Abstract
Cerebral palsy (CP) is a complicated disease with varying causes and outcomes. It has created significant burden to both affected families and societies, not to mention the quality of life of the patients themselves. There is no cure for the disease; therefore, development of effective therapeutic strategies is in great demand. Recent advances in regenerative medicine suggest that the transplantation of stem cells, including embryonic stem cells, neural stem cells, bone marrow mesenchymal stem cells, induced pluripotent stem cells, umbilical cord blood cells, and human embryonic germ cells, focusing on the root of the problem, may provide the possibility of developing a complete cure in treating CP. However, safety is the first factor to be considered because some stem cells may cause tumorigenesis. Additionally, more preclinical and clinical studies are needed to determine the type of cells, route of delivery, cell dose, timing of transplantation, and combinatorial strategies to achieve an optimal outcome.
Keywords
Introduction
Cerebral palsy (CP) is a heterogenous group of neurological disorders that are mainly observed in infants. It results due to a static brain lesion at the time of pregnancy or early life. However, CP is restricted to lesions of the brain. Therefore, diseases specific to the peripheral nerves of the spinal cord, such as traumatic peripheral nerve lesions, spinal muscular atrophy, myelomeningocele, vascular malformations, or tumors of the spinal cord, or inherited metabolic disorders, metabolic myopathies, metabolic neuropathy, or movement disorders, although causing early motor abnormalities, are not considered CP (96). CP involves damage to several types of cells in the brain with complex manifestations, so that a series of interventions of effective therapeutic strategies, including pharmacological and surgical treatments, and a multidisciplinary team composed of physiatrists, orthopedic surgeons, neurologists, specialists in genetics, pulmonologists, gastroenterologists, nutritionists, ophthalmologists, audiologists, and dentists are required to provide a comprehensive therapeutic plan for patients with CP (29). In fact, CP, an incurable illness, has caused significant burdens to both affected families and societies, not to mention the quality of life of the patients themselves. Stem cells are multipotent and take cues from their environment to guide differentiation and cellular activity. Although the mechanisms have yet to be determined, the transplanted stem cells have been proven to be capable of replacing damaged neurons, glia, and vasculature, and/or releasing trophic factors in the treatment of CP in animal models. Also, there are several early human trials showing effective and promising results in the functional improvements of CP patients (Table 1). These encouraging results have opened up a new era for utilization of stem cells for patients with CP.
Stem Cell Trials for Cerebral Palsy
See www.clinicaltrials.gov for more information. NK indicates that the information is not known. Reh, rehabilitation; BM, bone marrow; BMMNC, bone marrow mononuclear cells; EPO, erythropoietin; GCSF, granulocyte colony-stimulating factor; UCMSC, umbilical cord mesenchymal stem cell.
Epidemiology
In general, population-based studies from around the world report prevalence estimates of CP ranging from 1.5 to more than 4 per 1,000 live births or children of a defined age range (10,15,97,164). In developed countries, the overall estimated prevalence of CP is 2–2.5 cases per 1,000 live births. In the developing world, the prevalence of CP is not well established, but estimates are 1.5–5.6 cases per 1,000 live births. There are about 1.7 to 2.0 per 1,000 1-year survivors in the US (164), 2.0 per 1,000 live births in the UK (137), 2.61 per 1,000 school-age children in France (116), 1.92 per 1,000 children aged 1–6 years old in China (73), and 3 in 1,000 inhabitants in Taiwan (23). Additionally, the prevalence of CP was 72–86, 32–60, 5–6 per 1,000 live births for extremely preterm children (<28 weeks), very preterm (28–31 weeks), and moderate preterm (32–36 weeks), respectively (6). The prevalence of CP ranges from 90 cases per 1,000 neonatal survivors weighing less than 1,000 g to 1–5 cases per 1,000 for those born weighing 2,500 g or more (121). These figures suggest that CP prevalence is inversely associated with gestational age and birth weight despite significant advances in the medical care of neonates (6). However, a decreasing prevalence in low birth weight infants has been noted in Europe (104) and in an Australian cohort. Inconsistent diagnostic criteria, the lack of healthcare access, and no responsible authorities or institutes regarding CP in some countries may cause discrepancies.
Risk Factors and Pathophysiologial Mechanisms
Causes of CP are multifactorial, including intrauterine growth restriction, prenatal and perinatal asphyxia, injury or abnormal development at prenatal, perinatal, and postnatal stages, postterm pregnancy at 42 weeks or later, low birth weight, prematurity, fetal infections, multiple pregnancies, periventricular leukomalacia (PVL), intraventricular hemorrhage, vascular insufficiency, toxins, chorioamnionitis, maternal factors, underlying genetic abnormalities, increased prevalence in male infants, and lower socioeconomic status (29,36,93,112,113,119,166). The pathophysiological mechanisms that result in CP can be classified as (i) hypoxia and ischemia that triggers a cascade of excito-oxidative events in the brain (45), (ii) intrauterine infections/inflammation that cause a fetal inflammatory response syndrome and neuroinflammation (112), (iii) prematurity (145), (iv) PVL (159), and (v) genetic or other congenital causes (119). The pathogenesis of CP may be explained by each or a combination of these mechanisms.
Clinical Manefestations
CP is characterized by abnormalities of movement and posture. Young children are often identified with CP if they fail to meet expected developmental milestones or to suppress obligatory primitive reflexes in the first year of life. Children with the presentation of hypotonia or hypertonia, weakness of one side, asymmetrical crawling or failure to crawl, growth disturbance, underdevelopment or absence of postural or protective reflexes, gait abnormalities, such as the crouch position with tight hip flexors and hamstrings, weak quadriceps, and/or excessive dorsiflexion, scissoring legs, abnormal flexion and extension of knees, and equinus or toe walking, and varus or valgus of the foot should be tested for CP. Children may also present with abnormal sensation and perception, impairment of speaking, eating, swallowing, hearing, eyesight and coordination of eye movements, lack of coordination and mental underdevelopment, difficulties in walking, balance, and motor control, learning disabilities, and bladder and bowel control (113). The 2003 American Academy of Neurology practice parameter suggests if any child with mental retardation, ophthalmologic and hearing impairments, speech and language disorders, and oromotor dysfunction should be screened for potential CP-associated deficits at the initial assessment (11).
Classification
The CP syndromes may be classified by the predominant type of motor disturbance, such as spastic, dyskinetic, and ataxic. Spastic CP is the most common subtype, occurring in 80% of cases. On the basis of affected extremities, spastic CP is further classified into diplegic (21%), hemiplegic (31%), quadriplegic (35%), and monoplegic with other forms of CP, including dyskinetic, hypotonic, mixed, and ataxic CP contributing to the remainder (13%) (128).
Spastic Hemiplegia
Spastic hemiplegia affects one side of the body, including an arm and a leg, with involvement of upper extremity spasticity more than lower extremity spasticity. In unilateral lesions of the brain, the vascular territory most commonly affected is the middle cerebral artery; the left side is involved twice as commonly as the right. Other structural brain abnormalities include hemibrain atrophy and posthemorrhagic porencephaly. In premature infants, this may result from asymmetric PVL.
Spastic Diplegia
Spastic diplegia is the more common form that occurs when the lower extremities are more affected by stiffness and tightness. This form may present a scissor gait. Overtime, additional problems such as hip dysplasia can occur. Patients with spastic diplegia often have a period of hypotonia followed by extensor spasticity in the lower extremities. Patients have a delay in developing gross motor skills, impaired intelligence, impaired vision, cognitive impairment, and seizures. In premature infants, spastic diplegia may result from parenchymal–intraventricular hemorrhage or PVL. In term infants, the etiology is still not clear.
Spastic Quadriplegia
Spastic quadriplegia is the most severe form, and it affects all four extremities, the trunk, and the head. The patient often has a history of a difficult delivery with evidence of perinatal asphyxia. Preterm infants may have PVL. Full-term infants may have structural brain abnormalities or cerebral hypoperfusion in a watershed distribution. Patients may present some cognitive impairment, oromotor dysfunction, full-body hypertonia or truncal hypotonia with extremity hypertonia, oromotor dysfunction, increased risk of cognitive difficulties, and seizures.
Dyskinetic CP
Dyskinetic CP is caused by damage to the basal ganglion and/or cerebellum and is illustrated by extrapyramidal movement patterns, decreased head and truncal tone, and defects in postural control and motor dysfunction, such as athetosis, chorea or choreoathetosis, and dystonia. This type of CP is often associated with kernicterus, hypoxic ischemic injury, and/or hyperbilirubinemia.
The classification of CP is based on the motor dysfunctions and the affected limbs. The Surveillance of Cerebral Palsy in Europe group recommended an alternative classification: unilateral versus bilateral motor involvement (21). The Gross Motor Function Classification System (GMFCS; for CP patients less than 12 years of age) (94) and the GMFCS-Expanded and Revised (GMFCS-ER; for CP patients 12 to 18 years of age) (95) are five-level classification systems that describe the gross motor function of children and youth with CP on the basis of their self-initiated movement with particular emphasis on sitting, walking, and wheeled mobility. The Manual Ability Classification Systems (37) are five-level classification systems that describe how children with CP use their hands to handle objects in daily activities. The Drooling Impact Scale can evaluate the impact of drooling in children with developmental disabilities (106). The Viking Speech Scale can be used to investigate the speech performance of children with CP (102). These systems can help clinicians, care providers, and family to subjectively conduct standardized functional assessments and comparisons.
Diagnosis
Laboratory tests are not necessary to confirm the diagnosis of CP, as it is based on the patient's medical history and examination. CP is nonprogressive, and children with CP usually present with a delay in reaching early developmental milestones. As the brain continues to develop postnatally, abnormalities of motor tone or movement in the first several weeks or months after birth may gradually improve over the first year of life, making a corrective diagnosis of CP much more difficult. Therefore, the following several tests are used to rule out other causes, including (11) coagulation studies, lactate and pyruvate, cerebrospinal fluid protein, thyroid function studies, electrophysiological studies, including electroencephalography, electromyography, nerve conduction studies, and evoked potentials (70), and neuroimaging, which are recommended in the evaluation of a child with CP if the etiology has not been established. The incidence of MRI abnormalities in the cohort of patients with CP has be reported from 70% to 90% (54). It should be emphasized that normal results from neuroimaging studies do not exclude a clinical diagnosis of this disorder. In these cases, it is important that metabolic or genetic screens, chromosomal analysis, and specific DNA testing be considered (119) and excluded before making a diagnosis of CP. A good example is that a normal neuroimaging study in a child with dystonic CP or spastic diplegia may suggest a neurotransmitter disorder (89). In children with spasticity of the legs and worsening of bowel and bladder function, a spine MRI should be considered (55,108).
Animal Models
There are several animal models for CP based on the possible pathophysiological mechanisms. (i) Hypoxia and ischemia: transiently ligated single or bilateral carotid arteries or middle cerebral artery, leading to brain injury is followed by reduced myelination, ventricular enlargement, loss of neurons, damage to axons and dendrites, and alterations in neurobehavioral performance (47,129,148, 149,175), supporting white matter damage. However, these models are associated with highly variable outcomes. (ii) Intrauterine infections/inflammation: lipopolysaccharide, a potent inducer of innate immune responses such as inflammation, administered intracerebrally during the early neonatal period or intraperitoneally to the pregnant dam, results in inflammation and hypomyelination in the offspring (19,161). (iii) Activation of the excitotoxic cascade via the N-methyl-D-aspartate and glutamate receptors to mimic white and gray matter damage in the premature brain (60). (iv) Ibotenate and quinolinate induce cystic formation in the white matter, cortical necrosis, which are similar to PVL (57). These animal models may help elucidate the possible mechanisms behind this disease and evaluate the efficacy of any potential treatments for CP.
Treatments
CP involves damage to several types of cells in the brain with complex manifestations. All treatments are targeted at improving function and reducing disability. Treatment programs for CP encompass (i) pharmacological interventions, include botulinum toxin, type A, phenol, benzodiazepam, trihexyphenidyl, tetrabenazine, carbidopa-levodopa, and dantrolene for dystonia and spasticity; antiepileptic drugs for the termination of clinical and electrical seizure activity (22,33); (ii) surgical interventions, including intrathecal baclofen pump insertion (33), selective dorsal rhizotomy (46), and basal ganglia stimulation (157) for antispasticity, improving rigidity, choreoathetosis, and tremor, and orthopedic surgical intervention for scoliosis or hip dislocation (62,167); (iii) physical and behavioral therapy; (iv) mechanical aids; and (v) management of associated medical conditions. Although these treatments are helpful, none of them have achieved a satisfactory recovery of the damaged brain. Recent strides in stem cell research provide the possibility of developing a complete cure in treating CP.
Human pluripotent stem cells have the capability to proliferate and self-renew over long periods of time and to differentiate into almost any cell type. Stem cells have been utilized in several animal models of degenerative neurological diseases such as Parkinson's disease (PD) (40) and PolyQ diseases (41). Grafting mesenchymal stem cells to a mouse cerebellum in an animal model of cerebellar ataxia improved motor functions and rescued degenerated Purkinje cells (61). The injection of fetal cortical grafts into immature rat brains with focal infarctions rescued motor function (133), spatial learning, and memory (52), suggesting that stem cell transplantation could be used therapeutically to restore the CNS functions through the replacement of lost cells, inducing neurogenesis, and/or releasing various molecules, which are neuroprotective or modulate inflammation. Stem cells have been generated from cells isolated from the inner cell mass of the blastocyst (embryonic stem cells, ESCs), from the subventricular and hippocampal zones of the brain and the ependymal zone of the spinal cord (neural stem cells; NSCs), bone marrow mesenchymal stem cells (MSCs), somatic cells (induced pluripotent stem cells, iPSCs), umbilical cord blood (UCB cells), and primordial gonadal ridge (embryonic germ cells), which will be discussed below.
Embryonic Stem Cells (Escs) and Induced Pluripotent Stem Cells (iPSCs)
ESCs are isolated and expanded from the inner cell mass of the mammalian blastocyst stage embryo (144). ESCs are proven to have a normal karyotype (18), express high levels of telomerase activity (49), and have specific pluripotent intracellular and cell surface markers (107). ESCs are pluripotent and are capable of unlimited self-renewal in vitro and have the ability to give rise to all tissue types in the body, including neurons, astrocytes, and oligodendrocytes (13,91). Human ESCs can effectively migrate and differentiate into mature cell types after implantation (173). Therefore, a number of human ESC lines have now been generated (107,144). However, undifferentiated ESCs cannot be used in the treatment of neurological illness because they appear to have immune privilege secondary to reduced major histocompatibility complex antigen expression (77), allowing them to escape immune surveillance in the host to form tumors such as teratomas. Moreover, ESCs and embryonic carcinoma cells share many similar properties (35,132), such as the same markers (81), suggesting an intimate link between stem and tumor cells as well as between pluripotency and tumorigenicity (67). Furthermore, lack of genetic stability and long-term survival, high expression of oncogenes, the moral and ethical issues, and possible accumulating genetic abnormalities in replication raise serious concerns on the use of these cells. Interestingly, intrauterine infection with cytomegalovirus (CMV) is thought to be responsible for CP (140). An interesting report shows that ESCs are more resistant to CMV than most other cell types because ESCs express less heparin sulfate, β1-integrin, vimentin, and nuclear pores (64). This phenomenon may reduce the ability of CMV to attach to and enter the cellular membrane, translocate to the nucleus, and cross the nuclear membrane in ESCs. This finding may open a new page for understanding the pathogenesis and creating a new treatment for congenital anomalies caused by CMV infection in humans.
iPSCs and ESCs are both pluripotent stem cells. iPSCs can be obtained by inducing terminally differentiated somatic cells (skin fibroblasts or blood cells) via nuclear reprogramming. This has been achieved by transducing mouse or human somatic cells with retroviral vectors containing cDNA encoding genes of the “Yamanaka” transcription factors, including octamer-binding transcription factor 4, c-Myc, sex-determining region Y-box 2, and Kruppel-like factor 4. These factors are capable of dedifferentiating the fibroblasts or blood cells into pluripotent cells (171). These cells can be further differentiated into cells of neural lineages (158). iPSCs exhibit several characteristics similar to ESCs, including morphology, proliferation, epigenetic status, and pluripotency (142,171). However, unlike ESCs, iPSCs can be generated from animals, healthy humans, or patients with degenerative neurological diseases, such as amyotrophic lateral sclerosis (ALS) (34) and PD (131). These cells also have the benefit of alleviating ethical concerns (90,143). Human iPSCs may be an ideal source for personalized cellular therapeutics for disease treatment, organ repair, screening new drugs and potential teratogens, and studying mechanisms of pathogenesis and toxicology (4,90). Although the transplantation of iPSCs with normal or better functional activity to patients with degenerative diseases is seemingly going to be a new therapeutic strategy, iPSCs are reported to produce teratomas, neuroblastomas, and follicular carcinoma (92). Mice genetically derived to contain some tissues from iPSCs have a malignant tumor incidence of 20% (92). Genetic changes may pose a risk of enhancing tumorigenesis in cellular therapy with iPSCs. Human ESC-derived oligodendrocyte progenitor cells (OPCs) were reported to ameliorate function in rats undergoing traumatic spinal cord injury (67,130). Similarly, human ESC-derived OPCs were transplanted in a rat model of spinal cord injury. Seven days later, the result showed significant improvement in motor functions compared to untreated animals (65). Therefore, iPSC-derived OPCs may be considered candidates for the treatment of CP.
Neural Stem Cells (NSCs)
NSCs, identified in the subventricular and hippocampal zones of the brain and the ependymal zone of the spinal cord in humans (3,43) are the earliest uncommitted multipotent cells in the CNS (9). NSCs can extensively proliferate, self-renew (9), and differentiate into regionally appropriate cell types, such as cortical projection neurons (38), interneurons (120), hippocampal pyramidal neurons (30), oligodendrocytes (127), and astrocytes (122) to replace and or repair damaged cells and participate in CNS development and functions. NSCs can be isolated from almost all parts of the embryonic/fetal brain, be propagated in vitro, and subsequently implanted into the brain of animal models of human neurological disorders (63). There is evidence to suggest that NSCs in different regions of brains may possess variable proliferative, self-renewal, and differentiation abilities (8). This may explain the existence of histological and imaging evidence of endogenous neurogenesis and subventricular zone growth following brain injury in mammals and young rodents that cannot always achieve a complete functional recovery (155). In a human trial, NSCs derived from autologous bone marrow (BM) MSCs were intrathecally injected into 30 patients with CP. On follow-up, no adverse events were recorded, and patients showed significant improvement in motor and language performance compared to the control group (25). Although this trial indicates that NSCs are effective for the treatment of motor deficits related to CP, the possible connection between tumorigenesis and NSC transplant still raises serious concerns (2,5).
Mesenchymal Stem Cells (MSCs)
BM contains hematopoietic and MSCs. MSCs are a subset of the adherent stromal cell fraction, including plate-adhering, fibroblast-like cells possessing self-renewal ability with the capacity to differentiate into skeletal, cardiac muscle, lung, and liver cells (154), functional neurons (68,86), astrocytes, and oligodendrocytes (50,138). Apart from the BM, MSCs can be isolated from cord blood and the stroma of the umbilical cord and placenta (99). MSC transplantation is reported to be able to minimize immune reactions, show no adverse reactions in allogenic MSC transplants (14,80), and not produce expression of major histocompatibility complex class II antigens. Therefore, MSC transplantation is able to prevent the development of graft-versus-host disease (118). Regarding the homing effect of MSCs, a study using magnetically labeled MSCs, which were transplanted in a model of perinatal brain injury, showed that cells migrated away from the injection site toward lesioned areas in both hemispheres by MRI scan and microscopy postmortem (24), suggesting that MSCs have a capacity for precise migration to even widespread and distant damaged areas of the CNS. The inflammatory response that accompanies many CNS disease processes and inflammation-associated mechanisms may play a pivot role in the migration of MSCs or other stem cells toward regions of brain injury and degeneration. In inflammation, chemokines are released by many different cell types and serve to guide cells of both the innate immune system and adaptive immune system (56,177). More than 50 different chemokines and 20 different chemokine receptors have been cloned. The α-chemokine stromal-derived factor (SDF-1α) binds exclusively to CXC chemokine receptor 4 (CXCR4) (12). CXCR4 is found to express on MSCs, and SDF-1α navigates MSCs to ischemic and degenerative lesions to repair injured tissues in the brain (7,79). The dynamic of SDF-1α interaction with CXCR4 appears to have at least one requirement for MSC homing toward sites of injury. Moreover, astrocytes and endothelial cells may play a role in neurogenesis because both can upregulate SDF-1α after injury (115,176). Chemokines are crucial in injury-mediated homing, tethering, rolling, and firm adhesion to inflamed endothelial cells, and extravasation into inflamed CNS areas, all of which are sequentially mediated by the constitutive expression of functional cell adhesion molecules (such as CD44), integrins (such as α4, β1) and chemokine receptors (such as CXCR4) on the surface of MSCs. The cross talk between these molecules and other unrecognized factors increases the complexity of the homing phenomenon. Importantly, MSCs can promote angiogenesis (74) according to pathological analyses, which revealed higher blood vessel densities surrounding the MSC-injected brain (42). MSCs can provide physical scaffolding for elongating axons (7), mobilize endogenous NSCs (16), be less likely to induce tumorigenicity (160), and be used in demyelinations (117). A phase I clinical trial confirmed that MSC transplantation into the spinal cord of ALS patients is safe (82). MSC transplantation was also performed in a 6-year-old girl with CP. Three months later, the results showed significant improvement in motor, sensory, cognitive, speech, bowel, and bladder control functions after intrathecally multi-injected autologous BM-derived mononuclear cell (more than 108 cells) transplantation (105). Another case is a 20-year-old male adult with diplegic CP and mental retardation. Six months after intrathecal autologous BM-derived mononuclear cell (106 cells) transplantation, he demonstrated distinct improvement in IQ, social behavior, speech, balance, and daily functions, and a significant increase in the metabolic activity in the lesioned areas of the brain, as determined by positron emission tomography and computer tomography (PET-CT) scans (126). Transplantation of MSCs can significantly attenuate inflammatory cytokines in the CSF and brain tissue, reduce brain apoptosis and reactive gliosis, and improve abnormal sensorimotor functions in severe intraventricular hemorrhaging (IVH). Therefore, MSC transplantation may be applied in the treatment of posthemorrhagic hydrocephalus after severe IVH in premature infants. This may lead to decreased mortality and amelioration of long-term neurological morbidities, such as CP in the survivors (1). These features demonstrate that MSC transplantation into human neonates with cerebral damage may constitute a promising and realistic treatment modality for regenerating the damaged brain.
Umbilical Cord Blood (Ucb) Cells
UCB and the placenta were regarded as medical waste after a baby's birth, but now they are treated as a source of stem cells because UCB is rich in highly proliferative MSCs (72) and progenitor cells (58). UCB is capable of self-renewal and giving rise to cells derived from the endodermal, mesodermal, and ectodermal lineages (69,83,84,111) to develop organs (17). Advantages of UCB cells include easy collection, little risk to infant or mother, few side effects, and low immunogenicity (110). UCB cells have been shown to lessen the clinical and radiographic impact of hypoxic brain injury and stroke in animals (26,85,141, 156,163). Hypoxic injury animals receiving UCB cells showed homing or engraftment of UCB cells in the lesion and improvements in sensorimotor reflexes through reduction of caspase 3 cleavage and activation of microglia and macrophages at various brain regions (103). Intraperitoneal or intrathecal injection of UCB cells into animals has been shown to ameliorate neurological and motor deficits in a CP model by reducing the levels of proinflammatory cytokines, including interleukin-1α (IL-1α), IL-1β, and tumor necrosis factor-α (114,162). UCB cells may lessen acute and chronic graft-versus-host disease (109), suggesting that UCB cells may possess antiapoptotic and anti-inflammatory properties. Kurtzberg's group reported that the neurological function of patients with Krabbe's disease recovered significantly better if treated early (like 6 months old) compared to those treated after 2 years of age by UCB cells intravenously transplanted (39). This novel finding inspires the idea that UCB cells maybe able to lessen the symptoms and facilitate endogenous repair and neuroplasticity after injury to the brain and spinal cord.
Human Embryonic Germ (hEG) Cells
hEG cells, also called primordial germ cells (PGCs), derived from PGCs in the gonadal ridge of first-trimester embryos (147), express a broad spectrum of gene markers and have been induced toward ecto- and endodermal lineages. hEG cells are unipotent in that they are lineage restricted to become germ cells. PGCs in mice are derived from the epiblast that mainly gives rise to the extraembryonic mesoderm. hEG cells in humans first appear between the third and the fourth week postfertilization in the endoderm of the dorsal wall of the yolk sac, very close to the allantois, then migrate through the hindgut during the fourth week and the dorsal mesentery in the fifth week to reach the genital ridge (101). hEG cells can replace the damaged neurons and oligodendrocytes in the forebrain of neonatal mice (88), and their neural progenitors are capable of robust and long-term growth. The growth factor bone morphogenetic protein 4 plays an important role in affecting hEG cell survival, and disruptions in this gene lead to defects (51). However, the low efficiency, limited proliferation rate, and difficult control of differentiation may restrain the use of hEG cells (123).
Factors Affecting Stem Cell Therapy
Stem cell therapy can restore or preserve brain function by replacing and repairing neurons. However, several critical factors related to the stem cell therapy should be determined and optimized before the treatment. (i) Selection of cells: ESCs, NSCs, and iPSCs may cause tumorigenesis and the safety and feasibility and MSCs and UCB in the treatment of patients with CP are verified. Therefore, neurons or OPCs derived from ESCs, iPSCs, UCBCs, or MSCs may be options for the treatment of CP. However, the therapeutic efficiency of these cells in CP has not been objectively compared. (ii) Selection of trophic factors: erythropoietin (EPO), insulin growth factor 1, granulocyte colony-stimulating factor, transforming growth factor-α, and others may significantly affect the growth and functions of stem cells. However, it is not known which factor(s) is/are essential to stimulate and assist specific stem cells in treating patients with degenerative neurological diseases. (iii) Selection of dosing: The minimum therapeutic cell dosage should be determined to prevent any potential microembolism. In general, the minimum therapeutic cell dosage for cell engraftment is 1 × 107 cells per kilogram. Additionally, the available dose of autologous cells obtained at birth may be insufficient for transplantation at an older age (135). Therefore, the status of recipients should be assessed before the stem cell treatment. Moreover, a study showed that two MSC injections at 3 and 10 days after neonatal hypoxia–ischemia (HI) markedly improved sensorimotor function 4 weeks after the insult more than a single injection (152), suggesting multiple injections may enhance the therapeutic effect of the stem cell treatment in the neurological illness. (iv) Selection of routes for stem cell administration: stem cells can migrate to the site of injury in hypoxic–ischemic encephalopathy (HIE) animal models with systematic or local injection (65,85,100,169). Local delivery of multipotent progenitor cells into the brain can increase the amount of stem cells and improve the motor deficits in the rat model of HIE (170), but no significant differences were found between the results of stem cell transplants made intraparenchymally or intravenously (28,169). Moreover, as the damaged regions in the CNS of patients with severe CP are extensive, it is impractical to perform intraparenchymal stem cell injection. Therefore, intravenous infusion is a relatively safe method of administration. Stem cells with intrathecal injection are expected to migrate to the CNS through CSF, resulting in a direct repairing or restoration of the damaged brain tissue by replication of cells or by the secretion of cytokines to stimulate the healing of these injured cells, and recover motor functions (20,31,32). A report showed that intrathecally injected autologous MSCs significantly improved a CP patient's motor, sensory, cognitive, speech, bowel, and bladder control functions (105). Another advantage of lumbar puncture is multiple dosing of stem cells to increase therapeutic effect. Other approaches such as intraventricular injection (76) and nasal administration are reported to have good results for stem cell delivery to the brain (151). However, the efficacies of stem cell therapy through these routes should be optimized and standardized. (v) Selection of timing: Several reports have indicated that NSCs or ESC-derived neural precursor cells transplanted intraparenchymally or intravenously 24 h postischemia can survive, migrate, and differentiate in an experimental rodent focal ischemia model (28,48,100,172). An experiment showed that the endogenous cell death after hypoxia insult is prominent in the first day, and the endogenous repair mechanisms are initiated by day 2 (150). A study using NSC transplantation showed that engraftment of stem cells was most exuberant 3–7 days following hypoxic injury (71), suggesting the first few days after the hypoxic injury be the golden period to conduct the stem cell therapy. However, an animal study showed that transplantation of murine MSCs to the ipsilateral hemisphere of the P9 mouse at day 3 post-HI resulted in not only functional improvements but also increased cell proliferation and differentiation at 21 days after the insult (150). Also, a study comparing the effects of single and repeated MSC treatment indicated that the first MSC injection on day 3 stimulated the growth or survival of new neurons and oligodendrocytes, while the second injection at day 10 further enhanced sensorimotor function improvement (152), implying that intervention at different time points may cause various effects. (vi) Conditions of recipient: age, may affect the effect of stem cell treatment. Neurotrophic factors, such as glial cell-derived neurotrophic factor, neurotrophin-3, brain-derived neurotrophic factor, vascular endothelial growth factor, and neural growth factor (134,146,153), can promote synaptogenesis and activate silent synapses (59,78), leading to early and prominent functional recovery. Silent synapses will gradually decrease with age, which may partially explain the better plasticity and capability in juvenile brains than adult ones (53). (vii) Acupuncture: acupuncture is an ancient Chinese medical practice using needles inserted along meridians in the body. Acupuncture is reported to promote cognition, brain proliferation and differentiation, behavior, regeneration of nerve fibers, and partial functional recovery in animal neurological disease models, and these effects may, at least in part, be associated with progenitor cell proliferation or stem cell mobilization and neuronal differentiation of proliferated surviving cells in the CNS (27,44,75,168), suggesting that a combination of acupuncture with conventional therapies may have beneficial effects in children with CP (139,174).
Clinical Trials
The clinical benefits of stem cell therapy for patients with CP are currently under investigation. Of the more than 500 clinical trials investigating treatments for CP, only 17 involve stem cells, and seven trials were completed. There are five phase I, four phase II, three phase I/II, one phase III, and four unknown phase clinical trials (Table 1). The Republic of Korea has six trials. Although some of them are completed, protocols and further results regarding these trials are not available. A trial using autologous UCBCs, which were administrated into two children with spastic diplegic CP, proved that the stem cell therapy with autologous UCBCs is safe and feasible and showed significant improvement in the motor functions of children with CP (98). The next step is to validate the storage period in which the clinical use of cryopreserved CB is effective and safe. A pilot nonrandomized clinical study at Duke University is assessing the safety of autologous UCBMNCs in newborn infants with HIE (clinicaltrials.gov identifier NCT00593242) and evaluating the efficiency of functional recovery of viable stem cells from 15-year-old cryopreserved CB infused in the first 2 weeks of life (17). A branch project verified the safety and feasibility of thawed and washed autologous CB infusions in 140 pediatric CP patients who had their CB banked at birth (136). The same group has initiated a prospective, randomized, placebo-controlled study of autologous CB infusions in older children (ages 2–10 years) with CP (clinicaltrials.gov identifier NCT01988584), and the results may be promising. Also, an open label study in India (125) enrolled 32 patients with autism receiving autologous BMMNCs via intrathecal transplantation. In a follow-up at 26 months (mean 12.7), 29 (91%) patients showed significant improvement in social relationships and reciprocity, emotional responsiveness, speech, language and communication, behavior patterns, sensory aspects, and cognition. Twenty patients (62%) showed decreased symptom severity. Few adverse events including hyperactivity (18.7%), seizures (9%), and procedural-related side effects (headache, nausea, vomiting, and pain) were reported. A clinical trial regarding CP in India (124) enrolled 20 patients with an average age of 8.6 years. Results showed that 75% reported improvement in muscle tone, 55% showed significant improvement in head and leg movements, and 50% improvement in speech after autologous BMMNC transplantation at15 ± 1 months follow-up. The encouraging results of this pioneering clinical study support that autologous BMMNC transplantation may be an effective potential option for the treatment of CP. Min et al. (87) conducted a double-blind randomized trial clinical trial (clinicaltrials.gov identifier NCT01193660) to assess the effects of allogeneic cord blood (UCB) transplantation with EPO on children with CP. This study enrolled 105 participants of both sexes aged 10 months to 10 years for a duration of 6 months after trial launch. These participants were divided into three groups, including group 1: intravenous allogeneic UCB infusion, EPO injection, and rehabilitation; group 2: EPO injection and rehabilitation; group 3: rehabilitation. The neurological functions of these patients were evaluated by brain MRI and PET. These were no differences in the frequencies of adverse events among the groups, but the trial demonstrated that the stem cell therapy combined with EPO and rehabilitation-treated group significantly improved CP patients' motor and cognitive function, especially in the younger children with CP, suggesting that combinatorial strategies, as opposed to a single therapeutic approach, can provide the optimal functional recovery. Additionally, the study needs to address whether better HLA matching, total nucleated cell number, and CD34+ cell number directly correlated with the better outcome in group 1. China is going to conduct the first phase III clinical trial (clinicaltrials.gov identifier NCT01929434) enrolling 300 CP patients, including 100 for UCMSC injection, 100 for rehabilitation, and 100 for the control group. Participants are generally of either sex, aged 6 months to 75 years, lack seizures or comorbidities, and have CP of any etiology. A randomized trial (clinicaltrials.gov identifier NCT01763255), which enrolled eight patients with CP, used multiple dosing intrathecal injection BM-derived CD133 cells with a goal to have a significant repair and regeneration after the therapy, but further results regarding the trial are unavailable. NSCs, whether intravenously, intrathecally, or intraparenchymally injected, are suggested to have potential benefits in clinical neurodegeneration. Table 1 shows in the last 2 years, there are seven trials that were registered/launched in Korea, India, Mexico, US, and China. The most interesting thing is that all allogeneic trials use cord blood.
Conclusion
A multidisciplinary team, including those who focus on social and emotional development, communication, education, nutrition, and mobility, targeted at improving function and reducing disability may be helpful in ameliorating the complex neurological deficits of patients with CP, but these therapies have not achieved optimal outcomes. Recent advances in regenerative medicine suggest that stem cell therapy may provide a potential complete cure to patients with CP. Although posttransplantation follow-up shows that most children with CP continue to improve with regard to muscle tone, motor function, limb coordination, and speech, some patients suffer from adverse effects, such as neuropathic pain (66), encephalitis (66), infections (130), multiple brain tumor formation (5), and fatalities (165) after stem cell transplantation. Therefore, apart from the safety, further research should standardize the stem cell therapeutic protocol. The parameters, including the type of cells, route of delivery, cell dose, and timing of transplantation, should be determined to achieve an optimal outcome.
Footnotes
Acknowledgments
This study was funded by Ministry of Science and Technology, Taiwan (MOST103–2320-B-039–021-MY3), Ministry of Economic Affairs (102-EC-17-A-19-I1–0051), Taiwan Ministry of Health and Welfare, China Medical University Hospital Cancer Research Center of Excellence (MOHW103-TD-B-111–03), Clinical Trial and Research Center of Excellence (MOHW103-TDU-B-212–113002) and CMU under the Aim for Top University Plan of the Ministry of Education, Taiwan, and TSGH-C104–021. The authors declare no conflicts of interest.