Cell therapy: replacement.

Not long ago, the ability of the brain to restore function through regeneration of neural elements was thought to be nonexistent. It is now known that not only does some regenerative capacity exist, but implanted cells can integrate into the host brain, survive, and reverse neurological deficits. Neural stem cells, fetal transplants, immortalized cell lines, and bone marrow stromal cells show promise in experimental models of neurological disease including stroke. Although it is clear that transplanted cells function, the mechanism by which neurological deficits might improve is less certain. Transplanted cells may preserve existing host cells and connections through secretion of trophic factors; establish local connections that enhance synaptic activity; provide a bridge for host axonal regeneration; or actually replace cellular elements. Several observations from animal and human studies of cell therapy support the possibility that transplanted cells exert at least some of their effect through cellular replacement. In the early stages of brain development, implanting neural stem cells leads to replacement of multiple cellular elements including neurons and glia.1 Thus, the potential for cell replacement exists, but whether it persists into adulthood is uncertain. Models of Parkinson’s disease (PD) provide the most direct support for cell replacement as an important effect of cell therapy. Fetal ventral mesencephalic neurons grafted into the striatum in animal models of PD restore dopamine levels and improve function.2 Similar grafts outside the striatum fail to achieve clinical benefit. In humans, such fetal grafts produce clinical benefit3 that accrues gradually rather than immediately, suggesting an accumulation of synaptic connections that eventually results in sufficient dopaminergic transmission to improve neurological deficits. Autopsy findings in patients receiving fetal grafts demonstrate implanted cell survival as well as axon growth and synaptic connections4. Additional support comes from positron-emission tomography studies showing a correlation between clinical improvement and increased uptake of [F]fluorodopa in the striatum. This favors the concept that the response to grafting is mediated by direct activity of the transplanted cells replacing the function of the degenerating dopaminergic cells of the host nigro-striatal pathway. The challenge of cell replacement for treatment of stroke is in some ways similar to that for PD but in other ways is very different. Like PD, the injury is focal but the neuronal loss typically involves many more cell types and neurotransmitters. Neural pathways are more complex, and the likelihood of implanted cells forming appropriately directed connections necessary to restore function seems remote, unless guided by the host brain. Despite the potential pitfalls, treatment of focal ischemia in animals has demonstrated promising results. Fetal cortical grafts placed in adult neocortex following ischemia make connections with host neurons including cortex, thalamus, and subcortical nuclei.5 Behavioral improvement occurs in response to these grafts when animals are exposed to an enriched environment. Neuronal cells derived from a human teratocarcinoma cell line (NT2 cells) implanted into the striatum following infarction survive and integrate into the host brain, growing axons and making synaptic connections.6 Neurological deficits due to stroke are reversed by implantation.7 The clinical benefit occurs only when a critical number of cells are transplanted, ensuring adequate cell survival. The fact that response depends on the number of cells transplanted suggests the benefit may be mediated by cell replacement. Extrapolating the results of cell implantation in animal models of stroke to humans is problematic, particularly because of the relative lack of adequate primate stroke models. Unlike PD, in which the motor manifestations of striatal lesions mimic the human disease, deficits in animals due to ischemia are more difficult to compare with human stroke. The first human trial of cell therapy for stroke included 12 patients treated with LBS neurons derived from a teratocarcinoma cell line.8 This trial was not designed to examine efficacy, but improvement in some patients on the European Stroke Scale scores and NIHSS scores was observed. As in PD, positron-emission tomography studies showed increased metabolic activity in the area of the grafts in several patients 6 and 12 months after implantation.9 The results of an autopsy in one patient 18 months after implantation documented survival of transplanted neuronal cells.10 Taken together, these data support the concept that activity of implanted cells is responsible for clinical changes. Further studies are needed to more precisely determine the role of cell replacement—whether the implanted cells form new neural pathways, make local connections, or work by neurohumoral mechanisms. The opinions expressed in this editorial are not necessarily those of the editors or of the American Stroke Association. From the Departments of Neurology (L.R.W.) and Neurological Surgery (D.K.), University of Pittsburgh School of Medicine, Pittsburgh, Pa. Correspondence to Dr Lawrence R. Wechsler, University of Pittsburgh Medical Center, Stroke Institute, Department of Neurology, C426 PUH, 200 Lothrop St, Pittsburgh, PA 15213. E-mail [email protected] (Stroke. 2003;34:2081-2082.) © 2003 American Heart Association, Inc.