Clinical trial questions about Muse cells are usually answered in one of two unhelpful ways: a list of conditions with no trial detail attached, or a claim that trials are “underway” without saying which, where, or what they found.
What follows is the published record of five trials, each with phase, registration, patient count, dose, and reported outcome. Where a claimed result could not be traced to a published paper, that is stated. Where no trial exists, that is stated too.
Five trials, fewer than seventy patients
Muse cell clinical trials to date are small, early-phase, and concentrated in a single country. Every published trial used the same investigational product, CL2020, international non-proprietary name nafimestrocel, developed by a Japanese pharmaceutical group. Combined enrolment across all published studies is fewer than seventy patients. Safety and tolerability were the primary endpoints in every case, and efficacy signals, where reported, come from secondary endpoints in trials not powered to establish them. [1][2][5][6][7]
That is the honest summary of the status of Muse cells clinical trials as of 2026: a consistent safety picture across several indications, one randomised placebo-controlled trial with a secondary-endpoint efficacy signal[1][2][5][6] , and nothing that has completed a confirmatory study.
One clarification before the details: CL2020 is a separate investigational product developed by another company. Nothing on this page describes CellGenic material, and no trial listed here was conducted with it.
What has been published, trial by trial
Subacute ischaemic stroke: Phase 2, randomised, placebo-controlled
This is the strongest design in the set. Niizuma and colleagues, in the Journal of Cerebral Blood Flow & Metabolism in 2023, ran a randomised, double-blind, placebo-controlled Phase 2 trial in moderate-to-severe subacute ischaemic stroke, registered as JapicCTI-184103. Thirty-five patients were randomised, twenty-five to CL2020 and ten to placebo, each receiving a single intravenous dose of 1.5 × 10⁷ cells 14 to 28 days after onset.[5]
The primary endpoint was safety to week 12, with no major safety issues identified. On the key secondary efficacy endpoint, modified Rankin Scale of 2 or below at week 12, the response rate was 40.0% on CL2020 against 10.0% on placebo, the lower bound of the 95% confidence interval (21.1%) exceeding the pre-specified threshold of 8.7%.
That result is the most cited number in the Muse field, so its status is worth stating precisely: it met a pre-specified criterion for potential effectiveness on a secondary endpoint, in a thirty-five-patient trial whose primary endpoint was safety. It justifies a confirmatory trial.[5] It is not one.
Amyotrophic lateral sclerosis: Phase 2
Yamashita and colleagues published a Phase 2 trial in Cell Transplantation in 2023, registered as jRCT2063200047, in five patients with sporadic limb-onset ALS. Each received 15 × 10⁶ cells intravenously, monthly, for six doses, with safety and tolerability assessed to twelve months.
The product was well tolerated. Twenty-eight adverse events were recorded, most commonly headache in four cases and fatigue in three; the single serious adverse event, a bone fracture, was judged unrelated to administration. ALSFRS-R scores showed a non-significant trend toward improvement at twelve months, with three of five patients showing a reduced rate of decline. The authors name the limitations themselves: five patients, no comparator, and the need for a larger double-blind study.
Cervical traumatic spinal cord injury: Phase 1/2a
Among Muse cells clinical trials, spinal cord injury has the most recent published data. Koda and colleagues reported a Phase 1/2a prospective, multicentre, non-randomised, single-arm study in Stem Cell Research & Therapy in 2024, registered as JRCT1080224764, in ten patients with cervical injury at C4–C7 and modified Frankel B1 or B2, each given a single intravenous dose of 15 × 10⁶ cells. [1]
Two severe adverse events occurred, both attributed to underlying conditions rather than administration, and the authors reported no safety concerns and good tolerability. Statistically significant improvements were recorded in motor scores, sensory scores, and activities of daily living, in a single-arm study of ten patients, a design that cannot separate an intervention effect from natural recovery. [1]
Neonatal hypoxic-ischaemic encephalopathy: Phase 1
The SHIELD trial, published by Sato and colleagues in Stem Cells Translational Medicine in 2024, was a Phase 1 dose-escalation study in nine neonates with moderate-to-severe hypoxic-ischaemic encephalopathy already on therapeutic hypothermia. Three received 1.5 × 10⁶ cells; six received 1.5 × 10⁷. [6]
The primary endpoint was adverse events at twelve weeks. No serious adverse events related to administration occurred; a single mild gamma-glutamyltransferase elevation resolved without intervention, and all patients survived with no physiological abnormalities during or after administration. Gross motor function was assessed at 78 weeks using GMFCS-E&R. [6]
Indications reported but not traceable to a published paper
Acute myocardial infarction and dystrophic epidermolysis bullosa both appear in review coverage as indications in which CL2020 was studied, and epidermolysis bullosa results have been reported in pharmaceutical trade press. [1]
No peer-reviewed publication for those results could be located, so no figures from them appear here. That is deliberate: an efficacy number that cannot be traced to its original paper should not be repeated to physicians.
Three fields to check in a registry entry
A physician who wants to verify any of the above can do it in under a minute, and the reason to bother is that registry and literature searches for Muse cells return two different kinds of human study.
Three fields settle it.
Study type. The decisive one. ClinicalTrials.gov defines an observational study as one where “participants are not assigned to specific interventions by the investigator (as in a clinical trial).” An entry marked observational is not testing whether administering anything does anything. [8]
Sponsor. Tells you whether a commercial developer is behind the entry or an academic centre is running an investigator-initiated study. Different entries, different weight, different likelihood of a confirmatory programme following.
Enrollment. The number, and whether it is recorded as actual or estimated. An estimated enrolment on an entry that has not been updated in years is not evidence of a running trial.
Applied to this subject: the four trials above are interventional, each with a registration number JapicCTI-184103, jRCT2063200047 and JRCT1080224764 among them. Published human Muse research also includes observational work that administers nothing. Toyoda and colleagues, in Clinical and Translational Science in 2024, counted endogenous Muse cells in biopsied myocardial tissue from seventeen patients with fulminant myocarditis against six controls, correlating the counts with cardiac biomarkers and recovery. [7]
That is a study of Muse cells in patients. It is not a trial of a Muse cell product, and reading it as one is the category error this section exists to prevent. [7][8]
No trial has cerebral palsy as its indication
This needs stating directly, because it is the most searched condition pairing in this subject and the answer is not the one the search volume implies.
There is no published clinical trial of Muse cells with cerebral palsy as its indication.
What exists is adjacent and worth understanding. The SHIELD trial above enrolled neonates with hypoxic-ischaemic encephalopathy a leading antecedent of cerebral palsy and assessed gross motor function at 78 weeks using GMFCS-E&R, the classification system used to grade motor impairment in cerebral palsy. That is a Phase 1 safety study of nine neonates in which a cerebral-palsy-relevant motor outcome was measured. It is not a trial of Muse cells for cerebral palsy, and the distinction is not a technicality. [6]
Searches pairing Mari Dezawa’s name with cerebral palsy are, as far as the published record shows, reaching for that neonatal work. Dezawa’s group at Tohoku University published the original Muse cell science and much of the subsequent research; the group is not the developer of the investigational product used in these trials, and “MuseCells®” as a product name is a trademark held by another supplier, unconnected to CellGenic. Any page that presents the neonatal HIE data as cerebral palsy evidence is overstating the record.
Autism: no registered trial exists
Shorter still. Searches for Muse cells and autism return no registered clinical trial in any registry located for this article, and no published study.
The term appears in Dan’s keyword set because people search it, not because there is research behind it. Writing a section that implies otherwise would be the single clearest way to turn this page into the kind of content it exists to displace.
Every trial so far has run in Japan
Every published Muse cell trial has run in Japan, and there are three reasons, none of them accidental. [2][5][6]
The science originated there, at Tohoku University. The developer of CL2020 is a Japanese pharmaceutical group. And Japan operates a dedicated regulatory pathway for regenerative medicine products, distinct from its conventional drug pathway, which has made it the practical venue for early cell-product studies. [1][5][6]
On that last point, the developer stated in December 2021 that it had elected to pursue full approval for the cerebral infarction indication rather than the conditional, time-limited route, a decision requiring a larger validation study than the exploratory trial, with the national regulator aligned on that strategy.
That programme did not continue. On 14 February 2023 the same group announced that it had discontinued development of CL2020, citing the latest clinical developments, timelines for commercialisation and its pharmaceutical business strategy. No safety concern was given as a reason. The product was never approved in any market.
That is the single most important thing to understand about the trial literature above. Every published Muse cell trial used a product whose development has since been discontinued. The findings stand as published findings but no active late-stage programme sits behind them, and the confirmatory trial the stroke signal called for has not been run.
What “regeneration” means in this literature
Searches for Muse cells clinical trials and regeneration, or Muse cells clinical trials in regenerative medicine, ask a mechanistic question the trials above do not answer. They report clinical endpoints Rankin scores, ALSFRS-R, motor and sensory scores, adverse events not tissue regeneration measured directly. [1][2][5][6]
The regeneration account comes from preclinical work on homing and differentiation, a separate body of evidence with its own limits: the homing mechanism. Reading a clinical endpoint as proof of a regenerative mechanism is the commonest error in this literature.
The position as of 2026
Five published trials, fewer than seventy patients, one randomised placebo-controlled design, a consistent tolerability picture, one secondary-endpoint efficacy signal never confirmed, no trial in cerebral palsy, no trial in autism and the product all five used discontinued by its developer in February 2023.
Nothing in this page establishes that Muse cells treat, cure, prevent or diagnose any condition, and nothing in it supports administration outside a properly authorised research setting under applicable regulation.[1][2][5][6]
For the safety literature underneath the trials, see what the published safety literature reports; for what the cells are, what Muse cells are and how they were identified; and for material characterisation, where a population was sourced and how it was isolated.
Certificates of analysis are issued per batch and released to verified accounts. Access begins with provider verification.
Common questions
Are there Muse cell clinical trials for cerebral palsy?
No. No published clinical trial has cerebral palsy as its indication. The nearest work is the SHIELD trial, a nine-patient Phase 1 safety study in neonatal hypoxic-ischaemic encephalopathy, a leading antecedent of cerebral palsy, which assessed gross motor function at 78 weeks. Presenting that as cerebral palsy evidence overstates the record.
Is there a Muse cell trial in autism?
No registered trial and no published study were located. The term appears in search data because people search for it, not because research exists behind it.
Has any Muse cell product been approved?
No. CL2020, the investigational product used in every published trial, was never approved in any market, and its development has been discontinued. The developer stated in December 2021 that it would pursue full approval rather than a conditional route, then announced on 14 February 2023 that it had discontinued development of CL2020, citing clinical developments, commercialisation timelines, and business strategy, with no safety concern given as a reason.
Why are all the Muse cell clinical trials in Japan?
Three reasons: the science originated at Tohoku University, the developer is a Japanese pharmaceutical group, and Japan operates a dedicated regulatory pathway for regenerative medicine products separate from its conventional drug pathway.
References
Sources cited or paraphrased in this article are listed alphabetically by first author.
- Multilineage-differentiating stress-enduring cells: a powerful tool for tissue damage repair. Frontiers in Cell and Developmental Biology. 2024;12:1380785. PubMed / PMC →frontiersin
- Safety and feasibility of intravenous administration of a single dose of allogenic-Muse cells to treat human cervical traumatic spinal cord injury: a clinical trial. Stem Cell Research & Therapy. 2024;15(1):259. PubMed →
- Development of a regenerative medicine product (CL2020) using Muse cells. Mitsubishi Chemical Group Corporate Announcement. 15 December 2021. Official Release →mcgc
- Discontinuation of the development of a regenerative medicine product (CL2020) using Muse cells. Mitsubishi Chemical Group Corporate Announcement. 14 February 2023. Official Release →mcgc
- Randomized placebo-controlled trial of CL2020, an allogenic Muse cell-based product, in subacute ischemic stroke. Journal of Cerebral Blood Flow & Metabolism. 2024;44(4):589–601. PubMed → [,]
- Safety and tolerability of a Muse cell-based product in neonatal hypoxic-ischemic encephalopathy with therapeutic hypothermia (SHIELD trial). Stem Cells Translational Medicine. 2024;13(11):1053–1066. PubMed → [, ,]
- Accumulation of endogenous Muse cells in the myocardium and its pathophysiological role in patients with fulminant myocarditis. Clinical and Translational Science. 2024;17(11):e70067. PubMed → [,]
- ClinicalTrials.gov study-type definitions (support knowledgebase). U.S. National Library of Medicine. ClinicalTrials.gov →
REGULATORY DISCLAIMER · CELLS
These products are intended for laboratory research use only. They are not drugs, foods, cosmetics, or medical treatments and must not be used for any form of human or animal administration. All information provided is for educational and scientific reference only and the products should be handled exclusively by licensed, qualified professionals. Misbranding, misuse, or mislabeling of these products as therapeutic or consumable substances is strictly prohibited by law.
