Stem-cell neurons for Parkinson's disease: what the first human trials measured

If you or someone close to you lives with Parkinson's disease, you may have seen headlines in 2025 about "new brain cells grown from stem cells" being put into patients. The natural question is whether this is a treatment you can ask for yet, and what exactly those trials showed.

This article goes through what was measured, in whom, and what is still unknown.

This material is for information only and does not replace a consultation with a doctor. Treatment of Parkinson's disease is decided by a neurologist.

What would the cells replace?

Parkinson's disease is characterised by the loss of dopamine-producing nerve cells in the midbrain. These cells send their fibres to a deep brain structure called the putamen, which helps control movement. Tablets such as levodopa replace the missing dopamine, but as the disease progresses, the response to medical and surgical treatment becomes weaker and less even.

The idea behind cell therapy is different: instead of supplying dopamine from outside, place new dopamine-producing cells directly into the putamen, so that they settle in and release dopamine themselves.

Has this been tried before?

Yes, and the history explains why the new trials are designed the way they are. Before stem cells, dopamine-producing tissue taken from human embryos was grafted into the brains of people with Parkinson's disease. Open studies without a control group had reported benefit. Then two trials compared the transplant with a control operation: Freed and colleagues used sham surgery, with holes drilled in the skull but the brain's outer membrane left intact; Olanow and colleagues used a placebo procedure.

• Freed and colleagues, 2001. 40 patients aged 34 to 75 with severe disease, on average 14 years since diagnosis. At one year, the primary outcome, a global rating of change, did not differ: 0.0 in the transplant group versus −0.4 after sham surgery. In patients aged 60 or younger, standard motor scores measured in the morning before medication were better in the transplant group than after sham surgery; in older patients in the transplant group there was no significant improvement. The grafts survived: fibre growth was seen in 17 of 20 transplanted patients. After the first year, involuntary movements (dystonia and dyskinesia) returned in 15% of transplanted patients even after levodopa was reduced or stopped.

• Olanow and colleagues, 2003. 34 patients with advanced disease, followed for 24 months. The primary outcome, the motor score off medication, showed no significant overall effect (p = 0.244). Dopamine uptake on scans rose and the grafted cells survived, but 56% of transplanted patients developed dyskinesia that persisted even after medication was withdrawn overnight. The authors concluded that the procedure could not be recommended.

So the grafts lived, yet under sham control there was no overall benefit on the main measure, and a new side effect appeared. The Kyoto authors also name the complications and ethical concerns of fetal tissue as the reason to look for another source. Stem cells offer a product made in a laboratory, in a set dose.

What did the 2025 stem-cell trials find?

Two trials were published together in Nature in April 2025.

Bemdaneprocel, from embryonic stem cells (Tabar and colleagues). An open-label phase I trial in 12 patients. Frozen, ready-made dopamine-neuron precursor cells were grafted into the putamen on both sides: 0.9 million cells in 5 patients and 2.7 million in 7. Everyone received one year of drugs that suppress the immune system.

• The primary goal was safety and tolerability at one year, and it was met: there were no adverse events related to the cell product.

• At 18 months, uptake of a dopamine tracer on PET scans of the putamen had increased, which indicates that the graft survived.

• In the high-dose group of seven patients, with no sham surgery and no blinding, the authors reported an average 23-point improvement on the motor score measured off medication. This was a secondary or exploratory outcome, not the trial's primary goal.

• There were no graft-induced dyskinesias.

Cells from induced pluripotent stem cells, Kyoto (Sawamoto and colleagues). A phase I/II trial in 7 patients aged 50 to 69. The cells came from donor stem cells (not the patients' own) and were grafted into both sides of the brain; patients were followed for 24 months.

• There were no serious adverse events; 73 mild to moderate events were recorded.

• Patients' Parkinson's medication doses were kept unchanged unless an adjustment was needed, and the authors report that this resulted in increased dyskinesia.

• MRI showed no overgrowth of the graft, and no tumours formed.

• Of 6 patients assessed for efficacy in this open-label trial with no sham group, the authors reported a better motor score off medication in 4 and on medication in 5. The authors reported an average change across all six of 9.5 points (20.4%) off medication and 4.3 points (35.7%) on medication.

• Dopamine uptake in the putamen on PET rose by 44.7%, more in the higher-dose group.

One person's own cells. In 2020, a team in Boston described a single patient who received dopamine precursor cells made from the patient's own reprogrammed cells, without immune suppression. PET suggested that the graft survived, and symptoms were stable or improved 18 to 24 months later. This is a case report of one patient, not a trial.

What do these results not tell you?

They do not show that the treatment works better than surgery without cells. Neither 2025 trial had a sham or placebo group, and both patients and doctors knew who had been treated. That is exactly where the fetal-tissue story turned: open studies reported benefit, and the sham-controlled trials did not confirm it on their main measure.

The numbers are very small. Twelve patients and seven patients, split further by dose. An average of 23 points comes from 7 people in one group, without blinding.

The trials were designed for safety, and safety is what they showed — over 12 to 24 months. Late effects matter here: in the 2001 trial, dystonia and dyskinesias recurred after the first year.

These are specific products. The results apply to these cells, made and tested under trial conditions. They say nothing about other "stem cell" injections offered outside registered trials.

What is being tested now?

• exPDite-2 (NCT06944522), a phase 3 trial of bemdaneprocel. The trial is designed to randomise about 102 adults aged 45 to 75 in a 2:1 ratio to the cells or sham surgery, with participants, doctors and assessors all blinded. Participants receive immune-suppressing drugs or a matching placebo for about 12 months. The primary outcome is the daily time with good symptom control and without troublesome dyskinesia, recorded in a patient diary, measured at 78 weeks. Among the trials described here, it is the first to compare a stem-cell product with sham surgery.

• ANPD001 (NCT06344026), a phase 1/2a trial in which cells are made from each participant's own skin cells. It is open-label, with planned dose escalation in 9 participants, and has safety at one year as its primary outcome.

What should you do with this now?

• Treat these results as a sign that the approach is being tested properly, not as a treatment that is available.

• If you are interested in a trial, ask your neurologist whether you could meet the criteria. Trials are listed on ClinicalTrials.gov; exPDite-2 is recruiting adults aged 45 to 75, and ANPD001 is enrolling by invitation adults aged 40 to 75.

• Be cautious with clinics that sell "stem cell therapy for Parkinson's" outside a registered trial: the data above do not apply to them.

• The tablets, adjustments and surgical options that your neurologist already uses remain the standard; nothing here changes them.

Sources

• Tabar V, et al. Phase I trial of hES cell-derived dopaminergic neurons for Parkinson's disease. Nature. 2025. PMID 40240592

• Sawamoto N, et al. Phase I/II trial of iPS-cell-derived dopaminergic cells for Parkinson's disease. Nature. 2025. PMID 40240591

• Freed CR, et al. Transplantation of embryonic dopamine neurons for severe Parkinson's disease. N Engl J Med. 2001. PMID 11236774

• Olanow CW, et al. A double-blind controlled trial of bilateral fetal nigral transplantation in Parkinson's disease. Ann Neurol. 2003. PMID 12953276

• Schweitzer JS, et al. Personalized iPSC-derived dopamine progenitor cells for Parkinson's disease. N Engl J Med. 2020. PMID 32402162

• exPDite-2: phase 3 study of bemdaneprocel in Parkinson's disease. ClinicalTrials.gov. NCT06944522

• Phase 1/2a study of ANPD001 in Parkinson disease. ClinicalTrials.gov. NCT06344026

• Phase 1 study of MSK-DA01 (bemdaneprocel) in advanced Parkinson's disease. ClinicalTrials.gov. NCT04802733