New high-risk form of diabetic kidney disease linked to immune cells revealed: Study

A detailed new map of the human kidney revealed a previously unrecognized form of diabetic kidney disease (DKD) marked by clusters of immune cells-specifically B cells-that are linked to faster disease progression. The findings could help guide more targeted treatments in the future, according to study results led by researchers at the Perelman School of Medicine at the University of Pennsylvania and published today in Nature.

“Diabetic kidney disease has often been treated as a single condition, but patients can have very different outcomes,” said senior author Katalin Susztak, MD, PhD, a professor in Renal Electrolyte and Hypertension and co-director of the Penn/CHOP Kidney Innovation Center. “By looking directly at the kidney tissue, we can now see different disease processes and start to match treatments to what’s actually happening in each patient.”

Diabetic kidney disease affects 20%-40% of all people with diabetes, making it a leading cause of chronic kidney disease (CKD) and end‑stage kidney disease (ESKD). In the United States, roughly one in three adults with diabetes has CKD. Globally, the burden continues to rise: chronic type 2 DKD affected more than 107 million people in 2021, an 85% increase since 1990, driven by population growth, aging, and longer survival with diabetes.

Doctors currently rely on measures like kidney function and protein levels in the urine to detect CKD, but these don’t explain why the disease progresses differently from person to person.

Mapping how the disease changes inside the kidney

The researchers used a new technology that allows them to study gene activity directly in tissue samples while keeping the structure of the tissue intact. This view reveals not only which cells are present, but also where they are located and how they interact. The team analyzed kidney samples from dozens of patients, studying more than five million cells and identified patterns of how cells group together and interact—essentially creating a map of how the disease changes the structure of the kidney. One important finding was a tissue pattern associated with scarring and inflammation that became more common as the disease worsened.

Discovering a surprising role for B cells

Within these inflamed areas, the researchers made a key discovery: in some patients, B cells—an important type of immune cell that produce antibodies and orchestrate broader immune responses—formed organized clusters that resemble structures normally seen in autoimmune diseases. Patients with these B cell–rich regions experienced much faster progression to kidney failure. “This was unexpected,” Susztak said. “B cells haven’t traditionally been seen as a major driver of diabetic kidney disease. But in some patients, they form organized immune structures that appear to make the disease worse.” These areas also contained other immune cells that support B cells, suggesting that active immune responses are happening directly within the kidney.

Creating a test for better prediction and treatment

The researchers then used this information to develop tools that could identify this high-risk form of disease without needing detailed tissue mapping. They created a gene-based signature and a blood test that can help predict which patients are more likely to experience rapid disease progression.

The study suggests that diabetic kidney disease is not a single condition, but a group of different disease types. Recognizing these differences could help doctors choose more effective treatments, especially as new therapies targeting the immune system become available. “Understanding how inflammation is organized within the kidney gives us a new way to classify disease,” said Bernhard Dumoulin, MD, a postdoctoral fellow in the Susztak lab and first author of the study. “This could lead to more precise treatments tailored to each patient.”

The study also demonstrates the power of interrogating disease directly in tissue using an unbiased, near–genome-wide spatial approach, revealing biological patterns and disease processes that may be missed when cells are studied in isolation. This strategy could reshape how complex diseases are discovered and treated.

Reference:

Dumoulin, B., et al. (2026). Spatial atlas of diabetic kidney disease reveals a B cell-rich subgroup. Nature. DOI: 10.1038/s41586-026-10363-4. https://www.nature.com/articles/s41586-026-10363-4