Seeing Cancer in Space and over Time

 

Feb 2025 Cancer Commentary

Cancer Ecology Commentary begins a new year of Blog posts by reviewing a comparatively new development in laboratory cancer evaluation, ‘Spatial Omics’, the mapping within a cancer of the spatial distribution of tumor cells, their particular molecular features and the accompanying tumor microenvironment in which the neoplasm has arisen. The use of the term ‘Omics’ stems from the Greek word meaning group or whole and is used by biologists to indicate the study of any group of biological molecules of one type or another.   

Methods are now available to define the presence of DNA mutations, RNA expression levels and accompanying proteins derived from that RNA expression which can now be visualized in tissue sections, cell by cell, while preserving information to define each tumor cell’s spatial location. Molecular clocks are now being developed using mutation rates or methylation rates of tumor DNA to retrospectively define the clonal origin and phylogenetic relationship occurring in a cancer. A helpful overview of these advances comes from Eric Topol, published in his newsletter, Ground Truths, found online at Substack, November 10, 2024. A graphic illustrating these strata, taken from Topol’s publication is seen in figure 1 inserted below. Incidentally this publication was shared with me by Dr. Ed Weber, who has also kindly been an adviser for this Blog over the past year which this author is grateful for. 

Figure 1.

 

Of much interest is the promise that Spatial Multi-omics, will allow a better, more ‘resolved’ description of what cellular interactions are occurring within a tumor between tumor cells and adjoining stromal and immune cells of the tumor microenvironment.  An analogy would be to think of a mapping application from your computer which allows the user to visualize the spatial information of a map in any of several modes such as street grid, satellite view, terrain, traffic and transit. Now imagine a means of superimposing these separate fields of spatial information aligned in space to reveal, at the molecular level, the full complement of phenotypic characteristics of individual cells comprising the tumor microenvironment. The figure below, again taken from Topol, is a diagram illustrating the superposition of histologic, genomic, epigenomic and proteomic features of the tumor.

 

Figure 2

 These techniques for spatial imaging of omic parameters such as gene and protein expression or genetic mutations arose from earlier success in immunohistochemistry (IHC) and in situ hybridization. At their core these new spatial imaging techniques take advantage of the ability to create thousands of probes, each individually labeled with unique oligonucleotides, termed ‘barcodes’, whose presence can be tracked in tissue sections using reporter oligonucleotides complimentary to the probe nucleotides. Those reporter oligonucleotides are labeled with a set of fluorophores which then can be identified through fluorescence microscopy and assigned a physical location within that tissue section. So, for example, in co-detection by indexing (CODEX) a set of antibodies to cell surface markers, labeled with barcodes are allowed to bind in tissue section and then reported using the immunofluorescence of the hybridized oligomers to locate one or another group of cells displaying targeted surface markers and and then record the corresponding x, y coordinates of that cell's location. The labelling and reporting process is iterative, proceeding in multiple cycles producing an index of information which can then be registered with other parameters, particularly histopathology. The advantage of this system over conventional IHC stems from the ability to multiplex targets, using scores of antibodies at one time compared to the otherwise low dimensionality of single probes applied one at a time in studies with traditional IHC. 

In a similar fashion, RNA expression can be spatially localized in tissue sections using a system in which tissue sections are overlayed by a grid of ‘spots’ , each spot containing as many as 5000 different oligonucleotides, measuring 55 micrometers to cover the area of interest. RNA from individual cells is then released from those cells using a technique of permeabilization allowing the RNA molecules to escape from cells to be hybridized to the bar codes contained within the overlying spots. Those barcodes can then be reported and imaged allowing a cDNA library to be created to permit sequencing. Once again, the information from RNA expression can be tracked to specific locations permitting cells with similar RNA profiles to be clustered together for principal component analysis. This preservation of spatial information aligning with histopathology can then be used to study the neighborhoods in which cancers form and the character of the interface between the tumor and adjoining tumor microenvironment. Other techniques using whole exome sequencing and copy number variation can also now be indexed in the laboratory to decipher the clonal structure of a neoplasm and the identification of emerging subclones to deduce the phylogenetic relationship occurring across the tumor. That information in turn can be used as a measure of clonal evolution. Diagrammed below in figure 3 is an example of this type of study looking at the evolution of a neoplastic colon polyp undergoing malignant transformation.

 Figure 3

The tools then provided by these new techniques of spatial biology such as determination of spatial gene expression or the localization of cell subtypes within a tumor, including the findings of rare cell subtypes will provide investigators an increasing dynamic appreciation of the clonal and molecular events characteristic of a particular cancer. In the future we can expect increasingly detailed atlases to provide high resolution views of the molecular components of a cell in its native state and native tissue architecture. Importantly, it also illuminates the character of the cellular community in which a cancer cell resides, side stepping the problem of tissue disruption and loss of potentially valuable geographic information. Given the complexity of that neighborhood, we are now able to visualize at high resolution the cellular interactions occurring at the vascular, stomal and immune system levels providing a composite picture of the cancer. This in turn promises to help oncologists judge the evolving heterogeneity developing within a cancer. By assessing the pace of clonal evolution we are better able to inform a deterministic relationship between cancer progression and potentially relevant underlying selection forces. Finding the influences and circumstances which attract cells together may prove to be an important pathway into ecological understanding of cancer.

 Jim Cunningham