The patient lies face down inside an MRI scanner, under general anesthesia, for roughly six hours. Somewhere in that time a narrow cannula, guided by a stereotactic navigation platform, reaches the striatum, the deep structure that Huntington's disease dismantles over decades, and delivers a suspension of neural stem cells manufactured at the UC Davis GMP facility. UCI Health described that procedure on June 23, 2026, when it announced that the first participant had been treated in REGEN4HD, the first in-human trial of an embryonic stem cell-derived cell therapy for Huntington's. The trial registration, posted on ClinicalTrials.gov on March 5, 2026 under the identifier NCT07451613, lists the University of California, Irvine as sponsor, an actual start date of April 6, 2026, and a study identifier, ASCC-26-01, that will mean nothing to most families and everything to the regulators reading it.

What the registry sets out is deliberately modest. This is a phase 1b/2a study, open label and non-randomized, with an estimated enrollment of 21 adults aged 18 to 65 who have a genetically confirmed diagnosis and early-stage disease. Everyone who enrolls gets cells. They are assigned sequentially to four cohorts: unilateral low dose, bilateral low dose, bilateral mid dose, and a fourth, higher step, with twelve people in the phase 1b escalation and nine in the phase 2a expansion at the highest dose judged safe, according to UCI's announcement. The primary outcome is not walking better or thinking more clearly; it is the number of participants with treatment-related adverse events in the first six or twelve weeks after implantation. Everything a family would actually want to know sits in the exploratory column: the Unified Huntington's Disease Rating Scale total motor score, structural MRI, FDG-PET of brain metabolism, and fluid markers including neurofilament light and PENK, all followed out to twelve months. The registry estimates primary completion in June 2028 and full completion in January 2031. HDBuzz, the researcher-written HD news service, put it plainly on June 15, 2026: full data are expected around mid-2028.

The logic separating this from every other clinical programme in Huntington's is worth stating slowly. Gene therapies and oligonucleotides lower a protein. This tries to put cells back, or at least to put supportive cells alongside the ones still alive. The preclinical case rests on mice. In Stem Cell Reports in December 2017, Thompson's group reported that a GMP-grade human embryonic stem cell-derived neural stem cell line improved rotarod, pole test and grip strength performance in R6/2 and Q140 Huntington's model mice, with the transplanted cells showing signs of differentiating into immature neurons and astrocytes. A follow-up in Molecular Therapy in 2023 reported that transplanted human neural stem cells rescued phenotypes in zQ175 mice and sent projections into the striatum. That is mouse work. It says the cells can survive, integrate and change behaviour in a rodent brain over months; it does not say what they will do in a human striatum that has been losing medium spiny neurons for years.

The history behind this trial is the reason to keep both feet on the ground. For three decades, surgeons grafted human fetal ganglionic eminence tissue into the striatum of people with Huntington's, and the results diverged from centre to centre. The largest attempt to settle it, the Multicentric Intracerebral Grafting in Huntington's Disease trial, was published in Movement Disorders in July 2020: of 54 randomized patients, 45 were transplanted, 26 immediately and 19 after a 20-month delay, and the mean total motor score at month 32 did not differ between the groups. The authors reported anti-human leucocyte antigen antibodies in 40 percent of patients and concluded that no clinical benefit was found, possibly related to graft rejection. That finding is why immune management is not a footnote in modern grafting trials. The nearest contemporary precedent comes from another disease: in Nature in April 2025, investigators reported an open-label phase 1 trial of bemdaneprocel, human embryonic stem cell-derived dopamine neuron progenitors grafted bilaterally into the putamen of 12 people with Parkinson's, all of whom received one year of immunosuppression, with safety and graft survival reported at 18 months.

So what does June 2026 actually mean in a household where someone has a CAG expansion and a calendar of appointments? It means one person has had one operation at one site, and the team told NeurologyLive that no serious adverse events had been reported in that first participant. It means twenty more people will be enrolled slowly, in ascending dose steps, precisely because nobody yet knows the safe dose. It does not mean a treatment exists, and the trial is not designed to prove benefit. Leslie Thompson, who has studied Huntington's for more than three decades and whose team received an $11,999,933 grant from the California Institute for Regenerative Medicine in December 2025 to run this study, told Inside Precision Medicine that even if the cells help, researchers "do not know whether this will be sufficient alone", raising the possibility of pairing cell therapy with drugs aimed at mechanisms such as somatic repeat instability.