What if we could map not just who the cells are, but how they talk, where they gather, and what happens when their networks break down? Spatial biology is redefining our approach to tissue analysis by tracing cellular conversations and influences within their natural context. Instead of merely identifying cell types, spatial techniques let us follow molecular “breadcrumbs”—the secreted signals and patterns that shape how cellular communities form, coordinate, or unravel.

“The New Cartography: Spatial Biology’s Future Routes” explores these fundamentals, blending scientific depth with creative metaphor. I’ll guide you through cutting-edge tools in spatial biology, highlighting both the powerful insights and persistent challenges that come with measuring complex cellular interactions. Which conversations can our current technology capture, and which remain elusive? How do we turn dense spatial data into actionable understanding of harmony and discord within tissues?

A particular focus will be on the language of secreted molecules—cytokines, chemokines, and growth factors—that mediate cellular communication and transformation. Unraveling these messages is central to understanding disease, development, and regeneration but remains a tough technical challenge.

Spatial biology is not just about creating more detailed maps; it is about asking more precise questions and uncovering the hidden dialogue of life that guides us toward our true north star.

In the past, single-cell data visualization has primarily focused on dimensionality reduction techniques and scatter plots. However, with the advent of spatial single-cell biology, the demand for interactive tools supporting visual exploration of integrated, multi-modal spatial datasets has surged. These datasets often encompass measurements of multiple molecular entities and are generated using diverse assays. To address the rapidly evolving landscape of data sets and analytical tools, we have developed the Vitessce visualization framework (https://vitessce.io). This talk will introduce the Vitessce framework, highlighting its capabilities and applications across various research studies and software tools and other tools that the HIDIVE Lab has developed for spatial and single-cell biology data based on this framework. The talk will also discuss needs for standardization of data schemas and frameworks as well as strategies to effectively support 3D tissue maps data exploration using the Vitessce Link hybrid mixed reality approach developed by the HIDIVE Lab.

Understanding prostate cancer progression requires resolving molecular signals in their histological context. We applied an integrated spatial omics framework to clinical prostatectomy specimens spanning distinct histological grades, combining mass spectrometry imaging (MSI), spatial proteomics, and spatial transcriptomics (GeoMx Digital Spatial Profiling) with expert histopathological annotation. MSI delineated grade-associated metabolic landscapes and lipid architectures across tumor, stroma, and glandular compartments, revealing spatially confined metabolic reprogramming at invasive fronts. Spatial proteomics quantified compartment-specific protein networks, highlighting gradients of androgen receptor signaling, immune modulation, and extracellular matrix remodeling that tracked with Gleason pattern complexity. GeoMx profiling captured transcriptomic heterogeneity within annotated regions of interest, uncovering co-localized programs of epithelial plasticity, neuroendocrine features, and immune exclusion in higher-grade lesions.

This presentation demonstrates that coordinated spatial omics with histology provides a high-resolution atlas of prostate cancer’s molecular and microenvironmental complexity, clarifying how heterogeneity and niche interactions drive progression. Our approach establishes a framework for translational deployment of spatial biology in clinical specimens, enabling refined grading and biologically informed stratification in prostate cancer.

Spatial transcriptomics is a cutting-edge technology that allows the simultaneous visualization of the spatial distribution of gene expression and histological features in tissue samples. In the field of cancer biology, spatial transcriptomics has the potential to unravel complex biological processes that contribute to tumor development, progression, and response to treatment. Herein, I will discuss how

analyzing the gene expression profiles of cancer cells within their microenvironment and maintaining the tissue architecture, spatial transcriptomics can be used to identify and characterize key cancer hallmarks, such as angiogenesis, immune evasion, and tissue invasion. This information can help researchers better understand the mechanisms underlying tumor growth and develop novel targeted therapies. Spatial transcriptomics can also be a promising tool to improve cancer diagnosis and treatment, where I will explain the example of the biology and prediction-of-origin in cases of cancer of unknown primary (CUPs). By analyzing the spatial distribution of gene expression patterns in CUPs, we open the door to tailor treatment accordingly. Finally, I will introduce how we iused spatial transcriptomics to investigate the location of the cellular and molecular changes that occur in other diseases, such as in the lung tissue of COVID-19 patients. Initial data suggest that important insights into the immune response and tissue damage caused by the virus, including the identification of cellular and molecular scars left by lethal COVID-19 in the lung, can be obtained. Overall, spatial transcriptomics holds great potential for advancing our understanding of cancer biology, improving diagnosis and treatment, and shedding light on other diseases such as COVID-19.

A CTRB2 exon 6 deletion variant is among the top GWAS hits linked to an increased risk of pancreatic ductal adenocarcinoma (PDAC). This variant has been predicted to lead to a truncated CTRB2 protein that misfolds and accumulates in the ER. Using CRISPR/Cas9, we generated a mouse strain carrying the orthologous mutation in Ctrb1 (Ctrb1^Δexon6) and profiled pancreatic tissue over 1.5 years. Ctrb1^Δexon6 mice express a truncated CTRB1 protein that accumulates in the ER, resulting in ER dilation, suppression of the acinar program, and activation of ER stress and inflammatory pathways within three months. With aging, we observed mild histological alterations, reduction of the acinar program, and evidence of adaptive dampening of inflammatory responses, highlighting remarkable long-term organ adaptation to the mutant allele. A key discovery of our study is the mosaic acinar phenotype. Using spatial transcriptomic analysis, we have uncovered discrete acinar cell subpopulations distinguished by the expression of AGR2, one of the most highly upregulated genes in Ctrb1^Δexon6 pancreata. AGR2 is an ER-associated protein disulfide isomerase involved in protein folding. Cells with high amounts of AGR2 displayed higher ER stress signatures and an enrichment of reprogramming (OSKM-associated), underscoring spatially restricted acinar adaptation to chronic ER stress. Inflammatory pathways peaked in cells with high amounts of AGR2 at 3 months. Interestingly, mice harboring the Ctrb1^Δexon6 mutation in heterozygosity displayed an intermediate phenotype. Importantly, human pancreatic tissues from organ donors imputed or genotyped for the deletion variant recapitulated these findings, exhibiting downregulation of the acinar program, upregulation of ER stress pathways, enrichment of the Ctrb1^Δexon6-derived transcriptomic signature, and patchy AGR2 expression. Together, these results demonstrate that spatially heterogeneous acinar remodeling is a central feature of the Ctrb1^Δexon6 phenotype. By linking cell-intrinsic ER stress adaptation to tissue-level mosaicism, our work provides mechanistic insights into how the CTRB2 deletion variant may contribute to PDAC risk.

Pancreatic ductal adenocarcinoma (PDAC) is characterized by a highly complex and immunosuppressive tumor microenvironment (TME) that contributes to therapeutic resistance. A deeper understanding of the cellular and molecular mechanisms orchestrating this ecosystem is therefore crucial for developing novel intervention strategies. In this context, we are employing a multiparametric approach—integrating high-dimensional flow cytometry and mass cytometry (CyTOF)—to dissect the immune landscape of pancreatic cancer in preclinical mouse models. Particular attention is given to the cytokine IL-17A, which has emerged as a critical modulator of tumor-associated immune networks. Our studies aim to unravel how IL-17A shapes immune cell populations and stromal interactions within the TME, and to explore how its depletion may synergize with immunotherapeutic approaches, including next-generation DNA vaccines. By mapping the spatial and functional heterogeneity of the TME, these analyses provide new perspectives on targeting immunoregulatory pathways to overcome resistance and enhance the efficacy of cancer immunotherapy.

Recent advances in spatial transcriptomics, proteomics, metabolomics etc. assays have revolutionized our understanding of the tumor microenvironment (TME). As these methods mature, users are increasingly combining multiple readouts for a holistic view of TME heterogeneity and complexity. Spatial multi-omics presents specific bioinformatics issues, requiring specific workflows to make sense of the data.

We present software and a computational framework for integration and data analysis of spatial biology data. Weave software addresses several spatial biology bioinformatics challenges, such as joint visualization of different spatial omics assays, providing common downstream multimodal analysis pipelines, integration of derived results, and enabling communication of results between collaborators. Users can explore multicellular environments, and uncover novel discoveries within both new and existing datasets, in areas spanning research, drug discovery and biomarker discovery.

Our approach to spatial multi-omic data integration is technology agnostic, enabling combined analysis of data acquired from different spatial platforms, from the same or serial sections. This is demonstrated via two different use cases. In the first, multiplexed immunofluorescence-based spatial proteomics was combined with serial-section multimodal mass spectrometry imaging of peptides, glycans and endogenous metabolites to investigate cellular and molecular heterogeneity, and mechanisms of intrinsic chemoresistance of high-grade serous ovarian cancer. In the second, lung cancer biopsy sections were sequentially analyzed via spatial transcriptomics and spatial proteomics using different commercial platforms. The use of same-section acquisition enabled single-cell level comparisons of RNA and protein expression, revealing segmentation accuracy and transcript-protein correlation analyses within individual cells.

Breast and gastric cancers are among the most common cancers worldwide and are strongly influenced by the p53–MDM2 axis, yet clinical development of MDM2 inhibitors has been limited by drug resistance and poor efficacy. We investigated K1, a quinazoline derivative, which demonstrated potent anticancer activity with an IC50 value of 18.5µM in AGS and 19µM in MCF7 cell lines 72 h post treatment, and it also significantly reduced their clonogenic potential. Transcriptomic data show that K1 targets multiple stress pathways in distinct cellular compartments. In MCF7, ribosomal stress genes (RPL11, RPL23, RPS7, RPL26) and ER stress markers (DDIT3, ATF4, EIF2AK3) were upregulated. In addition, the expression of apoptotic protein mediators (PMAIP1, BBC3) and DNA damage regulators (CDKN1A, CDKN2D, FHIT) was also upregulated, whereas in AGS cells, mitochondrial dysfunction and mitophagy indicators (PINK1, PRDX3, PARK7), oxidative stress genes (CYP1A1, BACH1), and apoptotic regulators (FOXO3, RHOB, STAT1) showed increased expression. Elevated expression of mitotic stress and spindle assembly checkpoint genes (PLK1, CCNB1, CDC20, BUB1, MAD2L) leads to G2/M arrest and spindle disruption. This compartment-specific activity indicates a spatial biology perspective, linking gene expression changes to localized stress within the tumor environment. K1 causes nuclear fragmentation, cytoskeletal disruption, ROS generation, mitochondrial dysfunction, and DNA damage. In vitro studies indicated decreased MDM2 with increased p53 and p21 expression in MCF7, while AGS showed a decline in MDM2 and p53, with increased p21 expression. Despite the differences in response, both cell lines ultimately culminated in apoptosis. Molecular docking of K1 with MDM2 exhibited key molecular interactions, binding conformation, and stable dynamics. Our data identifies K1 as a promising anticancer candidate that activates compartment-specific, spatially distinct stress responses, leading to apoptosis in two distinct cancer types.

Navignostics is a precision oncology company pioneering the use of spatial single-cell proteomics to transform cancer diagnostics and treatment selection. Founded in 2022 as a spin-off from the University of Zurich’s Bodenmiller Lab, Navignostics combines highly multiplex tumor imaging with advanced data analytics to guide clinicians in identifying the most effective, personalized therapies for each cancer patient. Navignostics’ diagnostics test delivers detailed insights into tumor and immune cell composition from a single tissue section, providing a report that supports both individualized treatment decisions for patients and the development of novel targeted therapies. A recently in nature medicine published, Navignostics-backed study demonstrates that multiomics tumor profiling can drive dramatically improved outcomes for patients with advanced melanoma, particularly those with few remaining treatment options (Miglino et al., 2025). In a matched analysis of patients who had received at least three prior lines of therapy, multiomics-guided treatment, led to a median progression-free survival (PFS) of 8.34 months, compared to just 2.0 months in the matched cohort. The disease control rate (DCR) in this group was 64.7%, nearly three times higher than the 23.5% seen in non-TuPro patients. These findings suggest a strong clinical benefit of multiomics tumor profiling in late-line settings where standard diagnostic approaches often fail to guide effective treatment.

Solid tumors are composed of a diverse mixture of cancer, immune, and stromal cells. Understanding their spatial organization within the tumor microenvironment (TME) is critical for deciphering mechanisms of tumor progression and response to therapy. Targeted spatial proteomics for spatial organization studies of the TME is poised to revolutionize clinical practice as a natural evolution of immunohistochemistry (IHC), a technique routinely used for clinical diagnostics. However, direct comparisons that evaluate the relative capabilities and trade-offs of the major available targeted spatial proteomics platforms are currently scarce. In this study, we systematically evaluate two leading targeted spatial proteomics platforms: Co-Detection by Indexing (CODEX), a cyclic immunofluorescence technology, and Multiplexed Ion Beam Imaging (MIBI), an imaging mass spectrometry technique. Using a tissue microarray from a clinical cohort of 85 patient-derived head and neck squamous cell carcinoma samples, we processed a section from the same tissue blocks on each platform. To enable a side-by-side evaluation, we applied a comparable Python pipeline for image pre-processing, segmentation, and iterative clustering. Our results demonstrate that despite inherent differences in sample preparation and imaging approaches, both technologies can yield consistent and spatially relevant biological conclusions. For example, using antibody markers such as CD4, FOXP3, γH2AX, and cytokeratin, we successfully differentiated and localized CD4+ T cells, regulatory T cells, and cancer cells while simultaneously assessing cellular states like DNA damage. In conclusion, both technologies enable multiplexed, spatially resolved characterization of cellular architecture and heterogeneity in the TME. Future work will focus on a deeper spatial characterization of the dataset, alongside a direct comparison of the resolution and dynamic range of the two platforms. Once completed, this study will provide a robust framework for understanding the respective strengths and limitations of each technology, helping researchers select the optimal platform to accelerate the application of spatial biology in cancer research.