Researchers from the University of Pennsylvania analyzed nearly one million immune cells from organ donors, identifying a unique subpopulation of CD4 cells in pancreatic lymph nodes and changes in B cells in the spleen. Their findings point toward developing a simple blood test that could detect the immune failure leading to type 1 diabetes before insulin-producing cells are permanently lost.
A Window Before Destruction
Type 1 diabetes (T1D) affects nearly two million Americans, and by the time most people are diagnosed, the majority of insulin-producing cells in the pancreas have already been destroyed. Now, identification of a group of "attacker" immune cells hidden in the pancreatic lymph nodes—appearing earlier in the disease process—may provide the first real opportunity for early detection and even prevention of type 1 diabetes. This emerges from new research published in Science Immunology by the Perelman School of Medicine at the University of Pennsylvania.
"For the first time, the research managed to catch the attacking cells 'in the act,' while the disease is still developing. We're not just seeing the destruction after the immune system has already eliminated the insulin-producing cells in the pancreas," said Professor Golnaz Vahedi, a genetics researcher and senior author of the paper.
The researchers analyzed nearly one million immune cells—one cell at a time—taken from pancreatic lymph nodes and spleens of 43 organ donors: some with active type 1 diabetes, some at an early stage of disease risk, and some healthy.
This analysis identified a unique subpopulation of CD4 T cells—a type of "helper cell" of the immune system—in the pancreatic lymph nodes of people with active type 1 diabetes. These cells strongly "turn on" two proteins, NFKB1 and BACH2, which function as master switches: they activate and deactivate groups of genes in a way that intensifies the immune attack on insulin-producing cells (beta cells).
"The study showed that the same cell pattern also appears in people at the pre-type 1 diabetes stage, who still show no clinical symptoms," Vahedi said. "This suggests that immune failure begins early, at a stage where healthy beta cell reserves still exist that could potentially be saved."
In the spleen, researchers also identified molecular changes unique to type 1 diabetes in B cells—another type of white blood cell. More importantly, these signals can also be detected in simple blood samples.
This points to the possibility of developing a blood test in the future that would identify risk for type 1 diabetes years before symptoms such as high blood sugar appear.
"The changes in the spleen that can be identified in blood mean we could monitor at-risk children—for example, family members of type 1 diabetes patients—without the need for invasive procedures," Vahedi explains. "If we succeed in blocking the pathways that fuel these 'rogue' CD4 cells, we might be able to delay or even prevent type 1 diabetes altogether."
The research team worked with pancreas samples and lymph node tissue donated by deceased organ donors and their families.
"The work of the Gift of Life Donor Program staff, Penn transplant surgeons, collection teams, laboratory personnel—and especially the donors themselves—is what made this research possible," Vahedi emphasizes.
To date, researchers have analyzed pancreatic tissue and lymph nodes from more than two hundred organ donors, producing a large-scale database available to the entire scientific community through PANC-DB, a public database shared with Vanderbilt University, the University of Florida, and Stanford.
"Each dataset represents countless late nights and one donor's gift," says Professor Robert Faryabi, a pathologist and disease researcher, co-editor of the study. "This is collaborative science at its peak—surgeons, scientists, families, and donors together making the impossible possible."
The research is part of the Human Pancreas Analysis Program (HPAP), led by Professor Ali Naji and Professor Klaus Kaestner. The program was established in 2016 with funding from the U.S. National Institutes of Health (NIH), with the goal of systematically studying the pancreas and understanding exactly what goes wrong before and during the development of type 1 and type 2 diabetes. Recently, the program's funding was extended for an additional four years.
Now, Vahedi and her colleagues aim to develop artificial intelligence models that can not only identify type 1 diabetes earlier but also map the disease at the molecular level.
"Our goal is to teach AI systems the molecular language of type 1 diabetes—to train them on disease cells in pancreatic lymph nodes so they can discover their subtle traces in blood, even when they're like a needle in a haystack," Vahedi said.
