Four People, Four Years, and No Tumours

A complete cervical spinal cord injury is one of the few diagnoses in medicine that has stayed genuinely fixed. The cord does not regenerate, the circuitry below the break does not reconnect, and the deficit a patient has some weeks after the accident is very close to the deficit they will have for the rest of their life. Rehabilitation can make enormous differences to what someone is able to do. Nothing in the clinic repairs the tissue itself.

A team at Keio University in Tokyo has now put the most-discussed candidate for changing that into people for the first time. Four patients with complete cervical injuries received transplants of neural stem and progenitor cells grown from induced pluripotent stem cells, the reprogrammed adult cells whose discovery won Shinya Yamanaka a Nobel Prize and who is a co-author here. After two to four years of follow-up, no tumours had formed, and two of the four patients recovered enough function to move off the most severe injury grade entirely.

Why it matters: The fear that has shadowed iPSC therapy since the beginning is that cells with the capacity to become anything might become a tumour. Four people carrying these grafts for years without one is the observation the field has been waiting for, and it is what makes the larger trials worth running.

Why safety was the whole point

The design here is a phase 1, open-label study in four people, and the primary endpoint was safety alone. That is the correct order of operations for a therapy of this kind, and it is worth understanding why the bar sits there. Induced pluripotent stem cells are adult cells reprogrammed back to an embryonic-like state, which is precisely what makes them useful and precisely what makes them dangerous. A cell that can become any tissue can also proliferate without restraint. Any residual undifferentiated cell in the transplanted population is a candidate seed for a growth in the spinal cord, which is close to the worst possible place to discover you have one.

The trial cleared that bar. No tumour formation and no graft-related adverse events across two to four years, with imaging showing the graft sites stable where they were placed. The patients were on immunosuppression only for a short period, which matters for a therapy meant to be practical rather than heroic.

What happened to the patients

Efficacy was a secondary, exploratory endpoint, which is the honest framing for four people with no control group. Measured on the standard international motor score for spinal cord injury, the median improvement from baseline at two weeks after injury to one year was 13 points, with a range running from 10 to 40. Two of the four patients improved on the American Spinal Injury Association Impairment Scale, one from grade A to C and one from A to D. Grade A is complete injury, meaning no motor or sensory function preserved in the lowest sacral segments. Moving off it means something below the injury started working again.

The authors compare these gains against spontaneous recovery in a registry-based cohort and describe them as numerically greater, which is careful and correct phrasing for a comparison that is not a randomised one. Their conclusion, as Sugai and colleagues put it in Nature Medicine, is that transplantation of human iPSC-NS/PCs into the injured spinal cord is feasible and safe under short-term immunosuppression, with the findings supporting further clinical evaluation. That is a claim about feasibility, not a claim about cure.

What the study can't say yet

Four patients cannot establish that this treatment works. There is no control arm, the efficacy analysis was exploratory, and spinal cord injury has a genuine and variable spontaneous recovery curve in the months after trauma, which is the single hardest confound in this field. A registry comparison narrows that problem without solving it. Two patients improving in grade is encouraging and is not evidence of efficacy in the sense a regulator would need.

The timing is also a real constraint on who this could help. These were subacute injuries, treated in the window after the initial trauma has settled but before the injury site has fully matured into chronic scar. Whether transplanted cells can do anything in an injury that is years old is a different and harder question, and this trial does not address it. Nor does it establish which of the recovered function came from the graft rebuilding circuitry as opposed to supporting tissue that survived.

Quick questions

Does this mean spinal cord injury is treatable now? No. This is a four-person safety trial, and the authors present it as grounds for further evaluation rather than as a demonstration that the therapy works.

Why does the absence of tumours matter so much? Because it is the specific risk that has held iPSC-derived therapies back. Years of follow-up in humans without tumour formation is the evidence needed before larger trials are ethical to run.

What's the one-line takeaway? Four people with complete cervical spinal cord injuries received iPSC-derived neural progenitor transplants, and after two to four years there were no tumours and two had regained some function below the injury.