More delta cells meant less insulin being secreted.
For generations, the mechanisms behind diabetes have been studied at a remove — through animal models, approximations, and inference. Now, a landmark analysis of pancreatic tissue from 299 human organ donors has brought the cellular architecture of diabetes risk into direct view, revealing that an imbalance in the tiny hormone-producing clusters of the pancreas — particularly an excess of insulin-suppressing delta cells — may be a key reason some people's bodies quietly lose the ability to regulate blood sugar. Published in May 2026 by researchers at the Integrated Islet Distribution Program, the findings remind us that the most consequential biological truths are often hidden in proportions, not absolutes.
- Diabetes claims more American lives than most people recognize, yet its root cellular mechanisms have long remained obscured by the limitations of animal-based research.
- The discovery that higher genetic risk for Type 2 diabetes correlates with more delta cells — which actively suppress insulin — gives scientists their clearest mechanistic picture yet of how the disease takes hold.
- Cell composition varies meaningfully by sex and ancestry, exposing the inadequacy of universal treatment models and pressing researchers toward more personalized approaches.
- The gene HHEX and over 300 other diabetes-linked variants appear especially active in delta cells, suggesting that genetics shapes not just risk but the physical structure of the islet itself.
- The full dataset has been made publicly available, opening a new collaborative frontier for stem cell therapy, CRISPR-based beta-cell replacement, and islet transplantation research.
- The work now turns toward studying how these cellular compositions shift in people with prediabetes or active disease — the next threshold in translating discovery into treatment.
Diabetes affects one in eight Americans, costs the country over $400 billion annually, and ranks among its leading causes of death — yet the precise cellular reasons why some pancreases fail have remained stubbornly difficult to see. For decades, researchers worked largely through mouse models, which differ from humans in fundamental ways. To move closer to the truth, scientists needed human tissue.
The Integrated Islet Distribution Program at City of Hope in California has built the largest repository of human islets for research in the United States. In May 2026, the program published a sweeping analysis in Nature Communications, drawing on pancreatic tissue from 299 organ donors without diabetes. The scale was unprecedented. So were the findings.
Pancreatic islets contain three main cell types: beta cells, which produce insulin; alpha cells, which raise blood sugar; and delta cells, which suppress insulin through a hormone called somatostatin. The researchers discovered that donors carrying genetic variants associated with higher Type 2 diabetes risk tended to have a greater proportion of delta cells — meaning more insulin suppression and less secretion capacity. The mechanism, long suspected, was now visible in human tissue.
The study also found that cell composition varied by sex and ancestry, challenging the assumption that a single prevention or treatment model can serve all populations. More than 300 diabetes-linked genes were examined, and several — including HHEX — proved especially active in delta cells, suggesting that genetics shapes the very architecture of the islet, not just its function.
The implications extend across multiple frontiers of medicine. Scientists engineering stem cell-derived islets now have a clearer blueprint for what a healthy islet should contain. CRISPR researchers developing immune-evasive replacement cells have better targets for cellular balance. And clinicians pursuing islet transplantation have new evidence about what makes some islets more viable than others.
The full dataset — islet images, genetic profiles, and functional measurements — has been made publicly available, inviting the broader research community to build on these findings. Future work will examine how these cellular compositions shift in people with prediabetes or active disease. The researchers closed by honoring the organ donors whose gift made the study possible, and by calling for fewer barriers to beta-cell replacement therapy as safer immunosuppressive options emerge. A door has been opened.
Diabetes kills more Americans than most people realize. It affects one in eight of us, costs the country over $400 billion a year, and ranks as the eighth leading cause of death. The disease boils down to a simple failure: the body cannot make or use enough insulin to keep blood sugar in check. But understanding why that happens—why some people's pancreases fail while others' do not—has proven far more complicated than the basic definition suggests.
For decades, researchers relied on mouse models to study diabetes. Mice are useful, but their pancreases work differently from ours. The tiny organs called islets, which contain the cells that produce insulin and regulate blood glucose, have fundamentally different structures and behaviors in rodents than in humans. To truly understand diabetes risk, scientists needed to study actual human tissue. That meant getting access to pancreases from organ donors and examining them with precision.
A consortium called the Integrated Islet Distribution Program, housed at City of Hope in California, has spent years building the largest repository of human islets for research in the United States. In May 2026, researchers from that program published their most comprehensive analysis yet in Nature Communications. They had studied pancreatic tissue from 299 organ donors without diabetes, applying standardized methods to measure cell composition, genetic risk, and how well the islets functioned. The scale was unprecedented. The findings were striking.
Pancreatic islets contain three main types of endocrine cells, each with a different job. Beta cells produce insulin. Alpha cells produce glucagon, which raises blood sugar. Delta cells produce somatostatin, which does something counterintuitive: it suppresses insulin secretion. Delta cells are relatively rare—they make up a small percentage of the islet—but their influence is outsized. The research team discovered that people carrying genetic variants that increase Type 2 diabetes risk were more likely to have a higher proportion of delta cells in their islets. More delta cells meant more somatostatin being released, which meant less insulin being secreted. The mechanism was suddenly visible.
The team also found that cell composition was not uniform across all humans. Women and men had different ratios of alpha, beta, and delta cells. People of different ancestries showed variations too. This matters because it suggests that diabetes prevention and treatment may need to be tailored to different populations, not applied as a one-size-fits-all approach. The researchers examined over 300 genes linked to Type 2 diabetes risk and found that several of them—including one called HHEX—were particularly active in delta cells, pointing to how genetics shapes the very architecture of the islet.
The implications ripple outward. Scientists working on stem cell-derived islets—lab-grown replacements for damaged pancreatic tissue—now have a clearer picture of what a healthy islet should look like. Researchers developing CRISPR-edited cells to evade the immune system have a better understanding of the cellular balance they need to recreate. And clinicians hoping to expand islet transplantation as a treatment for diabetes have new evidence about what makes some islets more functional than others.
The study was made possible by thousands of organ donors and their families who consented to have pancreatic tissue used for research. The researchers emphasized this in their closing statement, noting that working with human islets is both a privilege and a responsibility. They hope the work will raise awareness of organ donation for research and remove barriers to beta cell replacement therapy, especially as new immunosuppressive drugs become available that could make transplantation safer and more practical.
The dataset is now public. Other researchers can access the images of islets, the genetic information, and the functional measurements through online platforms. This democratization of data—making it available to the broader diabetes research community—accelerates the pace of discovery. Future studies will layer on additional complexity: examining how these cellular compositions change in people with prediabetes or active diabetes, integrating new types of molecular data, and testing whether the findings hold across different donor populations. The work has opened a door. What comes next depends on how many researchers walk through it.
Citas Notables
Working with human islets is a great privilege and responsibility, as each human islet preparation is accompanied by family consent for organ donation.— The IIDP research team
La Conversación del Hearth Otra perspectiva de la historia
Why does it matter that delta cells suppress insulin? Couldn't the body just make more beta cells to compensate?
The body doesn't work that way. If delta cells are constantly suppressing insulin secretion, no amount of extra beta cells can overcome that brake. It's like pressing the gas and the brake at the same time. The genetic variants that increase diabetes risk seem to tip the balance toward more delta cells, which means the brake is always partially on.
So this explains why some people with the genetic risk actually develop diabetes and others don't?
It's part of the explanation. Genetics loads the gun, but environment pulls the trigger. This study shows the mechanism—the cellular composition that makes someone vulnerable. But diet, exercise, weight, and stress all matter too. What's new here is understanding the biological substrate that genetic risk acts on.
You mentioned the findings differ by sex and ancestry. Does that mean we need completely different treatments for different groups?
Not necessarily different treatments, but personalized ones. If women tend to have different islet compositions than men, or if people of African ancestry have different ratios than people of European ancestry, then prevention strategies and drug targets might need adjustment. One diabetes prevention program won't fit everyone equally well.
The researchers mentioned CRISPR-edited islets. How close are we to that actually working in patients?
Still years away, but this study helps. You need to know what you're trying to build before you can build it. Now researchers have a detailed blueprint of a healthy human islet—the right balance of alpha, beta, and delta cells, the genes that should be active. CRISPR can edit genes, but you also need to understand cell composition and function. This study provides that understanding.
Why did it take until 2026 to do this study if the islet repository has existed for years?
Scale and standardization. You can't just grab 299 pancreases and compare them. Every islet preparation is slightly different depending on how it was isolated, how long it was stored, who processed it. The consortium spent years developing standardized methods so that data from different labs could be meaningfully combined. That rigor takes time, but it's what makes the findings trustworthy.