Neural stem cell-like cells derived from autologous bone mesenchymal stem cells for the treatment of patients with cerebral palsy

Study design and patient enrolment

An open-label, non-randomised, observer-blinded controlled clinical study was conducted. The study was approved by the Ethics Committee and the Science Committee of the Armed Police General Hospital of Zhejiang Province, China and was registered in the China Clinical Trial Registry (ChiCTR-TRC-12002056). Informed consent was obtained from all of the patients or their parents. A total of 60 patients with cerebral palsy with Gross Motor Function Classification System (GMFCS, corresponding to GMFCS E&R 2007) levels III-V were enrolled in this study from June 2010 to June 2011. Overall, 30 patients were treated with conventional methods and stem cell transplantation and were included in the transplantation group. An additional 30 patients with cerebral palsy were only treated with conventional methods and were included in the control group. The study protocol is shown in Figure 1.

The patients were enrolled according to the following inclusion criteria [3]: (1) the presence of non-progressive, neurological disorders that appeared in infancy or early childhood; (2) the appearance of retardation including the lack of development or ability in sporting activities and the persistence of posture obstacles, which may be associated with lesions that affect feeling, perception, cognition, communication, and behaviour, epilepsy or secondary musculoskeletal problems; (3) no diagnosis of other diseases, such as metabolic or degenerative diseases, that could induce central paralysis and transient retardation in a normal person; and (4) negative serological markers for AIDS, hepatitis, and syphilis. Patients with severe anaphylactic or autoimmune diseases were excluded from the transplantation group.

Autologous MSC cultures

Patients fasted for 4–6 hours and went without water 1 hour before the collection of autologous bone marrow specimens. The patients were injected with 5 mg/kg phenobarbital, 0.01 mg/kg atropine, and 4 mg/kg ketamine in the aseptic collecting room 30 minutes before specimen collection. In addition, 8–25 mL of bone marrow (0.6-1.0 mL/kg of body weight) was collected from the right posterior superior iliac spine according to conventional bone marrow aspiration procedures. A total of 100 U/mL heparin was added to the collecting tube as an anticoagulant. Following current manufacturing best practices, mononuclear bone marrow cells were isolated by Percoll (1.073 g/mL) centrifugation and were allowed to adhere to a flask for 72 hours in low glucose Dulbecco’s Modified Eagle’s Medium (Gibco-Invitrogen) at 5% CO₂ and 37°C, and the media were changed every 3 days. The cells with a strong positive signal for the multipotential marker Oct 4 and Nanog were screened for further cloning and culturing (Additional file 1: Figure S1A). The cell phenotypes were assessed by flow cytometry. The results indicated that CD34 and CD45 were negative, and the positive rates of CD29 and CD44 were above 95% (Additional file 1: Figure S1B). The ability of the cells to differentiate into adipocytes and osteocytes in culture was confirmed in vitro following the criteria of the 2006 International Society of Cellular Therapy [21]. At 70%–80% confluence, the cells were detached and replated at 5 × 10⁶/175 cm² in culture to process for neural differentiation and confirm negativity for endotoxin, hepatitis C virus, hepatitis B virus, HIV, syphilis, fungus, Myco plasma species, and Chlamydia before infusion. A G-banding karyotype analysis was performed to confirm the absence of chromosomal aberrations in the final cellular product. The MSCs were continuously cultured without cryopreservation and were thawed before transplantation. Supplementary methods were shown in more detail (see Additional file 1).

Autologous NSC-like cells induction and differentiation

At 30%-40% confluence, a combination of 20 ng/ml recombinant human FGF basic (bFGF) and 10 nM retinoic acid was added to the medium for 12 hours to transform the MSCs, and the differentiation capacity was evaluated after 14 days [14]. Growth curves were drawn to assess the proliferation ability of the MSCs and the transformed cells (Figure 2A). In total, 1 × 10⁴ of the transformed cells were collected to prepare slices for immunofluorescence staining, and 50%–70% of the cells expressed Nestin and Tuj-1 (Figure 2B). A total of 1 × 10⁶ cells were co-labelled with antibodies against Nestin (BD Pharmingen) and Tuj-1 (Abcam) , Sox2 (Abcam) or Sox1 (Santa Cruz). According to the flow cytometry analysis, 66.83%, 90.92% and 64.71% of the induced cells expressed both Nestin and Tuj-1, Sox2 and Sox1, respectively (Figure 2B). These results suggest that the MSCs differentiated into NSC-like cells. The NSC-like cells were detached, washed three times in saline, re-suspended in saline with 5% human albumin and filtered in a 100-μm nylon filter. Approximately 100 μl of this solution was used for safety and viability tests. No chromosomal aberrations or increased telomerase activity was found for the induced NSC-like cells. In addition, a high cell viability (>90%) and the negative microbiological results ensured the quality of the cells before the transplantations. To determine whether the induced NSC-like cells could differentiate into different neural lineages, MSCs were induced for 14 days under the same conditions and were stained using antibodies against Tuj-1 (Abcam), NeuN (Abcam), Tau (Abcam), Nestin (Santa Cruz), CD44 (Santa Cruz) and GFAP (Santa Cruz). The induced cells expressed both neuron- and glial cell-specific proteins (Figure 2C).

Safety evaluation of NSC-like cells in vivo in a mouse model

The MSC-derived NSC-like cells (1.0 × 10⁶) were implanted subcutaneously into the flanks of nude mice or were transplanted intraspinally into the subarachnoid space of nude mice. No tumour formation or other adverse events were observed in mice after 3 months. The animal studies were approved by the Institutional Animal Care and Use Committee of the Second Military Medical University in Shanghai, China.

Cell transplantation therapy

Before the lumbar puncture, the patients fasted without food and water. The patients were injected with phenobarbital sodium, atropine, and ketamine according to the dose and usage guidelines before the collection of bone marrow. When the patients fell asleep, a local anaesthetic was infiltrated under the skin and a spinal needle was inserted between the lumbar vertebrae L3/L4 or L4/L5 and pushed in until the needle traversed the dura mater and the thin arachnoid membrane into the subarachnoid space. The stylet from the spinal needle was withdrawn, and drops of cerebrospinal fluid (up to 5 ml) were collected. Then, 5 ml cell suspension, which contained 1-2 × 10⁷ of NSC-like cells, was gently injected into the subarachnoid cavity through the spinal needle. The opening pressure of the cerebrospinal fluid was measured during specimen collection using a simple column manometer. The procedure ended by withdrawing the needle while placing pressure on the puncture site. The cell transplantation was administered at intervals of 3 weeks for two cycles. In addition, the patients in the transplantation group received rehabilitation treatments at the same time.

Observation indices

As the primary measurement, the motor and language abilities of the patients were observed before treatment and at 1, 3, and 6 months post-treatment. Motor ability was assessed using the Gross Motor Function Measure (GMFM-88 or GMFM) [22] questionnaire. An example was shown in Additional file 1. Language ability was measured using the Gesell questionnaire. The language developmental quotient (LDQ) was determined using the following formula: language developmental age/present age × 100. All of the assessments were conducted by a specifically assigned person. For the observer-blinded design of the study, the patients were randomly arranged before the tests, and the examiner was blinded to the names and the grouping of patients. The assessment results were recorded in the case reports.

The secondary measurement was patient survival and the incidence of the following adverse events: (1) fever, headache, allodynia, vomiting, infection, or other reactions; (2) clinical seizures and/or epileptic discharges on a serial electroencephalogram (EEG) at any time during hospitalization; (3) new lesions in the skull according to MRI at 6 months post-procedure; and (4) other clinically significant complications from the procedure during the long-term folow-up.

Statistical analysis

The data were expressed as the mean ± S.E.M. The statistical significance of the patient ages, GMFM scores and language developmental quotients was detected using the one-way ANOVA analysis, LSD test and two-tailed Student’s t-test. The statistical significance of gender and the GMFCS levels was detected using the chi-squared test. The statistical analysis was performed using SPSS 11.0.

Baseline characteristics of the patients

A total of 30 patients with cerebral palsy in GMFCS levels III-V were enrolled in this study and were included in the transplantation group. The group consisted of 14 males and 16 females with a mean age of 5.53 ± 1.20 years (range, 1–32 years). An additional 30 patients with cerebral palsy were treated with conventional methods and were included in the control group, which consisted of 14 males and 16 females with a mean age of 4.66 ± 1.31 years (range, 1–35 years). The baseline characteristics of the patients in the two groups at the study onset are shown in Tables 1–2. In addition, a comparison of the patient ages, LDQs and GMFM scores in the two groups are shown in Figure 3. No significant differences were found.

No.	Gender	GMFCS stage	Age (years)	LDQ	GMFM
1	Female	V	1	72	36
2	Female	III	3	31.7	176
3	Male	III	4	28	189
4	Male	III	4	92	185
5	Male	V	3	100	50
6	Male	IV	2	38	101
7	Male	V	4.5	15	23
8	Male	IV	4	26.8	115
9	Female	V	3	75	42
10	Female	IV	13	82	105
11	Female	V	1.5	62	38
12	Male	V	1	70	31
13	Female	V	1	75	20
14	Male	IV	2	90	63
15	Male	III	4	98	191
16	Female	IV	3	30	96
17	Male	IV	21	80	82
18	Female	III	2	98	120
19	Male	IV	5	85	86
20	Female	III	8	10	115
21	Female	III	4	90	183
22	Female	V	5	10	23
23	Female	III	12	100	195
24	Male	IV	2	98	83
25	Male	V	1	15	12
26	Female	III	6.5	90	193
27	Female	III	4	92	196
28	Female	IV	32	73	117
29	Female	V	1	54	56
30	Male	III	8.5	97	201

The clinical characteristics of the 30 patients in the transplantation group before the cell transplantation treatment are shown. GMFCS: Gross Motor Function Classification System levels; LDQ: language developmental quotient; GMFM: Gross Motor Function Measurement scores.

No.	Gender	GMFCS stage	Age (years)	LDQ	GMFM
1	Female	V	1	72	36
2	Female	III	2.5	35	183
3	Male	III	3.5	28	198
4	Male	III	4	92	211
5	Male	V	2	100	35
6	Male	V	2	38	23
7	Male	V	4.5	15	33
8	Male	IV	3.5	27	110
9	Female	V	2	75	51
10	Female	IV	2.5	82	99
11	Male	IV	1	60	80
12	Female	III	15	95	201
13	Female	III	1	62	135
14	Male	IV	5	86	110
15	Female	IV	2.5	95	98
16	Male	IV	21	98	135
17	Male	IV	35	82	90
18	Male	IV	9	80	112
19	Female	V	4	12	30
20	Female	V	1	23	43
21	Male	V	1.7	25	45
22	Female	V	1	30	50
23	Female	III	1.5	80	146
24	Female	V	3	39	25
25	Male	IV	2.5	96	99
26	Male	V	2.6	100	28
27	Female	III	2	86	100
28	Female	V	1	65	45
29	Male	III	1.5	85	125
30	Male	V	1	64	43

The clinical characteristics of the 30 patients in the control group are shown. GMFCS: Gross Motor Function Classification System levels; LDQ: language developmental quotient; GMFM: Gross Motor Function Measurement scores.

The MRI results from the skulls of patients with cerebral palsy can show a periventricular white matter injury (PWMI), and the pathological findings can include periventricular leukomalacia and diffused dysmyelination [23, 24]. Prasad [25] studied 102 children 1–3 years of age with cerebral palsy and found that 47.1% had a PWMI in the skull based on MRI findings. In this study, 16 of the 30 patients (54.5%) in the transplantation group had white matter injuries according to MRI.

Functional outcomes of motor development among the two groups

The motor functions of the patients were observed and evaluated using the GMFM scores before treatment and at 1, 3, and 6 months post-treatment (Table 1). The GMFM scores of all of the patients in the two groups gradually increased during the 6 months post-treatment, however, there were no significant differences in the control group. In contrast, the GMFM scores in the transplantation group at months 3 and 6 post-treatment were significantly higher compared with the baseline scores (P= 0.011 and 0.001), whereas there were no significant changes in the GMFM scores at month 1 post-treatment (P= 0.265). The GMFM score increase from baseline to month 6 post-treatment was 58.6 (95% CI: 25.8-91.4) in the transplantation group. This finding suggests that there was a significant recovery effect in motor function after the transplantation treatment. Additional movie files were shown in more detail (see Additional file 2 and 3). In addition, we found that the GMFM scores at months 3, and 6 post-treatment were significantly higher in the transplantation group compared with those in the control group (P= 0.003 and P< 0.001, respectively). However, there were no differences in the GMFM scores at month 1 post-treatment between the two groups (P= 0.089) (Figure 4A). To evaluate the impact of NSC transplantation in patients with different levels of cerebral palsy, we divided the patients according to their GMFCS levels. The results indicated that patients in the transplantation group with levels IV and V had a better recovery of motor function (Figure 4B-D).

Functional outcomes of language development among the two groups

The language abilities of the patients were observed before treatment and at 1, 3, and 6 months post-treatment. The means of the language developmental quotients of the patients in the two groups gradually increased during the 6 months post-treatment (Figure 5). However, no significant differences in the language developmental quotients were observed at months 1, 3, and 6 post-treatment when compared with the baseline quotients in both groups (P>0.05 for all). In addition, we did not find any significant differences in the language developmental quotients at months 1, 3, and 6 post-treatment between the two groups (P= 0.751, 0.522 and 0.304, respectively). This finding suggests that the recovery of language functions was not accelerated in the transplantation group.

Adverse reactions in the transplantation group

No new neurological deficits were immediately identified after NSC treatment. All of the patients in this study were observed for more than 6 months after treatment. None of the 60 patients experienced fever, headache, allodynia, vomiting, or other serious adverse events that were related to the experimental procedure. Follow-up MRI evaluations did not show any significant anatomical or structural changes that could indicate adverse events. There was no evidence of any new ischemic, haemorrhagic, or neoplastic lesions. However, an increase in the frequency of crying was reported in one patient but resolved spontaneously 2 days after the treatment without any intervention. Other possible adverse effects or complications in the control and transplantation groups were evaluated, but no differences were observed in the frequency of new-onset morbidities between the groups.