Safety and Preliminary Efficacy of Autologous Bone Marrow-Derived Mesenchymal Stem Cell Transplantation in Hereditary Cerebellar Ataxia: Phase I/IIa Clinical Trial

Major Points and Findings:

In this small Iranian safety trial, ten people with hereditary ataxia (eight with Friedreich ataxia, one with SCA1, one with SCA2) each received a single intrathecal injection of their own bone marrow mesenchymal stem cells (BM-MSCs). No serious adverse events occurred. The average ataxia score (SARA) was lower six months later.

Few recent clinical reports deal with intrathecal cell therapy in hereditary ataxia, and this is one of them. As evidence of benefit it is weak, which the authors acknowledge. There was no control group, the planned comparison arm was dropped and follow-up was six months. The cells were the patients’ own and did not come from a donor or from umbilical cord.

Aim:

To test whether one intrathecal dose of autologous BM-MSCs is safe in hereditary cerebellar ataxia, and to explore changes in the SARA score, GAD-65 antibodies and three cytokines (IL-6, TNF-alpha, IL-10) in serum and cerebrospinal fluid (CSF).

Methods:

This was a single-centre, single-arm, open-label phase I/IIa trial at Pastoor Hospital, Mashhad University of Medical Sciences, Iran (registry IRCT20160809029275N3).

The registered protocol included a comparison group and further outcome measures. The authors report that funding limits and international sanctions prevented this, and they describe the change as a protocol deviation that raises the risk of bias.

Eligible patients were aged 10 to 65, of either sex, with a definite diagnosis of hereditary cerebellar ataxia by brain and cervical spine MRI and/or genetic testing, and atrophy of the cerebellum, brainstem or cervical cord. Exclusions:

• Pregnancy
• Heart, kidney or liver failure, cardiac arrhythmia, diabetes, leukaemia, epilepsy, lung disease, thyroid disease
• Other central nervous system disease such as Parkinson’s, or structural brain disorders
• Total bilirubin above 1.5 times the upper limit of normal
• Alcohol misuse, any sign of infection
• Severe psychosis or cognitive impairment preventing consent
• Acute (non-hereditary) causes of ataxia

Fourteen people were screened, 11 enrolled, and one withdrew before treatment because of dissatisfaction with the study logistics. Ten were treated and all ten completed follow-up. Seven were male. Table 1 lists ages of 18 to 43 (mean 32), disease duration of 2 to 13 years, and baseline SARA scores from 2 to 22. Patients kept their usual care, with no added rehabilitation.

For cell preparation, about 50 mL of bone marrow was taken from the posterior iliac crest under local anaesthetic, and patients were discharged the next day. Mononuclear cells were grown in medium with 10% fetal bovine serum to passage 3, then checked for CD73 and CD90 (positive), CD34 and CD45 (negative), and for the ability to form fat and bone cells. Batches were tested for sterility and endotoxin. Viability before injection was 95% to 98%. Cells were washed three times in saline and shipped at 2 to 8 °C. The authors flag the use of bovine serum as a limitation and say xeno-free culture is preferable.

Patients received a single dose of 1 × 10⁶ cells per kg body weight in 2 mL of saline, given by standard lumbar puncture at L3-L4 over about two minutes. A clinician observed them for 24 to 48 hours.

Adverse events, CRP, liver enzymes, SARA, and ELISA measurements in serum and CSF were recorded one month before treatment and then at 1, 3 and 6 months. No correction was made for multiple statistical comparisons.

Results:

Safety:

• No severe adverse events in any of the 10 patients.
• 3 of 10 had a mild headache after the injection, which settled within 24 hours on 500 mg of paracetamol (acetaminophen).
• 1 of 10 had a low-grade fever, which also settled with paracetamol.
• CRP, ALT and AST did not change significantly (CRP 2.86 before and 2.23 mg/L after, p = 0.84).

SARA: The group mean fell from 14.9 ± 2.1 one month before treatment to 10.36 ± 1.94 at six months (p = 0.002), a drop of about 4.5 points. The authors note that a change of 3 to 4 points is often treated as clinically meaningful. They caution that statistical significance does not amount to functional improvement, that one patient started at a SARA of 2 with almost no room to improve, and that SARA can move with expectation, practice and day-to-day variability. Results for Friedreich ataxia and the two SCA patients were not analysed separately. The ten baseline scores printed in Table 1 average 13.5. The text gives 14.9. The paper does not explain the difference.

GAD-65 antibodies: Nine patients had normal levels throughout. One patient with Friedreich ataxia had very high levels at the start (serum 814 IU/mL, CSF 781 IU/mL). These were 748 and 678 at one month, then fell into the normal range at three and six months (serum 1.7, CSF 1.3 to 1.8). The authors call this a single-patient observation that cannot show cause and effect.

Cytokines:

• Serum IL-6 fell from 4.51 to 3.25 at month 3 (p = 0.03), then returned to 4.57 by month 6.
• CSF IL-6 did not fall and was slightly higher at month 6 (4.84 against 4.15, not significant).
• CSF TNF-alpha fell from 6.35 to 3.23 at month 1 and 3.34 at month 3 (both p = 0.002). Serum TNF-alpha did not change significantly.
• Serum IL-10 (anti-inflammatory) rose from 2.87 to 6.51, 6.81 and 4.63 at months 1, 3 and 6. CSF IL-10 did not change.

Conclusions:

The authors conclude that one intrathecal dose of autologous BM-MSCs was safe and well tolerated over six months in this small group. On benefit they write that “no definitive conclusions regarding efficacy can be established from the present study.” The limitations they list are a very small sample, a mix of genetically different diseases, no control group, short follow-up, no correction for multiple comparisons, and no MRI or other objective endpoint.

The dates printed in the paper are inconsistent. Data collection is given as September 2021 to June 2022, while the ethics approval is dated 2 July 2022 and the registry entry 31 August 2022. The paper does not comment on this. Several authors declare affiliations with a private personalised-medicine centre in Düsseldorf and state that it gave no funding or commercial support.

Background Information:

The authors do not expect injected cells to become new Purkinje cells. Their hypothesis rests on paracrine effects: MSCs release growth factors such as BDNF, GDNF and VEGF and calm activated microglia, which may protect surviving neurons from stresses that hereditary ataxias share (mitochondrial failure, oxidative stress, secondary inflammation).

They also raise a point about cell source. A patient’s own MSCs carry the same genetic defect, and MSC quality declines with donor age. For future studies they suggest that allogeneic cells from young healthy donors, such as Wharton’s jelly umbilical cord MSCs, may offer more consistent potency. This trial did not test that.