Let's BioTech

This article is based on my Extended Project Qualification (EPQ) research paper, in which I investigated the current evidence on stem cell-based therapies for Parkinson’s disease.

Parkinson’s disease can be managed, but it cannot be cured. Stem cell therapy holds something more — but how close is it to delivering?

Parkinson’s disease is best known for its visible symptoms: the tremor, the slow shuffling walk, and the mask-like stillness of the face. But the deeper story is what’s happening beneath the surface. Neurons in a region of the brain called the substantia nigra are dying slowly, progressively, and permanently. With them goes the brain’s ability to produce dopamine, the chemical that keeps movement smooth and coordinated.

The treatments we have today are good at masking that loss. Levodopa, the gold-standard drug for decades, essentially tops up dopamine levels and may be transformative, especially in the early years. But here’s the thing: it doesn’t stop the neurons from dying. It compensates for their absence. As the disease progresses and more neurons are lost, levodopa becomes increasingly difficult to manage, leading to unpredictable swings in motor control. Deep brain stimulation, another widely used option, works by electrically modulating the brain’s misfiring circuits, again, a workaround rather than a solution.

The scale of the problem

6M+ people living with Parkinson’s in 2015

12M projected by 2040 (Dorsey et al., 2018)

0 treatments that slow neurodegeneration

This is the gap that stem cell research is trying to close. Instead of masking dopamine loss, the idea is to restore it by transplanting new dopaminergic neurons into the brain to replace those that have died.

The biological logic is solid. The clinical reality is harder.

The concept itself is elegant. Take stem cells , either embryonic stem cells or induced pluripotent stem cells (iPSCs), which are adult cells reprogrammed back into a stem-like state, and coax them into becoming the exact type of neuron Parkinson’s destroys. Transplant them into the putamen, a key structure in the brain’s motor circuit. Let them take root, produce dopamine, and restore function.

Pre-clinical work backs this up. In animal models of Parkinson’s disease, transplanted dopaminergic progenitor cells have survived, matured, and measurably improved motor function. The biology works in a dish and in a mouse. The harder question is whether it works in a human brain that has been adapting to neurodegeneration for years.

“The Parkinsonian brain is not an empty system awaiting replacement, but a complex and progressively adapting network.”

The history here is instructive. Before modern stem cell techniques existed, researchers tried transplanting fetal brain tissue — actual dopamine-producing neurons harvested from aborted foetuses. Some patients improved. Others didn’t. A subset developed a disturbing complication: graft-induced dyskinesias, involuntary movements caused by the transplanted cells themselves. The results were too inconsistent to build a therapy from, as well as the ethical and administrative problems of sourcing fetal tissue at scale were unbeatable anyway.

The TransEuro trial, which attempted to revive and standardise fetal tissue grafting, told a similar story. Only 11 patients could be grafted due to tissue shortages, and after three years, the primary clinical endpoint showed no significant improvement. A proof of concept — yes. A viable treatment — no.

What the earliest human trials with stem cells actually show

Modern approaches aim to fix both problems: consistency and scale. Stem cells can, in principle, be manufactured in unlimited quantities under controlled conditions. Three early human trials published in 2025 have now put that potential to its first real test.

The bemdaneprocel trial (Tabar et al., 2025) enrolled 12 patients who received embryonic stem cell-derived dopaminergic progenitors transplanted directly into the putamen, with a year of immunosuppression to prevent rejection. The study met its primary endpoint — safety. PET imaging showed increased fluorodopa uptake, a signal that the grafted cells were surviving and producing dopamine. A phase 1/2a trial with similar cells (Chang et al., 2025) found no dose-limiting toxicities and reported improvements in motor scores at higher doses. An iPSC-based trial (Sawamoto et al., 2025) found no tumor formation, a major theoretical risk with pluripotent cells, and again showed positive imaging signals.

This is real progress. These trials show that cells can be placed into a human brain, survive, and perform a biological function. That is nothing. A few years ago, this was purely theoretical.

But these are Phase I trials. They were designed to ask: Is this safe? Not: Does this slow Parkinson’s disease? Cohorts of 7–12 patients, followed for 12–24 months, with no control group. This evidence base cannot yet answer the bigger question. Symptom improvements in open-label trials (where both the patient and the doctor know who received treatment) are notoriously hard to interpret, because the placebo effect and intensive clinical monitoring both tend to improve outcomes temporarily.

The obstacles that still stand in the way

Beyond the trial design limitations, there are deeper scientific and practical barriers:

• Parkinson’s is more than dopamine. Even a perfect dopaminergic graft just addresses one part of a disease that causes widespread neurodegeneration, non-motor symptoms, and cognitive changes. Replacing what’s lost may compensate for some of the damage while leaving the rest untouched.
• Grafted neurons can get sick, too. Post-mortem studies of patients who received fetal grafts more than a decade earlier found Lewy body pathology, the hallmark of Parkinson’s, in the transplanted neurons themselves. The disease process had spread into the graft. This suggests that cell replacement alone may not be durable.
• Surviving isn’t the same as integrating. A transplanted neuron that shows up on a PET scan still needs to make the right connections, release dopamine in a regulated way, and function as part of a circuit the brain has been reorganizing for years. That’s a much higher bar.
• Immunosuppression adds its own risks. Keeping a foreign cell graft alive requires suppressing the immune system, which increases the risk of infection and limits who can realistically receive the therapy. Autologous iPSC approaches sidestep this but are extraordinarily expensive and complex to manufacture for each patient.
• Manufacturing at scale is unsolved. Producing consistent, GMP-grade cell batches that can be shipped to hospitals around the world, used reliably, and stored safely is a major challenge. The therapy needs to work not just in specialist centres but throughout healthcare systems serving millions of patients.

So – viable or not?

The honest answer is: becoming viable, but not yet there. Stem cell therapy for Parkinson’s has cleared the lowest bars — the cells can be made, implanted, and shown to survive. Safety signals so far are reassuring. The field has moved from theory to early human data. That matters.

What it hasn’t done is demonstrate that neurodegeneration is actually slowing. Symptom improvements are promising, but they’re not the same thing. Disease modification, changing the underlying biological trajectory of Parkinson’s, requires larger controlled trials, longer follow-up, and outcome measures that can distinguish genuine slowing of degeneration from temporary compensation. None of that exists yet.

Where things stand in 2026

Early human trials have established safety and biological activity. Larger, controlled trials with multi-year follow-up are now the critical next step. The field is genuinely advancing — but the gap between “promising” and “proven” is still substantial, and closing it will take years, not months.

That’s not a reason for pessimism. It’s a reason for precision. Parkinson’s disease is one of the most important unsolved problems in medicine, and stem cell therapy is the most scientifically credible regenerative strategy currently in human trials. The question was never whether it would work in principle. The question has always been whether it can be made to work reliably, safely, and at scale — for the millions of people who need it.

The answer is getting closer. We’re just not there yet.