More than 100,000 Americans are waiting for a kidney transplant, and most will wait years. A study published in May 2026 in npj Biomedical Innovations, a peer-reviewed Nature Portfolio journal, offers a small but concrete reason for hope: a team led by Ryuji Morizane at Massachusetts General Hospital and Harvard Medical School grew sheets of human kidney tissue from stem cells, implanted them in mice, and watched the tissue filter molecules out of the bloodstream on its own, performing the same sieving job a natural kidney does every second of every day.
It is not a replacement organ. It is not close to one. But it is the first time lab-grown human kidney tissue has demonstrably done real filtration work inside a living animal, and that distinction matters.
What the researchers actually did
The team, whose work is detailed in the journal, started with induced pluripotent stem cells (iPSCs), adult cells reprogrammed to behave like embryonic ones. From those, they generated what they call “nephron sheets,” flat expanses of tissue containing the structural building blocks of the kidney’s filtration units, known as glomeruli. Critically, the sheets also incorporated human endothelial cells, the type that line blood vessels, an engineering choice designed to help the tissue connect with a host’s circulatory system after implantation.
Once implanted in mice, the nephron sheets did exactly that. Over the days following surgery, mouse blood vessels grew into the engineered tissue, supplying it with flowing blood. To test whether the tissue was actually filtering, the researchers injected two fluorescent tracer molecules into the animals’ bloodstreams: a small 3 kDa dextran, light enough to pass through a healthy kidney filter, and a large 500 kDa dextran, far too bulky to cross.
Using intravital multiphoton imaging, a technique that allows scientists to watch biological processes in real time inside a living animal, the team confirmed that the small tracer crossed the engineered barrier while the large one was held back. That size-selective sieving is the defining function of a working glomerulus.
Why this is different from earlier organoid work
Kidney organoids, tiny three-dimensional structures grown from stem cells, have been around for roughly a decade. Previous studies showed that transplanting them under a mouse kidney capsule could coax them to develop blood vessels, and that exposing organoids to controlled fluid flow on microfluidic chips accelerated their maturation. A prior study in the Journal of the American Society of Nephrology even used the same two-dextran imaging protocol to assess glomerular sieving in transplanted iPSC-derived kidney organoids.
What separates the new paper is its focus on scalability and uniformity. Rather than growing individual organoids, the researchers produced nephron sheets with integrated vasculature-friendly cells, a format intended to yield larger, more consistent tissue. The functional result, filtration confirmed in a living system, builds directly on the earlier proof-of-concept work but pushes it closer to the kind of engineered tissue that might one day be manufactured at a meaningful scale.
The enormous gap that remains
Filtering two sizes of sugar molecules in a mouse is a far cry from replacing a human kidney, and the researchers’ own data make that clear.
First, the implanted tissue does not produce urine. A working kidney does far more than sieve blood: it reabsorbs water and essential nutrients, concentrates waste, balances electrolytes, regulates blood pressure, and secretes hormones like erythropoietin. The verified filtration covers only the very first step in that chain.
Second, the study tracked function over a short post-implantation window. No published data show whether these structures maintain filtration over weeks or months, whether they provoke immune rejection, or how they would behave in a larger animal. The paper includes no comparison of the engineered tissue’s filtration rate against a native human glomerulus or against existing dialysis technology, making it difficult to gauge how much functional ground still needs to be covered.
Third, scale is a formidable obstacle. A single human kidney contains roughly one million nephrons. The current nephron sheets are sized for microscopy, not transplantation. No regulatory pathway for human use has been proposed, and no timeline for clinical trials appears anywhere in the published research.
Where this fits in the bigger picture
The kidney is one of the hardest organs to engineer because of its intricate architecture and the sheer number of functional units it requires. Other approaches to the organ shortage are further along in some respects: pig kidneys modified with gene editing have been transplanted into living human patients in recent clinical cases, though those efforts carry their own unresolved questions about long-term viability and immune compatibility.
Stem-cell-derived tissue and xenotransplantation are not necessarily competing strategies. If lab-grown nephron sheets can eventually be scaled and shown to function long-term, they could offer a patient-specific alternative, built from a recipient’s own reprogrammed cells, that sidesteps the immune barriers inherent in using animal organs. That possibility is what makes even a modest functional milestone worth paying attention to.
What the nephron-sheet filtration result does and does not prove
The verified claim is narrow but significant: human stem-cell-derived kidney tissue, engineered as nephron sheets by Morizane’s group at Massachusetts General Hospital, performed size-selective filtration after implantation in a living animal. For the patients on transplant waiting lists measured in years, the practical impact remains distant. But the science has crossed a threshold that earlier organoid work had not, proving that lab-grown human kidney tissue can do real work inside a living body. The next questions, durability, scale, and safety, will determine whether that threshold leads anywhere patients can follow.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
