Even cells are subject to peer pressure.
For years, scientists have studied cancer cells in depth to gain a better understanding of the disease. However, they are now finding out that the non-cancerous cell population near cancerous cells has a strong influence on the tumor trajectory. Sylvia Plevritis PhD, Stanford Medicine’s chair of the department of biomedical information science, said
“Not all cells in a tumor are cancer cells — they’re not even always the most dominant cell type,” . Plevritis, along with a group of researchers from Stanford Medicine’s department of biomedical data science have created a system they refer to as the “colocatome,” or co-locateome (pronounced “co-locate”). The nomenclature is based on the terminology used to describe other classes of molecules or aspects of human biology. For example, information about the genes and proteins are called the genome, metabolites the metabolome. The colocatome records the number and type of cancerous cells in their immediate surroundings. Gina Bouchard PhD, an instructor in biomedical information science, said
“We’ve been studying cancer cells for so long, but the picture is still incomplete,” . “Understanding tumor biology is not only about cancer cells; there’s a whole ecosystem that needs to be studied. Cancer cells need help to survive, to resist, to thrive and even sometimes to die.”
The study that describes the findings has been published last month in Nature Communications . Bouchard, the principal author is Plevritis.
Mapping Influence
The environment of cancer cells is surprisingly important. The behavior of the cancer cells can be affected by the type, quantity and location of the non-cancerous surrounding cells. This could result in a faster cell growth rate, a decreased susceptibility for drugs, or an increased metabolism. Bouchard stated that
“The questions we’re asking are very simple. We want to know who the neighbors are for each cell. Who likes whom? Who doesn’t like whom? It’s all about which cells tend to be together, and which ones are rarely found together,” . The cells that are attracted to each other can be described as “colocalizing” while the ones that repel each other appear to form “anti-colocalizations.” . These colocalizations, which are linked to cancer state — aggressiveness, resistance to drugs, and susceptibility to drug treatment — are then logged into the colocatome.
In the laboratory, the team created experimental lung cancer models, and then used artificial intelligent to analyse them. They identified non-cancerous cell and their organization within and around tumor cells. The colocalizations were compared with biopsies from tumors taken from patients. They confirmed, after mapping hundreds of configurations of cells, that most of the colocalizations observed in primary tumors of patients are also seen in experimental models. Bouchard said that this overlap was crucial. The models represent lung cancer accurately and are valuable. Plevritis’s research and that of others has shown strong interactions between cancer cells and fibroblasts, although the exact way fibroblasts work with cancerous cells remains unclear. Plevritis demonstrated in an experiment that lung cancers die after being treated with anti-tumor drugs that inhibit cell growth. When fibroblasts are added to the equation, the landscape literally changes. Plevritis studied the tumor models that had been treated and found that the cancer cells were left in roughly the same proportion as the fibroblasts. They had rearrange themselves. Plevritis is the William M. Hume professor at the School of Medicine. “It was like changing the furniture in the room, then finding the exits are blocked.”
New leads
The team hopes to uncover more spatial configurations to help doctors understand why certain cancers persist even after being treated. The researchers hope that the colocatome will provide valuable information to guide treatment for cancer patients. If, for example, a colocalization is associated with resistance to one drug, doctors can look for another. The researchers also hope that the maps of colocalization will help generate hypotheses for aspects of cancer biology which are still unclear. The team hopes to use AI as they gather more data to create maps and catalogs that correlate to various cell states in cancers. “Then we can begin to see whether certain spatial motifs are shared between cancer types, regardless of where they originate in the body. That could reveal universal rules of tumor behavior and guide the design of more broadly effective treatments,” Plevritis said. This research was conducted by a researcher at the University of Oxford. The National Institute of Health funded this study (grants R25CA180993, K99CA255586), as did Les Fonds de Recherche du Québec. Stanford University’s Department of Biomedical Data Sciences supported this work.
