Clinical feasibility of umbilical cord tissue-derived mesenchymal stem cells in the treatment of multiple sclerosis

Multiple sclerosis (MS) is an immune-mediated inflammatory disease in which the immune system progressively destroys its own myelinated axons in the central nervous system, in episodes lasting from a few months to many years in duration. The eventual demyelination and axonal degeneration can cause serious and debilitating motor, sensory, balance and cognitive problems, disability, serious complications, and negatively impact quality of life [1–3].

While there is no known cure for MS, up to 82% of costs incurred by MS patients are spent on drugs [4]. Treatments available include steroids for temporary flare-ups, disease-modifying drugs, and drugs targeting specific symptoms. While these may reduce the frequency of exacerbations and slow disease progression, none have myelin or nerve regenerative capability to restore the cumulative damage already in place [5].

Mesenchymal stem cells (MSC) derived from bone marrow, adipose, or other sources can exert inhibitory effects on immune-mediated disease states [6–10]. In particular, MSC derived from umbilical cord (UC) Wharton’s Jelly possess a high proliferative and expansive ability, an enhanced therapeutic activity compared to other MSC [11–14], and superior production of growth factors that stimulate secretions responsible for therapeutic potential [15].

The safety of MSC therapy for MS has been demonstrated in several trials [16–20]. We reported three subjects treated with MSC and stromal vascular fraction with no adverse effects; all showed clinical improvements in cognitive and motor function and presented no new lesions on magnetic resonance imaging (MRI) [21]. More recently, trials with placenta-derived MSC [22], or with intravenous UCMSC [23] reported few mild or moderate adverse effects, as well as some improvement in Expanded Disability Status Scale (EDSS) scores.

In this study, we sought to determine the safety and the efficacy of allogeneic UCMSC treatment in subjects with MS.

This open-label, single-arm, single-center phase 1/2 study was designed to assess the safety and efficacy of the intravenous administration of UCMSC for the treatment of MS. The study was approved by the Panamanian Institutional Review Board (Comité Nacional de Ética de la Investigación) and registered in the ClinicalTrials.gov database (NCT02034188). The study sponsor was Translational Biosciences. All treatments were administered at the Stem Cell Institute in the Republic of Panama, under Protocol Number TBS-UCMSC-001. Safety was defined as absence of treatment-associated adverse events at 1, 3 months, and 1 year post treatment. Efficacy was assessed with traditional MS evaluation instruments and a quality of life questionnaire at follow-up intervals, as detailed in the “Treatment protocol” section.

Subjects were enrolled under the following criteria: men or non-pregnant women ages 18–55 diagnosed according to revised McDonald criteria [24] for clinically-defined MS; an EDSS score of 2.0–7.0 assessed at least 3 months after the last acute attack of MS; willingness to keep a weekly diary and undergo observation for 1 year, and provision of documented health insurance in their home country. Enrolled subjects were not required to refrain from taking other medications or supplements prior to study entry.

Subjects were excluded if they presented active proliferative retinopathy, poorly controlled diabetes mellitus (glycosylated hemoglobin HbA1C > 8.5%), renal insufficiency (Creatinine > 2.5 mg/dL) or failure, infection (white blood cell count of > 15,000 K/cumm and/or temperature > 38 °C), history of organ transplant, previous or active malignancy, or cardiovascular conditions. All participants provided written informed consent before study participation.

All subjects received all of the infusions specified by the treatment protocol, and attended the 1- and 3-month visits. Nineteen subjects were followed for 1 year: the 1-year visit was completed for 17/20 subjects, two subjects partially completed the 1-year requirements. One subject was lost to follow-up.

EDSS scores were recorded at baseline and 1 month post-treatment for all subjects; scores at 1 year were available for 17 of 20 subjects (Fig. 1). At baseline, EDSS scores ranged from 2.5–7.0, with a mean score of 5.23 (SD = 1.50). At 1 month, the mean score decreased to 4.75 (SD = 2.00), a mean reduction of 0.48 (SD = 0.85) or about one category. Scores reduced further at the 1-year time point, to 4.62 (SD = 2.72), a mean reduction of 0.68 (SD = 1.49), a little more than one category. Differences were statistically significant at 1 month (p < 0.03), and 1 year (p < 0.04) when compared to baseline.

SNRS scores were recorded at baseline and 1 month post-treatment for all subjects; scores at 1 year were recorded for 17 of 20 subjects (Table 3). The SNRS total mean score rose slightly from 75.0 (SD = 11.9) at baseline to 75.5 (SD = 16.5) at 1 month post-treatment, and decreased to 73.5 (SD = 19.5, N = 17) 1 year post-treatment. A statistically significant improvement (p < 0.01) was seen for the bladder/bowel/sex dysfunction category at 1 month from the baseline assessment (Fig. 2). A statistically significant worsening was seen in the study group in four categories compared to baseline: visual acuity at 1 month (p < 0.02) and 1 year (p < 0.03), right upper extremity motor function at 1 year (p < 0.02), right lower extremity sensory function at 1 month (p < 0.03) and 1 year (p < 0.01), and left lower extremity sensory function at 1 month (p < 0.01).

SNRS component	Baseline	1 month	1 year
Mentation and mood (normal = 10)	8.4 (2.28)	9.4 (1.57)	9.3 (1.69)
Visual acuity (normal = 5)*	4.9 (0.45)	4.2* (1.35)	4.2* (1.24)
Visual fields (normal = 6)	5.7 (0.73)	5.6 (1.05)	5.6 (0.79)
Eye movements (normal = 5)	4.8 (0.62)	4.8 (0.62)	4.8 (0.66)
Nystagmus (normal = 5)	5.0 (0.00)	4.8 (0.62)	4.9 (0.49)
Lower cranial nerves (normal = 5)	5.0 (0.00)	4.9 (0.45)	5.0 (0.0)
Motor function R upper extremity (normal = 5)*	4.8 (0.62)	4.5 (0.89)	4.1* (1.44)
Motor function L upper extremity (normal = 5)	4.3 (0.98)	4.4 (1.14)	4.1 (1.44)
Motor function R lower extremity (normal = 5)	3.2 (1.66)	2.6 (1.82)	2.4 (1.91)
Motor function L lower extremity (normal = 5)	2.7 (1.81)	2.7 (2.03)	2.7 (2.23)
DTR upper extremity (normal = 4)	3.6 (0.95)	3.3 (1.22)	3.6 (0.79)
DTR lower extremity (normal = 4)	2.4 (1.35)	2.1 (1.62)	2.3 (1.61)
Babinski sign L side (absent = 2)	0.6 (0.94)	0.9 (1.02)	0.7 (0.99)
Babinski sign R side (absent = 2)	0.8 (1.01)	0.8 (1.01)	0.5 (0.87)
Sensory R upper extremity (normal = 3)	3.0 (0.00)	2.7 (0.47)	2.8 (0.56)
Sensory L upper extremity (normal = 3)	3.0 (0.00)	2.6 (0.50)	2.8 (0.56)
Sensory R lower extremity (normal = 3)*	2.9 (0.31)	2.5* (0.51)	2.5* (0.72)
Sensory L lower extremity (normal = 3)	2.9 (0.31)	2.5* (0.51)	2.5 (0.80)
Cerebellar signs upper extremity (normal = 5)	4.0 (1.52)	4.0 (1.52)	4.1 (1.60)
Cerebellar signs lower extremity (normal = 5)	2.9 (1.46)	3.6 (1.87)	3.4 (1.77)
Gait trunk and balance (normal = 10)	5.1 (2.74)	5.6 (3.17)	4.6 (3.69)
Bladder, bowel, and/or sexual dysfunction (normal = 0)*	− 4.7 (2.52)	− 2.7* (2.89)	− 3.1 (3.31)
Total SNRS	75.0 (11.90)	75.5 (16.54)	73.5 (19.52)

*Statistically significant change, p < 0.05

The 9HPT was recorded for all subjects at baseline and 1 month post-treatment; scores were available for 19 of 20 subjects at 1 year (one lost to follow-up). Overall, subjects saw improvements in their scores. Statistically significant improvements from baseline were seen in non-dominant hand scores for both best (p < 0.01) and average (p < 0.02) times at the 1-month assessment (Fig. 3). The best score for the non-dominant hand was 37.56 s at baseline, 32.49 at 1 month (5.1 s improvement), and 31.55 at 1 year (6.0 s improvement). The average mean score for the non-dominant hand was 40.80 s at baseline, 35.01 at 1 month (a 5.8 s improvement), and 34.09 (a 6.7 s improvement).

For the 25FWT, subjects were categorized into one of four categories, based on their need for assistance to complete the test (in order of increasing need): no assistance; cane; walker, and wheelchair. While scoring guidance for the 25FWT directs to use the average of the two completed trials, the minimum time was also analyzed to enable inclusion of any/all subject(s) who completed at least one full trial. At baseline, all 20 subjects were available for the 25FWT, and 19 subjects were available at 1 month (one did not take the test). At 1 year, results were recorded for 17 subjects (two did not take the test; and one was lost to follow up). Some subjects were unable to complete one or both trials due to ambulation status (two were wheelchair-bound at baseline, one at the 1-month and at 1-year follow-ups), or fatigue (one subject was fatigued at the 1-year follow-up).

At baseline, average time for the 25FWT Trial 1 was 17.43 s and 17.51 s for Trial 2 (Table 4). Overall mean time was 17.47 s, and average minimum time was 15.85 s. One month after treatment, average walk time for Trial 2 improved by 6.01 s, but average time for Trial 1 took longer than baseline, as did overall mean time, and average minimum time. At 1 year, all walk times were reduced when compared to baseline. Since not all subjects completed both trials of the 25FWT at a given time point, and not all subjects completed all follow-ups, statistical comparisons between baseline and follow-up scores could not include all subjects. Twelve subjects completed Trial 2 of the 25FWT at the three follow-up time points (baseline, 1 month and 1 year), with a statistically significant reduction (p < 0.02) in walk times found between baseline and 1 month (Fig. 4).

	Trial 1—average time	Trial 2—average time	Overall mean time	Average minimum time
Baseline	17.43 (N = 18)	17.51 (N = 18)	17.47 (N = 18)	15.85 (N = 18)
1 month	29.19 (N = 18)	11.50 (N = 13)	28.63 (N = 18)	28.01 (N = 18)
1 year	16.18 (N = 15)	13.29 (N = 13)	15.70 (N = 15)	14.71 (N = 15)

Some subjects changed ambulatory status at the time of the 25FWT at the 1-month visit: two subjects reduced their dependence upon assistive devices (one from wheelchair to walker, one from walker to cane); one became more dependent (from no assistance to a cane). At the 1-year visit, two subjects reduced their dependence upon assistance devices from baseline to 1-year visit (one from wheelchair to walker, one from walker to cane to no assistance); two increased dependency (one from unilateral cane to walker, one from no assistance to cane). One of the subjects, wheelchair-bound at baseline and thus unable to complete the test then, was able to complete it at the 1-month and 1-year follow-ups.

Scores for the RAND SF-36 test were recorded for all subjects at baseline and 1 month, and for 17 of 20 subjects at 1 year post-treatment. Overall, subjects reported improvements in their health during the study. Total QOL scores increased over baseline for 15 subjects at the 1 month follow up visit, while only five decreased. At 1 year, total QOL score improved over baseline for 11 subjects, while six decreased (Fig. 5). At 1 month and 1 year, more than half of the subjects reported their condition as better or the same on all scale scores (Table 5). At 1 month, more than 50% of subjects reported improvements on five of the eight scale scores (role limitations-physical, energy/fatigue, emotional well-being, social functioning, and general health), as well as health change; 45% (9/20) of subjects reported improvements in the pain scale, and 35% (7/20) of subjects reported improvements in their physical functioning and role limitations-emotional. At 1 year, more than half of subjects reported feeling better in five of the eight scales (physical functioning, role limitations-physical, energy/fatigue, emotional wellbeing, and general health), as well as health change; 47% (8/17) reported improvements in their social functioning, 35% (6/17) in pain scale, and 29% (5/17) in role limitations-emotional.

	Baseline to 1 month (N = 20)			Baseline to 1 year (N = 17)
	Better	Same	Worse	Better	Same	Worse
Physical functioning	7 (35%)	4 (20%)	9 (45%)	10 (59%)	1 (6%)	6 (35%)
Role limitations—physicala	13 (65%)	6 (30%)	1 (5%)	11 (65%)	4 (24%)	2 (12%)
Role limitations—emotional	7 (35%)	10 (50%)	3 (15%)	5 (29%)	9 (53%)	3 (18%)
Energy/fatigueb	16 (80%)	1 (5%)	3 (15%)	11 (65%)	0 (0%)	6 (35%)
Emotional well-being	13 (65%)	2 (10%)	5 (25%)	10 (59%)	2 (12%)	5 (29%)
Social functioning	12 (60%)	2 (10%)	6 (30%)	8 (47%)	1 (6%)	8 (47%)
Pain	9 (45%)	5 (25%)	6 (30%)	6 (35%)	8 (47%)	3 (18%)
General health	11 (55%)	1 (5%)	8 (40%)	9 (53%)	2 (12%)	6 (35%)
Health changea	13 (65%)	6 (30%)	1 (5%)	11 (65%)	4 (24%)	2 (12%)

^aDifferences between baseline and 1-month and 1-year scores statistically significant (p < 0.03)

^bDifferences between baseline and 1-month scores statistically significant (p < 0.001)

Statistically significant changes from baseline occurred at both 1-month and 1-year assessments for the RAND SF-36 role limitations-physical (p < 0.002 and p < 0.03, respectively) and health change (p < 0.004 and p < 0.02, respectively) categories. A statistically significant change occurred from baseline at 1 month in the energy category (p < 0.006) and the average score (p < 0.001). Bonferroni corrections retained statistically significant differences at 1 month in these four categories (p < 0.02). Most categories in the RAND SF-36 showed improvement over the baseline at 1 month and 1-year evaluations, with the exception of the 1-month physical function score. Most categories scored highest at the 1-month evaluation with the exception of the physical function category (Fig. 6).

Pre- and post-treatment MRIs from the brain and the cervical spinal cord were reviewed qualitatively for every subject. MRI was unavailable at the 1-year follow-up for two subjects. Not all subjects had the second MRI taken at exactly 1 year after treatment; there was some variation in completion of the second MRI, up to 4 months after the 1-year mark. No common themes were detected regarding number of lesions or disappearance of lesions. Of the 18 subjects that did complete MRI both at baseline and at the 1-year follow-up, 15 (83.3%) showed no disease progression or no new or active lesions. Two subjects (11.1%) showed progression in their lesions. One patient (5.6%) showed near complete resolution of the plaques in the brain, in previously noted areas of abnormal signal intensity consistent with demyelinating disease (Fig. 7).

The demographics of this study correspond to those of an expected MS population, where women are typically affected twice as frequently as men, and most patients are diagnosed between 20 and 40 years of age [26–29].

No Serious Adverse Events occurred in the study. One single Adverse Event (AE), headache, was noted as probably related to treatment. Of the AEs denoted as possibly related to treatment, most were headache or fatigue. Headache is a known complaint noted to occur during or just after MSC infusions, most of which quickly resolve [30]. Additionally, Foley et al demonstrated that patients with MS commonly experience headache [31]. Thus, headache is not unexpected at the time of or just after UCMSC infusions. Similarly, fatigue is one of the most common complaints [32]. Fatigue appears in all ages and phenotypes of MS [33–35] and is a primary determinant of poor QOL [35], affecting both physical and mental components independent of disability level [34]. Thus, although fatigue is a possibly related AE in this study, it is also a very common disability symptom of MS.

Previous studies using MSC treatment for MS have reported improvements in EDSS scores [19, 22, 23, 38, 39]. In our case, the statistically significant (p < 0.03) change in EDSS mean scores from baseline to 1 month reflects a change in disability category, which could translate into an improved ability to walk and work a full day with minimal, if any, assistance. Although other categories showed worsening typical of disease progression, the bladder/bowel/sexual dysfunction category of the SNRS showed statistically significant improvement at 1 month (p < 0.05). This finding may be encouraging in that up to 60% of patients with MS report sexual dysfunction problems [40] while over 50% experience bowel dysfunction and up to 75% will report bladder dysfunction [41].

Changes to ambulatory status from baseline classifications were noteworthy in this study for the 25FWT. Increasing disability (e.g., going from unassisted walking to a cane) is a typical disease progression in the MS population in general, but the reverse would not be expected. However, subjects in our study improved from wheelchair status to using a walker, and from walker status to requiring no assistance.

In general, QOL for patients with MS is diminished by physical, emotional, and cognitive symptoms and comorbidities [33, 42]. However, subjects enrolled in this study reported consistent improvements in their RAND 36 SF QOL in these areas, particularly at the 1-month evaluation (p < 0.001). Most categories improved at 1 month, and then slipped slightly at the 1 year evaluation period while still remaining improved over baseline, suggesting that treatment frequency greater than once annually could further improve the treatment subject’s QOL outlook.

Most subjects (83.3%) showed no disease progression or new lesions in their MRIs. The near complete resolution of the plaques of the brain in one patient (Fig. 7) is a particularly encouraging finding that should be further investigated by comparing it to similar MS cohorts in a standardized time period.

The small sample size is the most significant limitation of this study in that it may impact the statistical significance of the results. When subjects were lost to follow-up, the sample size was reduced; due to the nature of certain tests used to measure efficacy signals, failing to complete one part of the test often invalidated obtaining an average for the score. While this issue could have been avoided with larger recruitment numbers, we are still in the lower threshold of the 20–80 subjects recommended by FDA for early trials. Additionally, subjects with certain forms of MS may be less likely to develop new lesions, which may impact the findings of the reported MRI results. As this was primarily a safety and proof of concept trial, we did not require subjects to stop their usual medications, which could have a confounding impact on our findings. Poor medication adherence is frequently seen among patients with MS (usually because of cost, perceived lack of efficacy, or adverse effects) [43]; in our sample, 25% were not taking any MS-specific medication at the start of treatment. However, it is noteworthy that a further 20% felt well enough to reduce their intake. In any case, efficacy of UCMSC therapy for MS should be confirmed with larger, controlled, randomized trials.

We have shown that the intravenous infusion of UCMSC over several days is safe in subjects with MS. Additionally, UCMSC infusions may hold benefits, since this small study group saw improvement in bladder, bowel, and sexual dysfunction, walking, upper extremity physical function, energy and fatigue, general perspective of a positive health change and improved quality of life, and MRI lesions. More clinical studies, particularly with a larger cohort, are needed to substantiate the specific benefits of UCMSC infusion as a potential MS therapy.