In Simple Terms
Cancer cells do not all act alike. Some can change how they behave when their surroundings change, which may help a tumor spread or resist treatment. Scientists are studying where these adaptable cells are and what signals affect them, hoping to find ways to limit their ability to change and spread.
A Tumor Is Not One Uniform Mass
A tumor can look like one solid mass, but the cells inside it may differ in what they do and how they respond to their surroundings. Some grow or react to nearby signals differently from the cells next to them. Describing a tumor with a single average can therefore hide small groups of cells whose behavior matters, including cells linked to the start of spread or to treatment resistance.
Phenotypic plasticity is the ability of a cancer cell to move between different states in response to what is happening around it. It is not the same thing as a change in DNA. The central idea is that a cell may alter its activity and behavior even when a new genetic mutation is not the explanation. This makes tumors harder to understand: researchers must ask not only what kind of cell they are looking at, but also what state it is in, what surrounds it, and which signals may be steering it.
This flexibility matters in colorectal cancer because it is linked to disease progression, the formation of metastases, and the possibility that treatment will not work. That does not mean every flexible cell will spread or resist treatment. Studying these cell states may, however, reveal some of the ways a tumor keeps growing and adapting. Researchers are working to understand what controls the shifts between states and where in the tissue those shifts take place.
Three Cell States in the Spotlight
The research focuses on regenerative and inflammatory cancer-cell states, as well as a hybrid state that combines features of regeneration and stem cells. The term regenerative describes a cell program associated with the potential for growth and tissue rebuilding, while inflammatory refers to a pattern connected with inflammation-related signals. The hybrid state brings together traits of the regenerative pattern and stem-cell-like features; it should not be reduced to a single, simple category.
These categories are especially interesting because the regenerative and inflammatory states observed in patients’ tumors resemble, according to the research, metastasis-initiating cells previously identified in mouse models. This resemblance provides a basis for comparison; it does not mean that human and mouse cells are identical in every respect. It also gives researchers a way to test whether traits seen in models correspond to states that are actually present in patient samples.
One notable finding is that liver metastases may recreate some of the diversity found in the primary tumor. In other words, when cancer cells reach the liver, the tumor there is not necessarily a simple, uniform copy of the cells that left the colon. A range of cell states may emerge again at the new site, making it important to study the metastasis itself rather than relying only on analysis of the original tumor.
This observation raises important questions about how that diversity arises. Do different cell states travel from the primary tumor to the liver? Do the cells change after they arrive? What role does the new environment play in reshaping them? The available description does not give numerical details or specify how many samples were studied, but it makes clear that comparing primary tumors with liver metastases is central to understanding the phenomenon.
Regulatory Clues: AP-1 and NF-κB
The cell states in this research are associated with the transcription factors AP-1 and NF-κB. Transcription factors are proteins that help switch groups of genes on or reduce their activity, thereby helping direct a cell toward a particular program. Their influence may help explain why regenerative or inflammatory traits appear, but that does not mean that any one factor can explain the behavior of an entire tumor.
Identifying these regulators matters because they may connect the signals cells receive with changes in gene activity. If the surroundings send signals that push a cell toward a particular state, understanding the molecular intermediaries that translate those signals could help researchers identify points to investigate as potential targets. This remains a line of research to be tested, however. Identifying a regulator does not establish that blocking it would be safe or effective as a treatment for patients.
Signals from fibroblasts, macrophages, and T cells add another layer to the picture. A tumor is not made up of cancer cells alone: other cells around them can help shape the local environment. The analysis explores how these cell types interact with different cancer-cell states, and whether the signals they exchange help maintain a particular state or encourage cells to shift into another one.
Where Cells Are Is Part of the Story
Knowing that a particular cell state is present in a sample is not enough; its location can change what that presence means. The research describes the cell states of interest as being at the invasive edge of the tumor, the region where cancerous tissue meets the surroundings it is advancing into. These states were also found in an immunosuppressive environment. The research question therefore brings together the cell’s identity, its location, and the conditions around it.
The invasive edge may be a useful place to study interactions between cancer cells and nearby tissue because it is where the tumor is advancing. But finding cells there does not, by itself, prove that they created the immunosuppressive environment or that they alone are responsible for invading tissue. Researchers need to interpret location alongside cell type and the signals being exchanged, rather than drawing a cause-and-effect conclusion from a spatial image alone.
An immunosuppressive environment adds another dimension to the problem. The question is not only how cancer cells move or multiply, but also how the tumor relates to the immune cells around it. Understanding the links among cancer-cell states, macrophages, T cells, and fibroblasts may clarify some features of the environment associated with spread. But identifying these links does not mean that changing them will automatically stop metastases. Moving from description to treatment requires further evidence and testing.
How New Techniques Reveal What a Sample Can Hide
The approach combines single-cell multi-omics analysis with spatial gene-expression imaging. The first method makes it possible to examine cells one at a time instead of treating the sample as a single average. The spatial method helps connect patterns of gene activity with the cells’ locations in the tissue. Used together, the techniques can provide answers to two related questions: what state is a cell in, and where is it located?
This combination is valuable because cell signals do not act in isolation. Learning that a group of cells has a particular pattern does not, by itself, show whether the cells are close to fibroblasts, macrophages, or T cells, or whether they are at the invasive edge or somewhere else. Adding location helps place a cell’s molecular state in the context of the tissue and may reveal relationships that are missed when cells are studied separately from their surroundings.
The study includes samples from primary tumors and liver metastases in patients with colorectal cancer. Comparing the two sites makes it possible to examine whether cell states change as the cancer spreads and how diversity is reshaped in the liver. The research announcement does not specify the number of patients or samples, so it is not possible to infer the size of the study group or how representative it is of everyone with the disease. Advanced analytical techniques also do not remove the need to verify findings in other studies.
From Mapping Cells to Looking for Treatment Targets
The long-term goal of this research is to find ways to limit cancer cells’ flexibility or prevent them from shifting into states associated with spread. If researchers can identify the signals that support regenerative or inflammatory states, and determine which surrounding cells contribute to them, they may find candidates for further investigation. This could point toward approaches that do more than kill cancer cells: they might also disrupt the programs that give some cells a greater ability to adapt.
The word candidate matters here. Identifying a cell state or regulatory factor does not yet prove that targeting it will prevent metastases, nor does it reveal the best way to treat it. Cancer cells may also occupy more than one state, and their responses may differ between a primary tumor and a metastasis. Molecular and spatial maps are therefore a step toward understanding the system, not an announcement of a ready-to-use treatment.
Mirjana Efremova leads this line of research at the Bart’s Cancer Institute at Queen Mary University of London, where she started her laboratory in May 2020 as part of the Centre for Cancer Evolution. Her lab studies non-genetic mechanisms that contribute to cancer initiation, spread, and treatment resistance. Efremova earned a PhD in computational biology from the Medical University of Innsbruck in Austria and later worked as a postdoctoral researcher in the Teichmann Lab at the Wellcome Sanger Institute. Her background provides a fitting foundation for research that combines analysis of cell data with the study of tumor development.
Efremova is scheduled to present this work in a live scientific seminar on Thursday, November 19, 2026, at 08:00 Pacific Time, 11:00 Eastern Time in the United States, and 16:00 GMT, followed by a question-and-answer session. The talk will focus on what samples of primary tumors and liver metastases reveal about cancer-cell states, the factors that regulate them, and their locations within tissue.
The Bottom Line
Research into phenotypic plasticity highlights one part of colorectal cancer’s complexity: cancer cells can adopt different states, and some of those states may be linked to regeneration, inflammation, and spread. The cells of interest have been observed at the invasive edge in an immunosuppressive environment. By combining single-cell analysis with spatial imaging, researchers can study cells, their signals, and their locations in primary tumors and liver metastases, with particular attention to AP-1, NF-κB, and interactions with fibroblasts, macrophages, and T cells. This work does not offer a ready-made treatment, but it points to questions that can be tested in the search for ways to limit a tumor’s ability to adapt and spread.