Multiplexed 3D atlas of state transitions and immune interactions in colorectal cancer

Abstract

Advanced solid cancers are complex assemblies of tumor, immune, and stromal cells that invade adjacent tissue and spread to distant sites. Here we use highly multiplexed tissue imaging, spatial statistics, and machine learning to identify cell types and states underlying morphological features of known diagnostic and prognostic significance in colorectal cancer. We find that a thorough spatial analysis requires imaging the entire tumor region, not small fields of view (e.g. those found in tissue microarrays). When this condition is met, the data reveal frequent transitions between histological archetypes (tumor grades and morphologies) correlated with molecular gradients. At the tumor invasive margin, where tumor, normal, and immune cells compete, localized features in 2D such as tumor buds and mucin pools are seen in 3D to be large connected structures having continuously varying molecular properties. Immunosuppressive cell-cell interactions also exhibit graded variation in type and frequency. Thus, whereas scRNA-Seq emphasizes discrete changes in tumor state, whole-specimen imaging reveals the presence of large- and small-scale spatial gradients analogous to those in developing tissues.

The spatial landscape of progression and immunoediting in primary melanoma at single cell resolution

“…The resulting molecular and morphological data were then analyzed with respect to histopathology annotations. CyCIF and H&E imaging were performed on whole slides, not tissue microarrays (TMAs) or small fields of view (FOVs) (Baharlou et al, 2019), to preserve the spatial relationships of different histologies and to provide sufficient statistical power (Lin et al, 2021).…”

Single Cell Spatial Analysis Reveals the Topology of Immunomodulatory Purinergic Signaling in Glioblastoma

“…Molecular analyses including single-cell genomic and proteomic approaches can yield important insights into the heterogeneous cell states within tumors. These methods typically employ tissue-dissociative methods which eliminate the spatial associations between tumor cells and other components of the tumor microenvironment thereby preventing the evaluation of tissuelevel phenomena which can be critical for promoting tumorigenesis [15][16][17] . An important recent study using dissociative methodologies identified a distinct population of tumor-infiltrating macrophages that express CD73 in glioblastoma 33 ; these cells may persist through recurrence and promote resistance to immunotherapeutic treatment.…”

“…Given cell-state specific expression of critical purinergic regulators such as CD73 and CD39, we next sought to explore the expression, spatial organization, and cell-cell interactions of these proteins in native tissue at sub-cellular resolution. Unlike dissociated scRNA-seq, multiplexed tissue imaging preserves spatial relationships between cells and proteins, enabling correlation of molecular states and tissue architecture 13,15 .…”

“…CyCIF can be used to identify multiple cell lineages and states and map their spatial organization and interactions 13 . Using such methods to generate 2D and 3D tumor atlases, such as those developed by the Human Tumor Atlas Network (HTAN) 14 , has revealed substantial spatial and molecular variation within human tumors, and population-level regulation of tumor-immune interactions [15][16][17] .…”

Temporal and spatial topography of cell proliferation in cancer.

“…Thus, in tumors, cancer cells exist in proliferative, non-proliferative, and arrested states 11 . In addition, many cells in tumor masses are not neoplastic, but rather are immune and stromal cells 12 that are non-uniformly distributed across the tumor 13 . The crosstalk between distinct cellular components of the tumor ecosystem creates local conditions that may either promote or inhibit cell proliferation 14 .…”