various diseases, including cancer. Two of these approaches—spatial transcriptomics and spatial metabolomics—have blossomed independently, fueled by technologies such as spatial transcriptomics (ST) and mass spectrometry imaging (MSI). Although powerful, these technologies only offer insights into the spatial distributions of restricted classes of molecules and have not yet been integrated to provide more holistic insights into biological questions. These techniques can be performed on adjacent serial sections from the same sample, but section‐to‐section variability can convolute data integration. We present a novel method combining desorption electrospray ionization mass spectrometry imaging (DESI‐MSI) spatial metabolomics and Visium spatial transcriptomics on the same tissue sections. We show that RNA quality is maintained after performing DESI‐MSI on a tissue under ambient conditions and that ST data is unperturbed following DESI‐MSI. We demonstrate this workflow on human breast and lung cancer tissues and identify novel correlations between metabolites and mRNA transcripts in cancer‐specific tissue regions.

Aim: In the present study, we aimed to evaluate serum IMA, IMA/albumin ratio, and DTDH levels in patients with early- and late-onset PE compared to healthy controls.Impaired homeostasis between oxidant and antioxidant mechanisms, inflammatory processes, and endothelial dysfunction play a key role in the pathogenesis of preeclampsia (PE). Serum ischemia modified albumin (IMA) and dynamic thiol/disulfide homeostasis (DTDH) levels are elevated in the presence of inflammation, oxidative stress, and endothelial dysfunction. Material and Methods: A total of 24 patients with early-onset PE and 62 patients with late-onset PE were included.The control group consisted of 46 healthy controls with similar gestational weeks. Serum samples were collected and IMA, albumin, and native, total, and disulfide thiol levels analyzed. Corrected IMA/albumin ratios were also calculated.Results: Disulfide levels, disulfide/native and disulfide/total thiol levels were higher in patients with early-onset PE compared to late-onset patients(p=0.008, p=0.022, and p=0.021, respectively). However, there was no significant difference between the late-onset PE patients and late-onset PE controls. Although there was no significant difference in the IMA levels between the patient and control groups, the IMA/albumin ratio was higher in the early-onset and late-onset PE patients, compared to the control group. However, there was no significant difference between the early-onset and late-onset PE patients.Conclusion: Our study results showed increased disulfide levels, disulfide/native thiol, disulfide/total thiol and IMA/albumin ratio in the early-onset PE patients, indicating increased oxidative stress in the pathogenesis of PE. In the late-onset PE patients, there was an increase only in the IMA/albumin ratio. However, further large-scale, prospective studies are needed to confirm the diagnostic value of these markers in the clinical practice.

Bladder cancer (BC) is a malignancy that originates from the cells lining the bladder and is one of the most common cancers of the urinary system, capable of occurring in any part of the bladder. However, the molecular mechanisms underlying the malignant transformation of BC have not been systematically studied. This study integrated cutting-edge techniques of spatial transcriptomics (ST) and spatial metabolomics (SM) to capture the transcriptomic and metabolomic landscapes of both BC and adjacent normal tissues. ST results revealed a significant upregulation of genes associated with choline metabolism and glucose metabolism, while genes related to sphingolipid metabolism and tryptophan metabolism were significantly downregulated. Additionally, significant metabolic reprogramming was observed in BC tissues, including the upregulation of choline metabolism and glucose metabolism, as well as the downregulation of sphingolipid metabolism and tryptophan metabolism. These alterations may play a crucial role in promoting tumorigenesis and immune evasion of BC. The interpretation of ST and SM data in this study offers new insights into the molecular mechanisms underlying BC progression and provides valuable clues for the prevention and treatment of BC. Schematic illustration of BC metabolic reprogramming.

BACKGROUND Astrocytoma is a diffuse and heterogeneous glioma that infiltrates the brain. Despite post-operative therapy including radiation and chemotherapy, 52% of patients’ tumours recur within five-years of surgery. These tumours are characterised 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 heterogeneous tumour microenvironment, we applied spatial metabolomics and spatial transcriptomics to sequential sections from resected tumour samples of patients with low grade astrocytoma. METHODS Spatial assessment of heterogeneity was performed on serial sections using spatial metabolomics (atmospheric pressure MALDI-TOF) and spatial transcriptomics (10X Genomics Xenium). This necessitated extensive optimisation of image registration across modalities and sequential sections to create three dimensional models to deeply interrogate heterogeneity within tumours. RESULTS Combining spatial transcriptomics and spatial metabolomics across 8 matched sequential slices of the same patient tumour, enabled the identification of metabolic niches associated with distinct cell type composition and transcriptional regulation. Strikingly, normal domains that were present at the leading edge of tumours could be identified both metabolically and transcriptomically. CONCLUSIONS Understanding the heterogeneity of astrocytomas on a spatial level reveals unique domains and organisation of tumour and immune cell types. Matching these transcriptomic and cell state information with metabolic insights leads to pathway insights that will be essential to the identification of druggable targets. Our pipeline combining spatial transcriptomics and metabolomics with tissue histology reveals these biologically important domains.

Spatial omics technology integrates the concept of space into omics research and retains the spatial information of tissues or organs while obtaining molecular information. It is characterized by the ability to visualize changes in molecular information and yields intuitive and vivid visual results. Spatial omics technologies include spatial transcriptomics, spatial proteomics, spatial metabolomics, and other technologies, the most widely used of which are spatial transcriptomics and spatial proteomics. The tumor microenvironment refers to the surrounding microenvironment in which tumor cells exist, including the surrounding blood vessels, immune cells, fibroblasts, bone marrow-derived inflammatory cells, various signaling molecules, and extracellular matrix. A key issue in modern tumor biology is the application of spatial omics to the study of the tumor microenvironment, which can reveal problems that conventional research techniques cannot, potentially leading to the development of novel therapeutic agents for cancer. This paper summarizes the progress of research on spatial transcriptomics and spatial proteomics technologies for characterizing the tumor immune microenvironment.

Introduction: In digital pathology, extracting comprehensive data from a single tissue sample is critical for enhancing diagnostic precision and therapeutic strategies. Techniques like multiplex imaging and spatial transcriptomics offer unprecedented depth of information. This study introduces a novel and innovative approach combining for the first time Phenocycler® (with a 34-marker panel) and Visium® spatial transcriptomics on the same tissue section of squamous cell carcinoma of the upper aerodigestive tract. A single tissue section was analyzed using Phenocycler® (PP) for proteomics with a 34-marker panel targeting microenvironment and structural proteins, followed by H&E staining and Visium® spatial transcriptomics (ST) to map RNA expression. This approach allowed for the thorough assessment of immune cell spatial distribution and transcriptomic profiles, while evaluating the impact of proteomic treatment on RNA integrity and gene expression. Results: Despite initial concerns about RNA degradation due to multiplex staining treatments, RNA integrity in the PP+ST sample was comparable to ST-only samples, as indicated by similar total gene expression distributions. The PP+ST sample exhibited high-density feature points across the tissue, suggesting that proteomic pre-treatment did not impair subsequent transcriptomic analysis. Unsupervised UMAP visualization revealed similar clustering patterns between PP+ST and ST-only samples, indicating preserved cellular heterogeneity and molecular signatures. Consistent spatial distributions and clustering were observed between Phenocycler® pseudospots and Visium® data, with significant correlations in the spatial distribution of key cell types, including malignant clusters and immune cells. This study is a pioneering demonstration of the feasibility and efficiency of integrating spatial proteomics and transcriptomics on the same tissue section, guaranteeing high fidelity in molecular profiling. This combined method is crucial for deepening the understanding of the tumor microenvironment, with potential implications for improving diagnostics and personalized cancer therapies. The ability to analyze multiple data layers on the same cells enhances the development of precise AI-based diagnostic tools and personalized treatment strategies. Citation Format: yasmine adimi, clement Levin, Emilia Puig Lombardi, Marion Classe, Cecile Badoual. A new spatial multi-omics approach to deeply characterize human cancer tissue using a single slide: a new spatial muti-omics approach to deeply characterize human cancer tissue using a single slide [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7483.

Simultaneous profiling of spatial transcriptomics (ST) and spatial metabolomics (SM) on the same or adjacent tissue sections offers a revolutionary approach to decode tissue microenvironment and identify potential therapeutic targets for cancer immunotherapy. Unlike other spatial omics, cross-modal integration of ST and SM data is challenging due to differences in feature distributions of transcript counts and metabolite intensities, and inherent disparities in spatial morphology and resolution. Furthermore, cross-sample integration is essential for capturing spatial consensus and heterogeneous patterns but is often complicated by batch effects. Here, we introduce SpatialMETA, a conditional variational autoencoder (CVAE)-based framework for cross-modal and cross-sample integration of ST and SM data. SpatialMETA employs tailored decoders and loss functions to enhance modality fusion, batch effect correction and biological conservation, enabling interpretable integration of spatially correlated ST-SM patterns and downstream analysis. SpatialMETA identifies immune spatial clusters with distinct metabolic features in cancer, revealing insights that extend beyond the original study. Compared to existing tools, SpatialMETA demonstrates superior reconstruction capability and fused modality representation, accurately capturing ST and SM feature distributions. In summary, SpatialMETA offers a powerful platform for advancing spatial multi-omics research and refining the understanding of metabolic heterogeneity within the tissue microenvironment.

The combination of spatial metabolomics and spatial transcriptomics can reveal powerful connections between gene expression and metabolism in heterogeneous tissues. However, section-to-section variability can convolute data integration. We present a novel approach, termed Desium, which integrates spatial metabolomics by Desorption Electrospray Ionization Mass Spectrometry Imaging (DESI-MSI) and Visium Spatial Transcriptomics (VST) on the same tissue section. This work demonstrates that DESI-MSI does not significantly impact RNA quality, enabling robust analysis of gene-metabolite relationships in human breast and lung cancer tissues. The Desium workflow involves sequentially performing DESI-MSI at 100 μm spatial resolution, hematoxylin and eosin (H&E) staining, and VST CytAssist analyses on the same tissue section. We applied this workflow to six human cancer tissues (3 breast, 3 lung) and analyzed a serial section of two samples (1 breast, 1 lung) with standard VST CytAssist as a control. H&E and VST data were automatically aligned, while DESI-MSI and H&E data were coregistered manually using nonrigid registration. The MSI data was then converted into VST spot coordinates using a Gaussian granularity matching algorithm. Pearson correlation scores were calculated for each tissue to link gene expression and metabolite abundances. Comparing sequencing results between Desium and the standard VST protocol using UMI counts/gene, we found a correlation coefficient of 1 for both lung and breast samples, indicating minimal impact of DESI-MSI on gene expression measurements. Across all samples, we identified 15, 111 transcript-metabolite correlations from 1, 136 transcripts and 439 metabolites, with correlation scores above 0.5 or below -0.5. Notably, we observed a 0.82 correlation between glycerophosphorylethanolamine and XBP1 in breast cancer and a 0.77 correlation between phosphatidylglycerol PG 22:6_22:5 and MSLN in lung cancer. Our study demonstrates that Desium enables unambiguous spatial correlations between metabolites and RNA transcripts, uncovering novel relationships and unique patterns of intratumor heterogeneity not evident through histologic analysis. These findings suggest that Desium could serve as a powerful tool to explore the gene expression-metabolic phenotype relationships within the complex tumor microenvironment. Citation Format: Trevor M. Godfrey, Yasmin Shanneik, Wanqiu Zhang, Thao Tran, Nico Verbeeck, Nathan H. Patterson, Faith E. Jackobs, Maheswhari Ramineni, Chandandeep Nagi, Livia S. Eberlin. Integrating spatial metabolomics and spatial transcriptomics on the same cancer tissue sections to detect gene-metabolite correlations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 150.

Recent advances in spatial transcriptomics (ST) enable us to measure gene expression from cancer tissues while retaining their spatial context. We present a novel bioinformatics pipeline to infer molecular changes from tumor and immune cell interactions in the tumor microenvironment (TME) from ST data. Latent space methods enable inference of biological patterns from ST without the need for spot deconvolution into cell-based spatial features. While linear latent space methods yield interpretable biological patterns, interactions between tumor and immune cells can be nonlinear. To enable comprehensive inference of the pathways in the TME, we developed novel algorithms to characterize biological patterns from ST data using linear latent space methods and further nonlinear effects from their interactions. For any given set of genes, the patternSpotter tool visualizes the spatial variation in the relative contribution of individual patterns to the aggregate expression at each location in the tumor sample. Application of this tool to latent features identified using CoGAPS non-negative matrix factorization on a Visium ST (10x Genomics) data from a lymph node with pancreatic cancer metastasis confirms its known immune cell architecture. Furthermore, we develop a patternMarker algorithm to identify sets of coexpressed genes associated with the patterns, which help us to pinpoint the underlying biological processes and cell types. Further analyzing a breast cancer sample with invasive carcinoma and multiple precursor lesions demonstrates that this approach can uncover tumor and immune regions without prior reliance on pathology annotations from H&E imaging. In this case, an ensemble-based factorization of multiple dimensions enhances our resolution of intra-tumor heterogeneity and identifies distinct hormone receptor pathways in different precursor lesions with the patternMarker algorithm. Additional latent features are associated with immune cells, revealing further heterogeneity in immune infiltration between the invasive carcinoma and distinct precursor lesions. Still, the molecular interactions resulting from this infiltration induce a further non-linear alteration to transcription not captured through the inferred latent spaces. To resolve this, we develop a further interactionMarker statistic to identify regions of inter-pattern interaction and the associated genes. We apply this approach to detect additional intra-tumor heterogeneity in immune signaling from infiltration suggestive of differences in immune attack of invasive lesions. Altogether, our pipeline for latent space analysis of ST can identify the location and context-specific molecular interactions within the TME, broadly applicable to a better understanding of the key drivers of tumorigenesis and resistance to immune attack in cancer. Citation Format: Atul Deshpande, Melanie Loth, Dimitrios Sidiropoulos, QingFeng Zhu, Genevieve Stein-O'Brien, NIkhil Rao, Cedric Uytingco, Stephen Williams, Cesar Santa-Maria, Daniele M. Gilkes, Lei Zhang, Elizabeth Jaffee, Robert Anders, Ludmila Danilova, Luciane T. Kagohara, Elana J. Fertig. Uncovering the spatial landscape of tumor-immune interactions using latent spaces from spatial transcriptomics [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 2130.