Encapsulated cell biodelivery of GDNF: A novel clinical strategy for neuroprotection and neuroregeneration in Parkinson's disease?

https://doi.org/10.1016/j.expneurol.2007.08.019Get rights and content

Abstract

The main pathology underlying disease symptoms in Parkinson's disease (PD) is a progressive degeneration of nigrostriatal dopamine (DA) neurons. No effective disease-modifying treatment currently exists. Glial cell line-derived neurotrophic factor (GDNF) has neuroprotective and neuroregenerative effects and it enhances dopaminergic function in animal models of PD. These findings raise the possibility that intrastriatal administration of GDNF might be developed into a new clinical strategy for functional preservation and restoration also in PD patients. Gene therapy is a novel tool to increase local levels of GDNF. Transplantation of encapsulated, GDNF-secreting cells is one strategy for ex vivo cell-based gene delivery which has the advantage to allow for removal of the cells if untoward effects occur. Here we summarize studies with such cells in animals, and discuss the results from previous trials with GDNF in PD patients and their implications for the further development of neuroprotective/neuroregenerative therapies. Finally, we describe the different scientific and regulatory issues that need to be addressed in order to reach the clinic and start the first trial in patients.

Introduction

The main pathology in Parkinson's disease (PD) is a gradual loss of dopaminergic neurons in the substantia nigra, leading to a progressive reduction of dopamine (DA) levels in the striatum. Symptoms include tremor, muscle rigidity, bradykinesia and postural instability. The first symptoms appear when about 50% of nigral dopaminergic neurons have died. These symptoms are usually mild and can be effectively treated with l-dopa. However, PD is progressive and the majority of patients show a gradual loss of l-dopa efficacy with “on–off” fluctuations and dyskinesias. In this complication phase, partial symptomatic relief can be accomplished by means of medications, such as DA agonists and COMT and MAO-B inhibitors, or by deep-brain stimulation (DBS). A major therapeutic problem is that no effective disease-modifying strategy has so far been developed.

Neurotrophic factors are essential for neuronal survival and differentiation during development and for maintenance of normal neuronal function in the adult. Glial cell line-derived neurotrophic factor (GDNF) is one of the most potent protective neurotrophic factors for dopaminergic nigral neurons in experimental models of PD (Lin et al., 1993; for references and review see, e.g. Kirik et al., 2004). Delivery of neurotrophic factors such as GDNF to the brain is, however, problematic due to (i) their inability to cross the blood–brain barrier, (ii) brain diffusion and pharmacodynamic issues, and (iii) side-effects associated with binding to extra-target receptors. Intracerebroventricular (ICV) administration of GDNF has been explored as a possibility to bypass the blood–brain barrier. However, the poor diffusion of GDNF from the ventricles into the brain parenchyma and the occurrence of severe side-effects limit the usefulness of this delivery approach in PD patients (Kordower et al., 1999, Nutt et al., 2003). Both animal experiments and human studies indicate that direct intrastriatal administration of GDNF reduces the occurrence of side-effects associated with ICV application, although the degree of symptomatic relief in the clinical trials has varied from major improvements to no significant effects (see below).

The observations made so far underscore the need for the development of new systems for localized, sustained, and safe delivery of GDNF into the striatum in patients with PD. For a clinically valuable neuroprotective action, GDNF probably has to be supplied long-term, perhaps over several years. However, GDNF can work also through other mechanisms, such as induction of neuroregenerative responses in remaining DA neurons, as well as direct stimulatory effects on DA release. It is conceivable that for the neuroregenerative effect, GDNF needs to be released only for a shorter period, such as a few months.

Direct localized CNS delivery is achievable either by pump or by cell-based (ex vivo) or direct (in vivo) gene delivery (Fig. 1). However, efficient intrastriatal delivery of GDNF via pumps may require two or three catheters on each side of the brain connected to pumps placed in the abdomen, making this a cumbersome system prone to complications. Direct gene delivery leads to efficient local synthesis of GDNF but has the problem that there is so far no possibility to completely turn off the protein synthesis in case of adverse effects. The same problem limits the clinical usefulness of “naked” cells, which have been genetically engineered to produce GDNF and are then injected into the brain. The migratory capacity of such cells could lead to efficient delivery of GDNF over the entire target area but there is a risk for untoward effects if the cells also reach other structures. The strategy with “naked” cells also poses issues surrounding allogeneic immune rejection (unless autologous cells are used) and transgene downregulation, thereby limiting the usefulness of this approach for PD.

Our strategy for delivery of GDNF to the striatum in PD patients is to use encapsulated cell biodelivery (ECB). The cell encapsulation aims primarily at providing a physical barrier to prevent immune rejection and direct host contact with the allogeneic, genetically modified therapeutic cells after transplantation, and to allow for the removal of the GDNF-producing cells if untoward effects occur. This technique is based on surrounding the cells with a synthetic perm-selective membrane whose pores are sufficiently large to allow for the inward diffusion of nutrients and the outward diffusion of GDNF, but sufficiently small to prevent interaction with the host immune system. The concept has already been used and validated in humans for intrathecal and ICV delivery of the neurotrophic factor CNTF (Aebischer et al., 1996, Bloch et al., 2004).

The objectives of this review are threefold: First, to summarize available evidence that encapsulated, GDNF-producing cells can have neuroprotective and neuroregenerative actions on nigrostriatal dopaminergic neurons in animal models of PD. Second, to discuss the contradictory clinical data from the human trials with GDNF infusion performed so far and their implications for the further development of GDNF-based therapies for PD. Third, to delineate the different scientific and regulatory steps that need to be taken in order to bring the strategy of encapsulated, GDNF-producing cells into clinical use.

Section snippets

Can encapsulated, GDNF-producing cells give rise to neuroprotection and neuroregeneration in animal models of Parkinson's disease?

Implantation of encapsulated GDNF-producing cells unilaterally into the rat striatum 1 week prior to intrastriatal injection of the neurotoxin 6-hydroxydopamine (6-OHDA) produced substantial protection of both striatal dopaminergic innervation and substantia nigra neurons (Shingo et al., 2002). In these animals, the apomorphine-induced rotational asymmetry was markedly reduced during 6 months after implantation as compared to rats which had received non-GDNF control devices. Thus, the

What is the rationale for GDNF delivery via encapsulated cells as a therapeutic strategy in patients with Parkinson's disease?

Four clinical trials with GDNF administration in patients with PD have so far been performed with diverse outcomes. In the first study, GDNF was injected into the cerebral ventricles but without any symptomatic improvement (Nutt et al., 2003). The most likely explanation to the lack of efficacy and trophic action on striatal dopaminergic fibers was that the protein had not diffused into the brain parenchyma (Kordower et al., 1999). Instead, side effects developed such as nausea, loss of

What is the road-map to the clinic for encapsulated, GDNF-producing cells?

The concept of using encapsulated cells in implantable and retrievable devices for therapeutic applications in the CNS including PD is not new (Lysaght and Aebischer, 1999). However, even though animal studies and a few human proof-of-concept trials have been performed, no late-stage clinical product development or marketed product has been realized. The reasons for the failures are several; regulatory and functional issues associated with the use of xenogeneic cells, problems with

Perspectives

In the future, intrastriatal transplantation of encapsulated, GDNF-producing cells could potentially be of value for PD patients both by inducing symptomatic relief and by modifying the course of the disease. The clinical usefulness of this strategy will be determined primarily by its ability to provide long-term suppression of the degeneration of dopaminergic neurons. Because the available animal models do not mimic the human disease process, GDNF's possible neuroprotective action in PD can

References (24)

  • D. Kirik et al.

    Localized striatal delivery of GDNF as a treatment for Parkinson disease

    Nat. Neurosci.

    (2004)
  • J.H. Kordower et al.

    Clinicopathological findings after intraventricular glial-derived neurotrophic factor treatment in a patient with Parkinson's disease

    Ann. Neurol.

    (1999)
  • Cited by (0)

    View full text