Researchers find link between nearest neighbors of immune cells and survival time of pancreatic cancer patients

Researchers at Johns Hopkins Medicine have found that the organization of different types of immune cells in pancreatic tumors is linked to how well pancreatic cancer patients respond to treatment and how long they survive. New results published on September 16 in Cancer researchcould eventually lead to new ways to treat pancreatic cancer, which has the highest death rate of all major cancers.

“Mapping the location of specific tumor-associated immune cells may be a new predictive biomarker patient survival“, says Alexander Popel, Ph.D., professor of biomedical engineering and director of the Systems Biology Laboratory at Johns Hopkins University School of Medicine and member of the Johns Hopkins Sidney Kimmel Cancer Center. “We hope that our results will not only lead to a better fundamental understanding of cancer, but also to the potential for prognostic guidance for clinicians treating pancreatic cancer.”

By 2022, the National Cancer Institute estimates that more than 62,000 Americans will be diagnosed with pancreatic cancer and nearly 50,000 will die from the disease. On average, only about 10% of people with pancreatic cancer will survive five years. It is difficult to predict which patients are most likely to respond to the few existing treatments; researchers have long sought additional tools — cells, molecules or genes — that stratify pancreatic cancer patients for survival.

In recent years, scientists studying various types of cancer have discovered the importance of non-cancerous cells, molecules and blood vessels that surround tumors, so-called tumor microenvironment. Immune cells are part of this tumor microenvironment; some have the ability to target the tumor for destruction, while others help the tumor escape immune system.

In a previous study of pancreatic cancer, researchers counted how many immune cells were present in the tumor microenvironment and found no link with patient outcomes, but Popel and Johns Hopkins graduate student Haoyang Mi suggested that the physical location of immune cells may be more important than the total number.

In the new study, Popel, Mee, and colleagues at Oregon Health & Science University used a technique called multiplex immunohistochemistry to pinpoint the location of 27 different immune system molecules in surgically removed tumors from 45 people with pancreatic adenocarcinoma, the most common form of cancer. pancreas. . Patients were 52% female, average age 63.5 years, and had all stages of cancer, with 41% of participants having cancer spread to at least four lymph nodes.

Molecules found in different combinations on the surface of different types of immune cells correspond to the relative locations of the subtypes of immune cells.

They then developed new computational algorithms to analyze how these cells differed in number, location and shape between patients who survived longer or shorter than the median survival time of 619 days.

“Z computational approaches we designed, we analyzed not only the density of each cell type, but also how they interact with each other in the spatial architecture of tumors,” says Mee, first author of the new paper.

The researchers found that among 22 patients who survived less than the average (313 days on average), immune cells called IL-10+ myelomonocytes tended to be located near a cluster of granzyme B+ CD8+ T cells (or cytotoxic T- lymphocytes). Among the 23 patients who survived longer than average (832 days on average), the same myelomonocytes were more clustered near another type of T cell known as PD-1+ CD4+ T cells (or activated T cells -helpers).

In light of what is known about the functions of each immune cells, Mee says, the results make sense. Each type of cell acts as a brake on the other. Cytotoxic T-lymphocytes produce a toxin that can destroy cancer cells, but researchers hypothesize that nearby myelomonocytes block this ability. However, in long-term survivors, they believe that activated T-helper cells turn off myelomonocytes, which in turn allow cytotoxic T-lymphocytes to fight the cancer more effectively.

More research is needed to test these hypotheses about how cells interact in the pancreatic tumor microenvironment, the researchers say, and to determine whether targeting either cell type could lead to new immunotherapies for pancreatic cancer. But the researchers hope that more studies will confirm that the association of the tumor microenvironment with survival can provide clinicians with prognostic information and potentially direct patients to specific treatments or clinical trials.

Other researchers involved in the study include Elizabeth Jaffee of the Johns Hopkins Kimmel and Chamilene Sivagnan Cancer Center, Courtney Betts, Shannon Ludal and Lisa Cousens of the Oregon Health & Science University.