Intrathecal Stem Cell Therapy for Neurological Conditions

Intrathecal stem cell therapy involves injecting stem cells directly into the cerebrospinal fluid that bathes the brain and spinal cord, bypassing the blood-brain barrier that blocks most treatments from reaching the central nervous system. It is being tested in clinical trials for conditions like amyotrophic lateral sclerosis (ALS), multiple sclerosis (MS), spinal cord injury, and cerebral palsy, with results that range from cautiously promising to clearly inconclusive. The approach has a generally tolerable safety profile in regulated trials, but real risks emerge when the therapy is offered outside formal clinical settings.

Why Inject Into the Spinal Fluid

The blood-brain barrier is a tightly sealed layer of cells lining the blood vessels in your brain and spinal cord. It keeps toxins out, but it also blocks most drugs and nearly all transplanted cells from reaching diseased neural tissue. Delivering stem cells intravenously means most of them get trapped in the lungs, liver, or spleen before they ever approach the nervous system. Intrathecal delivery sidesteps this problem by placing cells directly into the cerebrospinal fluid, giving them a clear path to the surfaces of the brain and spinal cord without having to cross that barrier.¹

Once in the cerebrospinal fluid, stem cells can migrate toward injured tissue. In animal studies of spinal cord injury, fluorescently labeled stem cells injected intrathecally were found on the surface of damaged spinal cord tissue and even in deeper perivascular spaces, with significantly higher concentrations near the lesion compared to uninjured areas.² That homing behavior is part of what makes the route attractive for neurological diseases, where the damage is locked behind a barrier that most therapies cannot penetrate.

How the Cells Actually Help

One of the bigger surprises in stem cell research is that these cells often do not integrate into damaged tissue and become new neurons. Instead, their therapeutic effects appear to be mostly indirect. A study using a pig model of ALS found that intrathecally delivered mesenchymal stem cells stayed in the fluid-filled space around the spinal cord without embedding in neural tissue. The researchers concluded the cells were working through a paracrine mechanism, releasing neurotrophic factors, anti-inflammatory molecules, and tiny vesicles that help calm inflammation and support neuron survival from a distance.³

This paracrine activity has several downstream effects that researchers have documented in animal models. Stem cell spheroids derived from peripheral nerves, for example, produced neurotrophic factors that promoted nerve fiber growth, reduced cell death caused by damaging molecules, and shifted the immune response from a destructive pattern to a more protective one.⁴ In a neuropathic pain model, bone marrow stem cells delivered intrathecally secreted an anti-inflammatory protein called TSG-6 that dialed down inflammatory signaling in spinal cord immune cells and reduced the production of pain-driving inflammatory molecules.⁵

The picture is not uniformly positive, though. A separate rat study of nerve-injury pain found that intrathecal mesenchymal stem cells did not significantly reduce pain behavior or change spinal cord inflammation compared to a vehicle injection.⁶ This kind of conflicting evidence is common in preclinical stem cell work and is one reason the field has been slow to produce definitive answers.

Evidence in ALS

ALS is one of the conditions where intrathecal stem cell therapy has the most human trial data. In a controlled trial, patients who received repeated intrathecal injections of mesenchymal stem cells showed a slower decline on the standard ALS functional rating scale compared to controls, with a significant difference persisting at both four and six months after injection.⁷ A phase 1/2 trial using stem cells engineered to secrete neurotrophic factors found the treatment was safe and well tolerated, with about 87% of treated patients classified as responders based on either functional or lung capacity measures at six months.⁸

A meta-analysis pooling data from multiple ALS stem cell studies found that functional decline was significantly slower when mesenchymal stem cells were injected intrathecally, but not when they were injected directly into the spinal cord tissue. Intravenous delivery was actually associated with faster deterioration at all time points measured.⁹ That distinction matters: it suggests the delivery route itself is a key variable, not just the cell type. But these remain relatively small studies, and slowing progression is not the same as stopping or reversing it. No trial has demonstrated a cure or long-term disease modification in ALS.

Evidence in Multiple Sclerosis

MS trials have produced more mixed results. A randomized, placebo-controlled phase II trial of intrathecal mesenchymal stem cell-derived neural progenitors in progressive MS found no overall difference in the primary outcome between the treatment and placebo groups. However, an exploratory subgroup analysis of patients who already needed assistance walking showed significant improvements in walking speed and walking endurance in the treated group compared to placebo. The treatment was also associated with improved bladder function and a reduced rate of brain grey matter shrinkage on MRI.¹⁰

A separate randomized, double-blind trial (the SMART-MS study) found no detectable neuroregenerative effect from intrathecal mesenchymal stem cells in progressive MS, though the researchers noted the small sample size may have limited what they could detect. More concerning, the adverse event profile suggested the stem cell injections triggered an acute localized inflammatory reaction, leading the authors to recommend caution with this approach in future MS studies.¹¹ Measuring treatment response in MS studies with cerebrospinal fluid biomarkers is one area being actively developed. One study documented the first objective in vivo evidence of neuroprotective effects from mesenchymal stem cells in MS patients using a cerebrospinal fluid biomarker, though the authors positioned it as a proof-of-concept finding rather than clinical evidence of benefit.¹²

The tension between those two MS trials is instructive. Both were randomized and placebo-controlled, both used intrathecal mesenchymal stem cells in progressive MS patients, and they reached different conclusions. Subgroup effects, different cell preparations, and differences in patient populations can all drive such discrepancies. For someone with progressive MS hoping for a breakthrough, the honest takeaway is that this therapy remains experimental with uncertain benefit.

Spinal Cord Injury

In spinal cord injury research, intrathecal delivery faces a unique challenge: the injury site is a localized zone of disruption, and cells floating in the cerebrospinal fluid may not concentrate there effectively. A rat study comparing intrathecal versus intraspinal (directly into the cord) delivery of neural precursor cells found that both approaches improved locomotor recovery, but the intraspinal route performed better on several measures. Cells injected intraspinally survived for two months and integrated into spinal tissue, while cells delivered intrathecally were undetectable in the spinal cord at the same time point. The intrathecal approach improved some measures of tissue preservation and nerve fiber growth but was less effective at reducing scarring.¹³

Still, intrathecal delivery is less invasive than surgery to inject cells directly into the spinal cord, and some studies show encouraging results. Umbilical cord mesenchymal stem cells given intrathecally in a rat spinal cord injury model led to improved motor function, reduced scar tissue, increased neuron survival, and higher levels of a brain-derived growth factor associated with neural repair.¹⁴ The tradeoff between ease of delivery and cellular retention at the injury site is one of the central design questions in this field.

Cerebral Palsy in Children

Cerebral palsy is one area where intrathecal stem cell therapy has been tested in pediatric populations with what appears to be a reasonable safety profile. A single-arm clinical trial of repeated intrathecal injections of bone marrow-derived mesenchymal stem cells in children with spastic cerebral palsy found only mild adverse events (low-grade reactions not attributed to the therapy itself) and showed statistically significant improvements in motor function, balance, and spasticity.¹⁵

A systematic review of umbilical cord-derived stem cell therapy for cerebral palsy echoed these findings, noting that intrathecal injection appeared safe and effective across the reviewed studies, with no serious or life-threatening complications reported from cord-derived cells.¹⁶ A randomized, double-blind, sham-controlled trial of umbilical cord tissue mesenchymal stem cells also found the treatment safe and potentially beneficial for both clinical and brain imaging outcomes.¹⁷ The cerebral palsy data is encouraging enough that larger confirmatory trials are warranted, though the existing studies have been small.

Safety Profile and Known Risks

Across regulated clinical trials, intrathecal stem cell therapy has generally been well tolerated. A study tracking adverse events across repeated intrathecal injections of bone marrow-derived mesenchymal stem cells in patients with various neurological diseases found the most common side effects were mild: injection-site ache in about 4% of cases, fever in about 3.4%, and mild headache in about 2%. No severe adverse events were reported. Some patients showed temporary increases in white blood cell counts and protein levels in their cerebrospinal fluid after injection, but none developed signs of central nervous system infection.¹⁸

Rare but serious complications have been documented, however, primarily in unregulated settings. A case report described a patient who developed a unique spinal lesion with inflammatory cell infiltration and arachnoiditis (inflammation of the membranes surrounding the spinal cord) after intrathecal stem cell treatment. The authors noted that while some patients have had abnormal imaging findings without symptoms after such procedures, extreme cases have resulted in central nervous system tumors.¹⁹ The localized inflammatory reaction observed in the SMART-MS trial also raises questions about whether the immune response to injected cells could sometimes do more harm than good, even in a controlled setting.

The Problem With Unregulated Clinics

A significant portion of the stem cell therapy available worldwide falls outside the framework of clinical trials, and this is where the risk profile changes dramatically. “Stem cell clinics” operating in loosely regulated markets may offer intrathecal injections using cell products with unclear scientific rationale, unknown mechanisms of action, insufficient safety data, unconfirmed product quality, and untested administration methods.²⁰ The serious adverse events reported in the literature, including tumors and severe inflammation, have often been linked to treatments received at such facilities.

For patients or families considering intrathecal stem cell therapy, the distinction between a registered clinical trial and a commercial clinic offering “stem cell treatments” is critical. In a trial, cell preparation, dosing, patient selection, and follow-up monitoring are standardized and overseen by institutional review boards and regulatory agencies. At a commercial clinic, you may not know what cells you are receiving, how they were processed, or whether anyone will track what happens to you afterward. Researchers studying neural stem cell therapies for progressive MS have specifically cautioned about unscrupulous clinics marketing unproven therapies to desperate patients.²¹

Keeping Cells Where They Need to Be

One persistent technical challenge is that cells injected into cerebrospinal fluid tend to disperse quickly and widely, diluting their concentration at any particular injury site. Researchers have been experimenting with embedding stem cells in hydrogels before injection to address this. In a study using pigs, stem cells suspended in plain saline were only detectable within about 0.25 cm of spinal cord tissue on MRI, while cells embedded in a hydrogel matrix were detectable for up to 12 cm. The hydrogel-embedded cells also survived better and proliferated at higher rates.²² That is a striking difference in distribution, and it points toward a future where the delivery vehicle matters as much as the cells themselves.

MRI guidance during the injection procedure is another refinement being tested. Real-time imaging allows clinicians to see where cells are going as they are being delivered, rather than injecting blindly and hoping for adequate distribution. The same pig study used MRI guidance to track cell-hydrogel composites as they spread along the spinal cord, providing a level of procedural control that has not been standard in most clinical trials to date.

Cell-Free Approaches Using Exosomes

An emerging twist on intrathecal stem cell therapy is the idea that you might not need the cells at all. Since much of the therapeutic benefit appears to come from molecules the cells release rather than the cells themselves, researchers have begun testing whether the tiny vesicles (called exosomes) that stem cells shed can deliver similar benefits with fewer risks. A first-in-human trial of intrathecally injected exosomes derived from umbilical cord mesenchymal stem cells in patients with complete subacute spinal cord injury found the treatment safe and well tolerated, with no adverse events attributed to the therapy. At 12 months, patients showed significant improvements in sensory function and self-care independence scores.²³

Preclinical work supports this direction. In a rat spinal cord injury model, intrathecally delivered extracellular vesicles from stem cells reduced inflammation, decreased cell death, increased the number of healthy neurons, and improved locomotor function.²⁴ Exosomes are smaller, easier to standardize, and potentially easier to store and ship than living cells. They also avoid concerns about transplanted cells growing uncontrollably or triggering immune rejection. Whether they can match the efficacy of whole-cell therapy in larger human trials is an open question, but the concept of delivering the cells’ molecular output without the cells themselves is gaining traction.

Scaling From Animals to Humans

A recurring obstacle in this field is that promising animal results often do not translate smoothly into human benefit. Part of the reason is sheer anatomy: a mouse spinal cord is a few centimeters long, while a human one stretches roughly 45 cm. The volume of cerebrospinal fluid differs by orders of magnitude. These physical differences affect how cells distribute after injection, how concentrated the therapeutic molecules they release remain at any given point, and how quickly the cells get cleared. Researchers have noted that anatomical and physiological disparities between animal models and the human nervous system complicate the translation of findings about drugs and cells delivered into cerebrospinal fluid.²⁵

This scaling problem partly explains why, for instance, the intrathecal route shows strong benefits in rat spinal cord injury studies but faces cell retention issues that make direct injection into the cord look better. A therapy that blankets a 2 cm mouse lesion in a bath of neurotrophic factors may barely register in a human injury where the relevant tissue segment is farther from the injection site and the cerebrospinal fluid volume dilutes the signal. The hydrogel and imaging advances discussed above are direct attempts to solve this problem, but the field is still working out how to reliably deliver the right concentration of cells or their products to the right location in a human-sized nervous system.