Emad Farag Discusses Retinal Pigment Epithelial Cell Implantation for Parkinson's Disease

November 10, 2009


Emad S. Farag, MD, Assistant Professor of Neurology at UCLA, discusses his paper, "Pathologic findings in retinal pigment epithelial cell implantation for Parkinson disease," which was recently published in Neurology® (2009;73:1095-1102). He spoke with José G. Merino, MD, MPhil, Science Editor of AAN.com.

AAN.com: Can you describe your study and summarize the major findings?

Farag: We studied a cerebral hemisphere from a patient who had undergone surgical implantation of retinal pigment epithelial (RPE) cells in gelatin microspheres as part of a clinical trial for Parkinson's disease. We used routine histological and immunohistochemical methods to examine the tissue. In addition, we characterized Alzheimer's disease pathology, Lewy body pathology, reactive astrocytosis, hemosiderin deposition, and cellular inflammation. Staining included H&E, Bielschowsky silver, Perl's iron, glial fibrillary acidic protein, alpha-synuclein, and CD 4, CD 8, and CD 19 B and T cell markers. Manual cell counts were performed on RPE cells. We found RPE cells in the brain, but overall survival was poor (0.036 percent). There was prominent local inflammatory and astrocytic reactive change along the implantation tract.

AAN.com: What was the clinical status of the patient at the time of transplantation? Why did he die?

Farag: The patient had advanced Parkinson's disease, and his symptoms did not improve after transplantation. We could not determine the cause of death at autopsy, as only the brain was available for examination. The patient had a history of a fall, and possible related complications, before his death. We do not know if the trauma had an effect on cell survival, but this is unlikely.

AAN.com: What are RPE cells and why they are being evaluated as a potential therapy for Parkinson's disease?

Farag: RPE cells are normal constituents of the human retina. They line Bruch's membrane and play a role in absorbing scattered light, transporting nutrients, and recycling vitamin A, among other things. There is ample data to suggest that RPE transplants may help patients with Parkinson's disease. In vitro studies showed that RPE cell culture media promotes survival of dopaminergic cells. In animals, RPE implantation led to improvement in dopaminergic function (as demonstrated using functional imaging) and motor deficits. Open label clinical trials in humans also showed improved motor function.

AAN.com: How did you identify the implanted cells at autopsy?

Farag: We identified the implanted cells at autopsy based on their morphologic characteristics; their dense neuromelanin granular contents and presence within characteristic matrix support material made them stand out from the background neuropil. The patient had neuropathological features consistent with a diagnosis of advanced Parkinson's disease, including Lewy bodies and neuronal loss in the substantia nigra and other characteristic sites. The RPE cells did not have Lewy bodies. We found a cellular inflammatory response near the matrix material and at needle tract sites away from matrix material. This inflammatory response consisted primarily of macrophages. The effect of the inflammation on cell survival is unclear, but it is possible that inflammation impacted it adversely.

AAN.com: How are the RPE cells prepared for implantation in the brain?

Farag: The details of the clinical trial in which the patient was a participant have not been published, and the precise implantation methodology has yet to be described. In other studies, cells have been implanted at five sites within the putamen using stereotactic procedures. In our patient, 325,000 cells were implanted. The cells were mounted on a matrix support material that acts as a support lattice for the RPE cells, can promote the survival of the RPE cells, and possibly furnish trophic support for them.

AAN.com: What do you think triggered the host's inflammatory response?

Farag: This is of course a key question, and cannot be definitively answered from our study. It is unclear if any immunogenic triggers were present in the tissue. Antigenic properties of the RPE cells or the support matrix are possible causes, but prior animal and human studies do not support this hypothesis. As noted in the editorial by James F. Morley, MD, PhD, and John E. Duda, MD, that accompanies our paper, another important consideration is that the neuropathologic findings, irrespective of cell type, may be time dependent, and that the inflammatory response may be phasic. This adds yet another variable.

AAN.com: What are other cell-based therapies under investigation for Parkinson's disease?

Farag: Several cell-based therapies have been tried, including fetal ventral mesencephalic grafting, adrenal medullary (chromaffin cell) grafting, and autologous sympathetic ganglion cell transplants. While Kordower and colleagues found limited cell survival but possibly preserved neuritic sprouting in the area of the cells at 30 months after implantation, others have not found adrenal cell graft survival years after implantation. Results from fetal nigral transplants have been more variable. Survival has been documented up to 14 years after surgery, with Lewy bodies seen in surviving cells. In a subject examined 18 months post implantation, significant differences in cell survival were seen across hemispheres, and from tract to tract within a single hemisphere. Factors such as the number of donors from which the cells were harvested—and the area of cell implantation—were associated with survival. Inflammation has been noted in some of these reports, but overall it has not been felt to be a major cause for poor survival. It is difficult to compare the role of inflammation among studies, however, because fetal cell implant protocols usually require the use of cyclosporine for a period of time, but this was not the case for our patient.

AAN.com: What are the barriers to cell transplantation?

Farag: Changes in the trophic support or other function of the matrix material is a possible barrier to transplantation. In addition, recent findings from fetal nigral cell transplants suggest that host neurodegenerative neuropathology can affect grafted cells over time. This has not been reported with RPE cell transplants, and our patient did not have Lewy bodies in the transplanted cells. Survival may depend on factors specific to the biology of the transplanted cells. Inflammation may also play a role.

AAN.com: What can be done to promote cell survival after RPE cell implantation? Are any interventions being evaluated in humans?

Farag: This is unknown. Initial studies in animals using RPE cells suggested that no immunosuppression was necessary for cell survival and clinical benefit. Our findings suggest that the effect of inflammation after cell transplantation should be reconsidered. Further study is needed into inflammatory and other factors that may influence long-term survival of grafted cells of various types in host brains.

Author Disclosures

Dr. Farag received funding support from VA Southwest Parkinson's Disease Research Education, and Clinical Center (PADRECC), an NIH grant, and a pilot grant from a Team Parkinson project examining Lewy bodies in PD and PD dementia.

Dr. Merino performed a one-time consultation with staff from Bell, Falla and Associates.