Novel Approach to Stem Cell Therapy in Parkinson's Disease

Cell Type Used in Clinical Trial: Rationale

The available pharmacological therapies in Parkinson's disease (PD) do not rescue the nigrostriatal system, do not deliver dopaminergic (DA) specifically to the brain regions needed, do not mimic the normal release of DA, lose efficacy over time, and are often limited by distressing side effects, including hallucinations and troublesome dyskinesia. Cell replacement therapies are a promising avenue for treatment of PD and other neurodegenerative disorders. The first studies investigating the effects of human fetal ventral mesencephalic (hfVM) tissue in PD patients in the late 1980s to early 1990s yielded promising results [1 –3]. However this approach demonstrated some limitations. For instance, a number of subjects were found to suffer from graft-induced dyskinesias (GID) [4,5].

Following the early encouraging reports of fetal transplants, other potential avenues have been explored. These include the differentiated derivatives from human pluripotent stem cell sources such as human embryonic stem cells (hESCs), induced pluripotent stem cells (iPSCs), and human parthenogenetic stem cells (hpSCs). Human pluripotent stem cells have the ability to proliferate indefinitely and differentiate into all three lineages, providing potential advantages and limitations.

We are transplanting ISC-hpNSC, which are neural stem cells (NSCs) derived from hpSCs. hpSCs are derived from the chemical activation of unfertilized oocytes [6,7]. Differentiated derivatives from parthenogenetic stem cells have been found to be safe and effective in various disease models by multiple laboratories around the world [8 –13]. In 2014 the US Food and Drug Administration (FDA) cleared the hpSC line used to derive ISC-hpNSC for investigational clinical use. hpSCs have a number of potential advantages compared to other sources of cellular replacement. hpSCs bypass the ethical concerns associated with hfVM tissue or hESCs because no fetus or viable embryo is used in their derivation [6,7].

The hpSCs used in this study were derived from a consenting young donor's oocyte. Derivation of iPSCs from older donors may pose the additional risk of acquiring cancer causing mutations [14,15]. HLA-matched iPSCs do not necessarily confer improvements in DA neuron graft survival (see Fig. 8n–p in [16]). Parthenogenetically derived stem cells may have other advantages: a recent study published in Cell Stem Cell showed that hpSCs have lower number of de novo coding mutations than iPSCs [17], a potential risk factor for tumorigenicity. Overall, hpSCs have comparable therapeutic potential and risks to other pluripotent stem cells. In this study we explain why we chose to differentiate hpSCs into NSCs to treat PD.

NSCs are self-renewing multipotent cells that generate the neurons and glia of the nervous system. NSCs were shown to be effective in treating PD models over a decade ago by Snyder, Redmond and colleagues [18 –22]. Neural stem/progenitor cells have also been investigated in PD by many other laboratories [23 –35]. In contrast to DA neurons, ISC-hpNSC have multiple mechanisms of action, including neurotrophic support, neuroregeneration, and immunomodulation [36]. ISC-hpNSC secrete various neurotrophic factors such as brain-derived neurotrophic factor (BDNF) and glial-derived neurotrophic factor (GDNF), which have been shown to increase the survival of DA neurons [37 –40]. ISC-hpNSC also increase the concentration of these neurotrophic factors in vivo [39].

ISC-hpNSC have been shown to efficiently differentiate in vitro into DA neurons that release dopamine, fire spontaneous action potentials, and express DA markers [41]. Together with our scientific collaborators, we have been able to demonstrate that ISC-hpNSC engraft and differentiate into tyrosine hydroxylase⁺ (TH⁺) DA neurons in rodent and nonhuman primate models of PD [13,39]. Although between 1% and 2% of the engrafted ISC-hpNSC differentiate into TH⁺ DA neurons in vivo, sufficient numbers can be implanted to allow improvements in motor symptoms. Assuming a 10% survival of the implanted cells, patients receiving 30–70 million ISC-hpNSC would have approximately 30,000–140,000 grafted TH⁺ DA neurons. Postmortem brain analyses from PD patients who received hfVM tissue have estimated that only 30,000–100,000 grafted TH⁺ DA neurons are necessary for long-term symptomatic relief [42 –45].

Most of the recovery observed in striatal DA neuron innervation is host derived and is due to the neuroprotection and neurotrophic support provided by ISC-hpNSC, where the majority of the engrafted cells remain as NSCs that undergo growth arrest and become quiescent and a small percentage differentiates into glia and neurons [13,39]. There were no signs of uncontrolled proliferation in any of the ISC-hpNSC transplanted animals.

The potential immunogenicity of the engrafted ISC-hpNSC may be low. Human parthenogenetic derived NSCs express HLA-G and show unique resistance to NK cell-mediated killing [46]. The binding of HLA-G to its receptors leads to the destruction of T and NK cells [47]. In general, graft failure in stem cell transplantation increases with increasing numbers of mismatches in the Host versus Graft direction. However, autologous or HLA-matched transplantation may not be strictly necessary based on observations that hfVM grafts, derived from multiple tissue sources, can show stable integration and long-term persistence—confirmed on functional imaging, as well as postmortem analyses—without use of concomitant immunosuppression [5,48].

The long-term survival of the grafts might be attributed to the fact that hfVM cells express HLA-G for immune tolerance during pregnancy [49]. Placental HLA-G proteins facilitate semiallogeneic pregnancy by inhibiting maternal immune responses to foreign antigens [50]. NSCs can further decrease neuroinflammation by reducing the expression of tumor necrosis factor-α⁺ (TNF-α⁺) and major histocompatibility complex II⁺ (MHC II⁺) activated inflammatory cells, which have been shown to be effective in multiple neurological disease models [51 –54].

Other potential mechanisms of action for NSCs include the observation that RNAseq analysis of brain tissues showed that transplantation of ISC-hpNSC in nonhuman primates induces the expression of genes and pathways downregulated in PD [13]. A recent study also showed that human neural progenitor cells reduce α-synuclein oligomers and rescue cognitive and motor deficits in a Dementia with Lewy bodies mouse model [55]. Normalization of α-synuclein aggregation was also observed after human NSC transplantation in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-lesioned nonhuman primates [18].

Transplantation of neural precursor cells induces the proliferation, neurogenesis, and migration to the graft site of host subventricular zone neural precursors that lead to the significant preservation of striatal TH expression and substantia nigra TH cell number in 6-hydroxydopamine (6-OHDA)-lesioned rats [56]. NSCs harness endogenous repair mechanisms to promote tissue regeneration and have the intrinsic capacity to rescue dysfunctional DA neurons in the brains of aged mice [22,57]. The combined actions of neurotrophic support, neuroregeneration, immunomodulation, downregulation of PD associated genes, and recruitment of host neural precursors promote behavioral recovery and increase striatal DA concentration, striatal DA neuron fiber innervation, and nigral DA neuron number after ISC-hpNSC transplantation in MPTP-lesioned nonhuman primates [13].

Comparison with DA Neuron Progenitors

In the preclinical program, the efficacy of NSCs has been compared to DA neuron progenitors in both rodent and nonhuman primate PD models. We observed that animals transplanted with DA neuron progenitors had lower levels of striatal DA and lower behavioral recovery than animals transplanted with NSCs. Animals transplanted with DA neuron progenitors had profuse host inflammatory factor 1⁺ (Iba 1⁺) microglia and astrogliosis surrounding the graft, consistent with other reports [16,58]. It has been shown that NSCs, but not neurons, suppress neuroinflammation [59,60].

While DA replacement strategies are effective in ameliorating dopamine responsive symptoms early in the disease process, more advanced stages of PD are associated with the development of dopamine-resistant symptoms such as gait dysfunction, freezing, falling, and dementia [61 –66]. Based on the importance of neurotrophic support and neuroinflammation reduction in the treatment of PD, NSCs have theoretical advantages over DA neuron progenitors in ameliorating the non-DA pathology characterizing advanced disease states [40,63,67,68]. Consistent with this, it has been reported that transplanted NSCs can rescue cognitive dysfunction in a transgenic mouse model of dementia with Lewy Bodies through a BDNF-dependent mechanism [40]. Furthermore, BDNF was shown to mediate antidepressant efficacy in BDNF conditional knockout mice studies [69,70].

Some challenges remain with purely dopamine cell replacement strategies for PD. Stem cell derived DA neuron replacement therapies will have to compete with established device aided therapies that can provide effective continuous DA treatments with relative safety (eg, levodopa-carbidopa intestinal gel [LCIG]) or with neuromodulation therapies (eg, deep brain stimulation). Moreover, it is also worth noting that none of the stem cell derived DA neuron replacement approaches are moving closer the clinical use of mature DA neurons, but only DA neuron progenitors [71]. Delivery and dosing of DA neuron progenitors will provide logistical challenges. Because DA neuron progenitors migrate less, precise localization of the implants will be critical. The smaller numbers of implanted cells will require the development of novel delivery systems.

Precise dopamine dosing could also be difficult—a recent article by Parmar and colleagues demonstrated that only a small percentage of the transplanted GMP grade DA neuron progenitors differentiated into TH⁺ DA neurons (∼3.7%) in rodents, even though more than 80% of the transplanted cells were of DA origin [72]. Similarly, studies in nonhuman primates by Takahashi and colleagues showed greater than threefold variations in the number of DA neurons surviving (between 4,300 and 13,000 TH⁺ DA neurons survived out of 4.8 × 10⁶ DA neuron progenitors transplanted, or 0.089%–0.27%) [73,74]. Based on these facts we propose that NSCs offer more repair mechanisms than just DA neuron replacement, but ultimately these approaches need to be tested in clinical trials.

Extensive Safety Data Supporting Clinical Trial

We have manufactured our master and working cell banks of ISC-hpNSC under cGMP and performed strict quality control measures to test sterility, purity, identity, potency, and safety before their use in the clinic. Our patented manufacturing protocol allows for the chemically defined production of a homogenous population of NSCs that is expandable and cryopreservable. In comparison, hfVM tissue is an extremely limited resource and not scalable. Therefore, despite significant resourcing from the European Union, the multinational TRANSEURO study of hfVM transplantation for PD recently failed to complete dosing of their planned participants, and only 20 out of a planned 90 or more surgeries have taken place because of tissue supply [71,75].

The clinical grade ISC-hpNSC population expresses neural markers and no pluripotency markers, has a normal karyotype, and is negative for bacteria, fungal, mycoplasma, and adventitious viral contaminants [76]. This is not the case for hfVM tissue, where 47% of samples isolated have microbial contamination and have to undergo decontamination [77]. As part of the release testing of ISC-hpNSC, sensitive analytical methodologies are used, including flow cytometry capable of direct detection of pluripotent stem cells and reverse transcription-polymerase chain reaction (RT-PCR) capable of selective detection of messenger RNA (mRNA) transcripts from pluripotent stem cells. These two approaches are supplemented by cell culture studies that demonstrate the inability of pluripotent stem cells to propagate under cultivation conditions used for the manufacture of ISC-hpNSC. All three methodologies confirm the absence of pluripotent hpSCs in clinical grade ISC-hpNSC [76].

To further mitigate this risk, we conducted a series of in vivo preclinical studies that demonstrate the absence of both residual undifferentiated pluripotent stem cells and tumorigenic transformed cells in the clinical grade ISC-hpNSC. One of these studies was a large-scale 9-month tumorigenicity study in 300 immunodeficient nude rats under good laboratory practice (GLP). Histopathological analysis by a board certified pathologist did not detect the presence of any teratomas or tumorigenic risk in any of the animals injected with ISC-hpNSC. Injection of ISC-hpNSC was not associated with any proliferative risk or other serious adverse event at any dose or time interval, and biodistribution analysis did not detect the presence of ISC-hpNSC outside the central nervous system [76]. No ISC-hpNSC related deaths or adverse clinical observations were observed [76].

A prevalent concern in the PD cell transplantation field is the appearance of GID [64 –66,78,79]. The GID observed in the hfVM studies might have been caused by the nonhomogeneous delivery of dopamine cells across the putamen giving rise to DA “hot spots” [64,65] or the presence of serotonergic neurons in the transplant which could be releasing DA in an unregulated manner [78,80]. In a 12-month GLP pharmacology and toxicology study in MPTP-lesioned nonhuman primates, we did not observe GID in any of the monkeys injected with ISC-hpNSC [13]. The lack of GID could be due to the fact that ISC-hpNSC migrate away from the injection site along the nigrostriatal pathway and do not form compact grafts with potentially contaminating serotonergic 5-hydroxytryptamine (5-HT⁺) cells as hfVM tissues do [5,80]. It is also possible that the lack of l-DOPA priming precluded from appropriately evaluating the risks of GID in this study.

There is extensive evidence of safety and therapeutic potential of NSCs in many other neurological indications. Preclinical studies conducted by many investigators, including Okano, Gage, Borlongan, Cummings, Loring, Thompson, and Svendsen, show that neural stem/progenitor cells are safe and effective in treating spinal cord injury [81,82], stroke [83], traumatic brain injury [84], multiple sclerosis [52], Huntington disease [85], and amyotrophic lateral sclerosis [86]. Furthermore, there are clinical data demonstrating that neural stem/precursor cells are safe in spinal cord injury, stroke, amyotrophic lateral sclerosis, age-related macular degeneration, myelination disorders, and lysosomal storage disorders [87 –91]. Up to 1 billion NSCs were intracranially injected in children with neuronal ceroid lipofuscinosis disease, a fatal lysosomal storage disease [87]. The cells were well tolerated, and there was no evidence of tumors or test article related adverse events [87]. Even conditionally immortalized NSCs have shown no safety concerns after intracerebral transplantation in chronic ischemic stroke patients [92].

Our therapeutic approach is not only based on cell replacement but also on neuroprotection, neurotrophic support, and immunomodulation and has the potential to rescue the nigrostriatal and extranigral systems and significantly improve motor and cognitive functions, as well as quality of life, offering hope to patients worldwide suffering from this devastating disease.