Cell-based advanced therapy is a novel and extremely promising option to cure otherwise untreatable neurodegenerative diseases but there are still several bottle necks that slow the progression and the widening of clinical trials in this context. Mainly, despite the overall positive results reported in several small studies in PD and MSA, there is a lot of uncertainness regarding the real capacity of MSC to reach the brain, the selectivity of their targets and their real mechanism of action. Moreover, in PD as well as in atypical parkinsonisms no validated biomarker is available to follow-up the neuroprotective effect of putative disease-modifyng treatments. So, why are we performing this trial? First: there are several positive results in other no-option parkinsonisms as well as in PD that deserve to be confirmed also in other pathologic context like PSP. Therefore, even thought the primary objective of this protocol is to demonstrate the safety of autologous MSC intra-arterial administration in subjects affected by PSP, we have designed a sham-controlled study, with the aim to document efficacy as well. Second, we would like to indirectly get information on the putative neuroprotective mechanism of action of MSC in PSP by comparing the in vitro results of the proposed potency assay and the clinical outcome. The biological hypothesis underlying the rationale of the proposed protocol is that MSC can reduce the neural cell loss in PSP by reducing cell apoptosis and the detrimental consequence of oxidative stress on neural cell homeostasis. Whit this premises, it should be clear to the reader that, the objective of making neurons from stem cells is out of the scope of our approach. Nevertheless, the story of replacing lost neurons in parkinsonisms by transplantation has been pursued for years. At first tissue was transplanted. In the 1980s autologous adrenal gland medullary tissue was transplanted first in animals and then in humans. Significant clinical improvements were achieved, but common and even serious postoperative complications induced clinicians to abandon this option [39–44]. In the 1990s non-autologous fetal neuron transplantation was attempted. Clinical benefits were modest and confined to younger patients (<60 years), and complications occurred i.e. the appearance of postoperative dyskinesias that in a few cases were severe and refractory to treatment. PET investigations suggested that these cases were due to an imbalance in the dopaminergic circuits induced by the transplanted cells [45, 46]. Recent in vivo brain imaging findings in two patients, who had exhibited major motor recovery after transplantation, suggested that the cause of the dyskinesias was a serotonergic hyperinnervation . An alternative explanation is that an immune response was responsible for both the appearance of dyskinesias and the modest extent of the therapeutic benefits . Furthermore, the autopsies of subjects who had received fetal mesencephalic dopaminergic neurons disclosed the presence of Lewy bodies, raising the possibility of host-to-graft disease propagation. Thus, the prospects for dopaminergic fetal neuron transplantation in PD do not appear to be promising . Autologous stem cell transplantation has been proposed for the most common form of parkinsonism, PD . This proposal overcomes the ethical reservations related to the use of fetal cells or embryonic stem cells, and removes the risk of transplant rejection. Even thought the ability of different bone-marrow derived stem cells to migrate into the brain and to differentiate into cells bearing neural markers have been postulated even in vivo [51–53], the proven evidence of the so-called neural plasticity have never been demonstrated. Indeed, in the last years the attention of most investigators has been addressed to study the potential of BM mesenchymal stem cells to positively influence neural cell survival and to reduce cell apoptosis [54–56]. MSC have proved to be able to produce different kinds of growth factors that increase neuronal survival, to possess immunoregulatory properties able to influence inflammatory conditions and to migrate to damaged tissue areas where they might contribute to counteract neurodegeneration. What is more, they can be easily harvested from the bone marrow of the patients, easily expanded on a large scale for autotransplantation, and administered to patients via various routes . MSCs have been isolated from PD patients and they do not differ from those of healthy subjects in terms of phenotype, morphology and ability to differentiate . Furthermore, it has been demonstrated that MSCs exert neuroprotective effects on dopaminergic neurons both in vitro and in animal models of PD (rats given the toxin MG-132 or 6-OHDA), mediated by anti-inflammatory activity and the secretion of growth factors [59–62]. Of particular interest is the fact that in the rats given MG-132, a proteasome inhibitor, MSCs also reduced the accumulation of polyubiquinated proteins, a finding that suggests that they also contribute to correct proteasome dysfunction, which is believed to play an important role in the pathogenesis of PD .
Autologous MSC therapy would therefore appear to be an important candidate for the development of a parkinsonism-modifying therapeutic strategy.
To our knowledge, four pilot clinical trials have been published on the use of autologous stem cell transplantation in patients with primary parkinsonism. Two studies [63, 64] have been conducted in patients with multiple system atrophy (MSA). The first one was an open-label, controlled trial designed to assess the feasibility and safety of MSC therapy by intra-arterial and intravenous route in which eleven MSA patients were infused MSC therapy through the internal carotid artery and the proximal portion of the vertebral artery once and by intravenous route thereafter once a month for three months. Treated patients were compared to 18 untreated control MSA patients. Follow-up was continued up to one year after the beginning of treatment and consisted in neurological examinations using the unified MSA rating scale scores (UMSARS). In addition, 5 treated patients and 10 untreated patients underwent brain metabolism imaging using PET and 18 F-fluoro-deoxyglucose (FDG). After 12 months mean total UMSARS score was similar to the score at baseline in the treated group (functional stabilization), whereas it had worsened in the control group (p = 0.002). Moreover, treated patients showed increased brain metabolism by means of FDG-PET, while the untreated group had decreased uptake. The most significant adverse events reported was the occurrence in 7 patients of small spotty lesions on MR images that were considered to be asymptomatic microemboli, a frequent complication of catheterization techniques. The same group recently published the results of a second protocol, in which thirty-three patients with probable MSA-C were randomly assigned to receive MSC via intra-arterial and intravenous routes or placebo. The primary outcome was change in the total UMSARS scores from baseline throughout a one-year follow-up period between groups. In this study, the MSC group had a smaller increase in total and part II UMSARS scores compared with the placebo group. Cerebral glucose metabolism and gray matter density were more extensively decreased in the cerebellum and the cerebral cortical areas, along with greater deterioration of frontal cognition in the placebo group compared with the MSC group. No serious adverse events directly related to MSC treatment were recorded. However, intra-arterial infusion resulted again in small ischemic lesions on MR. In another open-label clinical trial , autologous BM MSCs were transplanted into a sublateral ventricular zone of the brain by stereotaxic surgery in 7 male patients aged 22 to 62 years with advanced PD. The diagnosis was based on the presence of the classical symptoms and a good response to levodopa. The patients were followed up for 10 to 36 months. Three of the patients experienced motor improvement compared to baseline; the mean extent of improvement in UPDRS score was 22.9% in OFF and 38% in ON. Two of the patients were able to reduce the dosage of their antiparkinson medications. No serious adverse events occurred. MRI imaging did not disclose any significant changes. The last study  was an open-label clinical trial conducted by interventional radiologists, who subjected 53 patients with a diagnosis of PD made according to UK Brain Bank criteria to intraarterial autologous implantation of mononuclear cells from bone marrow. The cells were introduced by intraarterial catheterization, infusing the cells into the posterior part of the circle of Willis, from which the perforating arteries that irrigate the basal nucleus and the substantia nigra originate. Four patients received a second implant. None of the patients had major complications. They experienced major significant changes in median disease severity scores: UPDRS, Hoehn & Yahr, Schwab & England and Northwestern University Disability Scale (NUDS). In eight patients follow-up MR spectroscopy revealed mean improvements in n-acetylaspartate/creatine ratio. These studies demonstrated the feasibility of autologous cell transplantation in patients with parkinsonism, but they did not demonstrate efficacy, because their design did not ensure objective measurements. Also, clinical improvement could have been influenced by the beliefs of patients and investigators alike as they were open-label studies. Neuroimaging changes could not be easily dismissed, but they were available only in a few patients and did not focus on the kind of damage that is seen in parkinsonism. A search in the WHO worldwide clinical trial database on August 1, 2013 disclosed 390 studies assessing MSC therapy. Out of these only other three studies are ongoing in patients with Parkinson. To our knowledge this is the first time that autologous MSC therapy is given to PSP patients.
In this kind of cell therapy protocols, a crucial question is how to fix the optimal cell dose to be administered to the patients and which is the best administration route, considering together the safety aspects and the efficacy objectives. When the protocol was conceived, the dose to be given was established based on several considerations. In particular, although there were data showing that MSCs in PD patients do not differ from those of healthy subjects in terms of phenotype, morphology and ability to differentiate into other cells , a preliminary study was conducted in PSP patients, to check the bone marrow function and MSC yield and to establish how many cells could be realistically produced for administration. The elements that have been used for the definition of the cell dose were the quaantity of MSCs that was obtained from maximum 30 mL of bone marrow and the cell dose that was given in the three protocols that were previously performed with MSC in parkinsonisms (1-2×106/kg). Thus, the cell dose was fixed at 1.5 ± 0.5 ×106/kg. Regarding the route of administration, several possibilities were considered. Systemically injected MSC undergo intra-pulmonary cell trapping  and therefore only a limited amount of cells might home to the brain. Stereotaxic-guided intra-striatal implantation was excluded in consideration of the pathologic characteristics of PSP that has a much wider distribution compared to classic Parkinson disease and for safety concerns, since there are several evidences that intra-striatal cell administration may cause harm and reduce efficacy by evoking a local cellular immune response . Therefore, we decided to use superselective arterial catheterization to implant stem cells throw the arteries that feed the brain regions affected by PSP to release in situ the highest concentration of MSC. This technique has been used in the three out of four previous clinical trial in parkinsonisms, with no major adverse events out of the report of asymptomatic microembolism. To minimize this risk, several precautions have been taken in our protocol during the preparation of the cell product such as dilution of the cells to less than 1 × 10E6 cells/mL and the addition of an anticoagulant (ACD-A) to the solution in which the cells are re-suspended before infusion.
In this study a “pure” control group treated with placebo solution administered by the same route in a double-blinded manner is missing since this option was considered not acceptable by the ethical point of view by the investigators, given the potential harm of a cerebral artery catheterization procedure that is not justified by a potential benefit. Nevertheless, the patients enrolled in the control harm receive a simulated arterial catheterization and administration procedure and the neurologist and the neuro-radiologist who perform the follow-up evaluation are blinded. In this way, we exclude the placebo effect with the best precision.
In conclusions, the presented protocol is the first attempt to understand if MSC can be safely administered and can exert a beneficial effect in PSP patients. We believe that the results of this trial will help to improve the knowledge around the neuroprotective properties of MSC that might be exploited in other several neurodegenerative disorders.