The potential of cell therapies for cerebral palsy: where are we today?

KEYWORDS:

• Cerebral palsy
• efficacy
• inflammation
• mesenchymal stem cells
• safety
• stem cells
• umbilical cord blood

1. Introduction

There are still no approved cell therapies for cerebral palsy (CP) in any global jurisdiction despite 15 years of human research and over 2400 children treated in clinical studies [1]. Meanwhile, families continue to purchase therapies from stem cell tourism operators and clinical trials go on. Pediatric research is fraught with inherent ethical complexities and small sample sizes. Moreover, the stem cell field faces additional barriers of cost, manufacturing scalability, and immature regulatory approval pathways, all of which further hamper progress. This article will critically evaluate whether there is hope beyond the hype.

CP is the most common physical disability of childhood, with a declining incidence of 1.6/1000 (95%CI 1.5–1.7) in high-income countries, but a higher burden of disease 3.4/1000 (95%CI 3.0–3.9) in low-and middle-income countries [2]. CP has no cure, despite over 180 different interventions in existence and great progress in the preventative sphere with a 30% reduction in incidence [3]. CP has multiple etiologies, including heritable causes. For this reason, cell therapy for people with CP is not one-size-fits-all. Knowledge is accumulating about the different cell types, doses, timing of treatment, and routes of administration that might be helpful for various sub-populations.

2. Cell therapy in CP

2.1. What is cell therapy for CP?

At a basic level, stem cells are cells that self-renew and differentiate i.e. specialize [4]. In their more specialized form (e.g. tissue stem cells) they are known as adult stem cells. There are many cell types under investigation for CP, predominantly using adult stem cells. Here we will focus on the four best-known cell types for CP: (1) Umbilical Cord Blood (UCB), a mixed cell population (including Mesenchymal Stem Cells (MSCs) and Hematopoietic Stem Cells (HSCs)) drawn from the umbilical cord after birth, which can be frozen and stored in a blood bank for later infusion. UCB can be autologous (child’s own cord) or allogeneic (sibling or unrelated donor). Autologous, sibling and unrelated donor cells have all been infused for the treatment of CP. Unrelated donor cords have the highest clinical convenience given there is no pre-birth test for CP to trigger autologous cord storage. The major drawback of UCB is that cords are dose-limited and doses vary between cords. An infant may be able to have two infusions from a single cord, provided the blood is stored across multiple bags since refreezing destroys cell viability. However long-term repeat dosing in children of larger body weight introduces the complexity of infusing multiple donor cord units, for which safety is less understood; (2) MSCs are a stromal cell type derived from a variety of sources including adipose and cord tissue. These cells can be commercially manufactured at scale and stored ‘off the shelf’ for rapid infusion in acute indications. MSCs are ‘immune privileged’ meaning donor cells can be administered to patients without matching or immunosuppression, thus producing high clinical convenience; (3) HSCs include mobilized peripheral blood cells, bone marrow mononuclear cells and expanded hematopoietic UCB cells [5]. Bone marrow cells are harvested from either a matched donor or the patient’s own bone marrow following hyper-stimulation. In CP, the child’s own bone marrow has been reinfused, which protects against rejection, but harvesting is painful and presents a serious drawback for children. Infants do not have a fully developed immune system, thus, early marrow harvesting followed by reinfusion is uncommon and tends to occur later in childhood; (4) Neural Stem Cells (NSCs) are typically sourced from embryonic stem cells or fetal brain tissue and give rise to all cell types found in the brain, including astrocytes, oligodendrocytes, and neurons.

2.2. How might it work?

Stem cells have three basic mechanisms of action: anti-inflammatory, replacement of damaged cells, and trophic effects promoting endogenous salvage and replenishment of cells [6]. Different cell types have different properties, and some cells have more than one action, with all mechanisms of action thought to be helpful for CP. Whilst the current definition of CP states that the brain injury is static, we identified that children with CP experience prolonged inflammation extending for years, suggesting an under-recognized and paradigm-shifting treatment target [7]. UCB, MSCs and HSCs all have anti-inflammatory properties and produce immunomodulatory effects. In the chronic brain injury phase, these cells produce a small treatment effect, but preclinical studies suggest these effects might be even greater if applied early. Only NSCs can strictly replace injured brain cells. Thus, NSCs offer the greatest scientific plausibility of brain repair and the possibility of a cure.

2.3. How well does it work?

Safety: There is considerable evidence that adult cells administered systemically, particularly from autologous sources, are safe for use in humans, including children with CP [4]. Various systematic reviews have demonstrated safety of several cell types for CP. Recently, we showed that allogeneic UCB has a good safety profile, without immunosuppression, paving the way for repeat dose therapy in adults and children with neurological conditions [8]. For NSCs, the obligatory neurosurgical delivery route and requirement for immunosuppression poses added safety considerations. However, from a small number of participants treated NSCs appear safe. More research is needed.

Efficacy: Meta-analyses indicate that cell therapy paired with rehabilitation confers greater gains in gross motor skills, than rehabilitation alone, on the CP gold-standard Gross Motor Function Measure (GMFM) (SMD 1.27; 95% CI 0.22, 2.33) [9]. UCB has the largest evidence base for CP, with over 22 trials treating 700+ children, including a number of high-quality Phase II trials. What is missing is a Phase III clinical trial to meet the regulatory threshold for an approved therapy. UCB is a ‘green light’ intervention when coded on the Evidence Alert Traffic Light System, which means it works, with a favorable balance of benefits over harms [3]. There is accumulating evidence that MSCs (1000+ treated) and HSCs cells (400+ treated) also have efficacy [1]. These are coded ‘yellow light’ ‘probably do it’ therapies, as more research would increase our confidence in the estimate of the effect [3].

2.4. What is involved?

The most efficacious stem cell therapies for fatal diseases such as blood cancers and multiple sclerosis, involve ablating the patient’s immune system prior to cell transplantation. Ablation mitigates against transplantation rejection and optimizes conditions for cell engraftment to enable resetting the immune system, akin to a forced reboot of the operating system. In stark contrast, children with CP have a healthy immune system and thus most cell therapies are an intravenous transfusion, not a transplant. Engraftment is not sought. Rather it is assumed that a transient treatment effect occurs, while the child’s immune system is fully clearing the transfused cells. This is why repeat doses may be required with UCB, MSCs and HSCs therapies. In contrast, with the goal of cell replacement, NSCs engraft and thus are a single-dose therapy transplant. NSCs must be administered neurosurgically, directly adjacent to the brain injury site within healthy tissue, as they do not migrate without brain structures, nor do they cross the blood-brain barrier. This adds additional complexity and cost, and consequently, NSCs for CP have been less researched.

2.5. Where is it done?

Cell therapies for CP are currently hospital-based procedures, typically led by a hematologist, with expertise in both clinical management of cell transplants and adverse events, plus proficiency in cell release quality control that is compliant with international regulatory procedures. Cellular infusions may also be performed in appropriate outpatient settings.

2.6. Who is it for?

Research to date has largely included all subtypes and ages of children with CP [10,11]. This is both a strength and a weakness of the evidence base. On the one hand, cells appear safe for all types of CP, but it is not yet possible to tease apart best-responders from non-responders. Timing is another key issue; short-acting transient cells (such as MSCs) are likely to have higher efficacy in the acute phase of brain injury, whereas engrafting replacement cells (such as NSCs) will plausibly work even in adulthood. A recent clinical trial that compared MSCs to UCB head-to-head in children with CP, found UCB produced superior results [12]. Results suggest that UCB might be helpful in both the acute and chronic phases of brain injury acting on prolonged inflammation. Thus, UCB might be helpful for all subtypes of CP at all ages. Whereas MSCs may be more efficacious in the acute injury phase and are thus being explored for a number of CP causal pathways including neonatal stroke and acute hypoxic ischemic encephalopathy.

2.7. How much is needed?

Extensive human safety data on adult cells indicate humans tolerate millions of cells being transfused [4]. Logically the CP field commenced cell therapy at low doses. Subsequently, the threshold dose needed to produce a therapeutic effect from UCBs and MSCs has been calculated from clinical trial data. These doses are not interchangeable between cell types, given the cells vary in their mechanisms of action and systemic uptake. Future clinical trials are likely to commence therapy at the known threshold doses and escalate upwards to explore dose-response effects.

3. Expert opinion

Cell therapies are safe and show promising efficacy for children with CP to improve their motor skills. A number of high-quality clinical trials have been completed but the weaknesses of the evidence base include: heterogeneous patient selection creating wide confidence intervals; inclusion of broad age ranges confounding data on the optimal timing of administration; varying dosage paraments; and uncontrolled concurrent rehabilitation confounding effects. Moreover, children with CP have multiple comorbidities, and families want gains beyond the motor domain. It is plausible that cells might deliver broader benefits given their anti-inflammatory actions, but this has not yet been measured rigorously using the right instruments [13]. Clinical trials have been painfully slow, allowing unregulated stem cell tourism to bourgeon. Reasons include: (1) the costs of cells and manufacturing often exceed maximum allowable grant budgets; (2) inappropriate blanket application of transplant safety protocols to transfusions causing risk aversion; (3) misaligned regulatory processes and differing risk appetites across jurisdictions hampering multicentre global collaborations; (4) an ill-equipped workforce, where clinicians, researchers, and reviewers have limited experience in cell therapies, onsite storage and cell release, and the catalysts needed to take the field to scale; and (5) limited robust manufacturing infrastructure for replicability and scalability as patients require millions of cells per treatment. Despite these challenges, the field still does have hope. The ultimate goal is to find a cell therapy that either partially or fully alleviates the symptoms of CP and the underlying brain injury. In the coming years, we can expect a Phase III UCB clinical trial to meet the necessary regulatory threshold for an approved treatment; an individual patient data meta-analysis to tease apart best-responders from non-responders; and Phase I and II safety and efficacy clinical trials in newborns.

Declaration of interest

I Novak, MC Paton, M Jackman hold funding from the National Health and Medical Research Council (NHMRC) and Medical Research Futures Fund (MRFF) of Australia as well as other philanthropic grant funding from the Cerebral Palsy Alliance. AR Griffin and M Finch-Edmondson hold philanthropic grant funding from the Cerebral Palsy Alliance. I Novak and M Finch-Edmondson hold funding from the Horizons 20:20 European Union. In the past, the authors have also been supported by the Ramsay Foundation and Advance Queensland as well as via IMPACT for Cerebral Palsy Macquarie Group grant funding. None of the work discussed in this article pertains to the NHMRC, MRFF, European Union or Cerebral Palsy Alliance funded research. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. This includes consultancies, honoraria, stock ownership or options, or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This manuscript has not been funded.