Broken Bones, Modified Cells: A Small Trial With Large Implications

Posted on 15 September, 2026 by Mehrdad Fathi

Osteoporosis is one of the most common conditions in aging medicine, and one of the most consequential. Affecting an estimated 200 million women worldwide, it progressively weakens bone architecture until fractures become almost routine — after a stumble, a cough, a minor fall. Existing drug therapies slow that deterioration, but they do not rebuild what has been lost. A small clinical trial published today in Cell suggests that a one-time infusion of modified bone-marrow cells may do something closer to that.

The Mechanism: Getting Cells Where They Need to Go

The central challenge with using mesenchymal stromal cells (MSCs) — progenitor cells capable of generating skeletal tissue — for bone repair has always been delivery. Injected into the bloodstream, MSCs simply do not make it to bone tissue in useful numbers. They lack the surface signals that would allow them to slow down at the bone marrow’s blood-vessel walls and migrate through.

Regenerative medicine specialist Robert Sackstein, now at the Miami Veterans Affairs Medical Center, spent decades working on exactly this problem. In 2008, his team reported that attaching fucose — a simple sugar molecule — to the surface of MSCs gave them a previously absent ability to seek out and enter bone. The fucose modification creates interactions with proteins lining blood vessel walls in skeletal tissue, slowing the cells enough to allow them to squeeze through into the marrow. The modification is trining blood vessel walls in skeletal tissue, slowing the cells enough to allow them to squeeze through into the marrow. The modification is transient,ice, Sackstein and collaborators spent several years developing a manufacturing process suitable for human use. By 2015, clinical teams in Spain, led by bone marrow transplant specialist José Moraleda at the University of Murcia, were ready to begin the trial.

What the Trial Did and Found

Ten women aged 51 to 72 — all with advanced osteoporosis and significant fracture histories — received a single intravenous infusion of their own bone marrow cells, extracted, fucose-modified in the laboratory, and returned. Most participants continued their standard osteoporosis medications throughout.

The follow-up period, with a median of six years, produced several notable signals. The most striking was fracture frequency. Before the infusion, participants collectively experienced fractures from low-impact events roughly every year or two on average. After treatment, the rate fell sharply — closer to once per decade. Bone density scans and biopsies indicated measurable improvements in bone size and tissue quality. Participants reported reductions in pain and functional disability. And levels of proteins associated with bone formation rose in blood tests.

No elevated incidence of cancer or other serious adverse events was detected across the follow-up period.

“It’s quite remarkable,” said Ajit Varki of the University of California, San Diego, who reviewed the findings. The trial “showed an almost 100% efficacy sustained for several years — and no side effects.”

The Caveats Are Significant

Any honest reading of these results has to sit with the study’s limitations, which the authors and outside commentators were candid about.

Ten participants is a very small number. There was no control group, making it impossible to isolate the treatment effect from concurrent drug therapy, natural disease variation, or regression to the mean. Most critically, the researchers did not directly track the modified cells after infusion. Whether enough fucose-tagged MSCs actually reached bone tissue in meaningful concentrations — the biological premise the entire intervention rests on — was not confirmed in the human subjects.

Richard Eastell, an osteoporosis researcher at the University of Sheffield, noted that the protein changes suggestive of increased bone formation could also reflect accelerated bone turnover, a process that can indicate breakdown as well as building. Without a control arm, distinguishing the two is not straightforward.

Where the Evidence Points Next

Moraleda is planning a follow-up trial of approximately 100 participants, randomized across three arms: fucose-enhanced autologous MSCs (derived from the recipient’s own marrow), fucose-enhanced allogeneic MSCs (from healthy donors), and standard drug therapy alone as a comparator. The donor-cell arm is particularly significant — if off-the-shelf MSCs from healthy donors perform comparably to patient-derived cells, the treatment becomes substantially more scalable.

That larger, controlled trial will be necessary to resolve the core ambiguity: are the modified cells doing what the mechanism predicts, and is the fracture reduction a reproducible treatment effect? As Jasmeen Merzaban of the King Abdullah University of Science and Technology, who trained in Sackstein’s lab and contributed to the original 2008 fucose research, put it: “Now, we need a larger, randomized trial to really answer some of the questions.”

What the current findings provide is a safety signal that is genuinely encouraging and a biological rationale that has been in development for nearly two decades. For a condition where the standard of care stabilizes bone loss but cannot restore it, that combination justifies the follow-up work.

Source: Moraleda, J.M. et al. Cell (2026). DOI: 10.1016/j.cell.2026.08.017. Reported by Nature News, doi: 10.1038/d41586-026-02828-3.

References:

1. Moraleda, J.M. et al. Fucose-engineered mesenchymal stromal cells for bone regeneration in osteoporosis: a phase I/II clinical trial. Cell (2026). DOI: 10.1016/j.cell.2026.08.017

2. Sackstein, R. et al. Ex vivo glycan engineering of CD44 programs human multipotent mesenchymal stromal cell trafficking to bone. Nature Medicine 14, 181–187 (2008). DOI: 10.1038/nm1703

3. López-Lucas, M.D. et al. Manufacture and characterization of fucosylated bone marrow-derived human mesenchymal stromal cells for clinical use. Cytotherapy 20, 1110–1123 (2018). DOI: 10.1016/j.jcyt.2018.07.006