Even cells experience group pressure.
Scientists have long studied the ins and outs of cancer cells to learn more about the disease, but they are increasingly discovering that non -cancer cells near cancer cells exert a powerful influence on the trajectory of a tumor.
“Not all cells in a tumor are cancer cells, they are not even the most dominant type of cells,” said the Plevritis of Sylvia, PHD, president of the Stanford Medicine Biomedical Data Department of Biomedical Data. “There are many other types of cells that support tumors.”
To better capture the complete image of the locations and interactions of the cells, the Plevritis and a team of researchers have developed something they call the “colocatomo” (pronounced co-lubocate-man). Modeling after the nomenclature that describes other classes of molecules and facets of human biology (collective information about genes is called genome; protein, proteoming; metabolites, metaboloma, etc.) The colocatoma documents the details of malignant cells in their neighbors, what are those cells and how many of them are present.
“We have been studying cancer cells for so long, but the image is still incomplete,” said Gina Bouchard, PhD, Biomedical Data Science instructor. “Understanding tumor biology is not just about cancer cells; there is a complete ecosystem that must be studied. Cancer cells need help to survive, resist, prosper and even sometimes die.”
A study that describes the findings was published in Nature communications last month. Bouchard is the main author, and Plevritis is the main author.
Mapeo influence
Cancer cells depend surprisingly on their surroundings. Depending on the location, type and amount of non -cancer cells that surround the tumor, cell behavior can change, either through faster growth, a decrease in susceptibility to medications or high cell metabolism.
“The questions we are asking are very simple. We want to know who the neighbors are for each cell. Who likes who? Who does not like who? Cells that attract with each other described as “colochantors”, while those that seem to repel with each other “anti -collocation.” These placements are related to the state of cancer, aggressive, resistant, susceptible to medications, and record the colocatomo.
The team developed experimental models of lung cancer in the laboratory, then used artificial intelligence to analyze them, identifying non -cancer cells and how they organized inside and around tumor cells. Then they compared the placement with those of the patient’s tumor biopsies. After mapping hundreds of cell configurations, they confirmed that most of the placuation in primary patient tumors are observed in experimental models. (That overlap is key, Bouchard said. It means that the models are a valuable and precise representation of what is happening in someone who has lung cancer).
Previous investigations of Plevritis and others showed strong interactions between fibroblasts and cancer cells, but it is not clear exactly how fibroblasts interact with cancer cells. In an experiment, the plevritis showed that lung cancer cells die when they sprayed with a type of antitumoral drug that agrees cell growth. But throw fibroblasts into the mixture, and the whole landscape changes, literally. The Plevritis mapped the tumor models treated and saw that the subsequent treatment, cancer cells and fibroblasts were generally left intact in the same amount. But they had reorganized.
“That spatial reorganization seems to have given rise to drug resistance,” said Plevritis, a professor at William M. Hume at the Faculty of Medicine. “It was like changing the room furniture, then finding the outputs are blocked.”
Chasing new potential customers
As the team continues to register space maps of treated and not treated tumors, they hope to unlock more configurations that help doctors on the track about why some types of cancer persist after treatment. Ideally, the researchers said, the colocatomo could provide information that guides patient cancer treatment: if a specific placement confers resistance to a common medication, for example, doctors can look for another that may have a better opportunity to work. They also expect colocalization maps to generate verifiable hypotheses to describe aspects of cancer biology that are still clear.
As they collect more data, the equipment plans to use AI to identify specific space motifs and create maps catalogs corresponding to different cell states for a variety of cancers. “Then we can begin to see if certain space motifs are shared between the types of cancer, regardless of where they originate in the body. That could reveal universal rules of tumor behavior and guide the design of more effective treatments,” said Plevritis. “That is something that excites me a lot.”
An Oxford University researcher contributed to this research.
This study was funded by the National Institute of Health (grants R25CA180993, U54CA274511 and K99CA255586) and Les Fonds de Recherche du Québec.
The Stanford Biomedical Data Department also supported the work.