A team led by researchers at Karolinska Institutet in Stockholm has grown three-dimensional clusters of insulin-producing cells from human stem cells and shown that, once transplanted into diabetic mice, the clusters matured enough to regulate blood sugar without any external insulin. The work, built across several peer-reviewed studies published between 2019 and 2025, represents one of the most detailed preclinical demonstrations to date that lab-grown pancreatic tissue can function on its own inside a living animal.
For the roughly 8.75 million people worldwide living with type 1 diabetes, a condition in which the immune system destroys the pancreas’s native insulin-producing beta cells, the findings sharpen a question that has driven regenerative medicine for more than a decade: can replacement cells grown in a lab eventually eliminate the need for daily insulin injections?
What the experiments showed
The research is a collaboration between Karolinska Institutet, Uppsala University, and the University of Helsinki. The team differentiated human pluripotent stem cells into islet-like clusters and then ran them through glucose-stimulated insulin secretion assays and deep metabolic profiling, as detailed in Nature Biotechnology. The clusters sensed rising glucose and responded by releasing insulin, closely mimicking the behavior of natural pancreatic islets.
A key insight, published separately in Nature Cell Biology, was that the physical architecture of the cells mattered enormously. Arranging endocrine cells into three-dimensional clusters rather than flat layers activated maturation programs that improved the cells’ ability to couple glucose sensing to insulin release. In practical terms, the clusters arrived in the body already partially functional, showing glucose-responsive insulin secretion shortly after engraftment.
The most striking result came from transplant experiments. When researchers placed sufficiently large quantities of stem-cell-derived islet cells under the kidney capsule of diabetic mice, the animals’ blood glucose levels normalized and their pre-existing diabetes was effectively reversed, according to a study in Stem Cells Translational Medicine. Crucially, this reversal held only while the graft remained in place; removing the transplanted cells caused blood sugar to rise again. Smaller doses of the same cells did not achieve that result, establishing that dosage is a decisive variable.
A 2025 study in Diabetologia added a further dimension: the transplanted beta cells continued to mature metabolically after engraftment. The tissue improved its performance over weeks inside the host rather than plateauing at the level it reached in the dish.
Separately, a related Nature Chemical Biology paper from the broader collaboration explored chemical strategies for steering stem-cell differentiation toward functional beta cells. Karolinska Institutet highlighted that work in institutional communications published on Phys.org, offering additional context on the group’s approach.
What has not been proven yet
Every result described above was obtained in mice. No clinical trial data exist for these specific stem-cell-derived islet clusters, and as of June 2026, no regulatory body has issued public guidance or approval related to this particular approach.
The gap between reversing diabetes in a mouse kidney capsule and doing so safely inside a human body is wide. Scaling production from laboratory batches to the billions of cells a single patient would likely require involves manufacturing, quality-control, and cost hurdles that the published papers do not address in detail. No institutional cost breakdown or production-yield analysis from the Karolinska team has appeared in the public record.
Immune rejection may be the tallest barrier. Type 1 diabetes is an autoimmune disease, and any transplanted beta cells face the same immune attack that destroyed the patient’s original cells. The published studies do not report long-term immunological outcomes. It remains unclear whether the clusters would survive in humans without lifelong immunosuppressive drugs or some form of protective encapsulation.
How this fits into the broader race for a cell therapy
The Karolinska-led work is not happening in isolation. Vertex Pharmaceuticals’ VX-880, a stem-cell-derived islet therapy, has already produced early human data: in a small trial, some patients achieved insulin independence, though they required immunosuppression. Vertex is also testing VX-264, which encases the cells in a device designed to shield them from immune attack, potentially removing the need for immunosuppressive drugs. Other groups, including programs spun out of Harvard’s Doug Melton lab, are pursuing related strategies.
What distinguishes the Swedish research is the depth of its characterization. The Nature Biotechnology paper offers unusually granular metabolic profiling of the stem-cell-derived clusters. The dose-response transplant data provide a clear framework for thinking about how many cells a therapy would need. Together, the studies give the field a more precise map of what functional maturity looks like in lab-grown beta cells and what it takes to push them across the threshold of therapeutic usefulness.
What this means for people managing diabetes now
The practical takeaway is measured but real. These findings confirm that human stem cells can be coaxed into functional, glucose-responsive beta-cell clusters and that those clusters can normalize blood sugar in an animal model when delivered in adequate doses. That is a meaningful preclinical milestone, not a treatment.
No human efficacy data, regulatory timeline, or cost estimate exists for this specific technology. Anyone managing type 1 diabetes today should continue following their current treatment plan. The next concrete step to watch for is the announcement of a formal clinical trial, which would move the work from the animal lab into the realm of human medicine for the first time.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
