To the editor

Much debate has been generated from the results recently published by Freed et al.1 describing the outcome of the first double-blind placebo-controlled clinical trial on the efficacy of embryonic neural tissue transplantation in Parkinson disease. Particular interest has been focused on the severe dyskinetic side effects observed in several patients in the study, from which the authors drew the conclusion, highlighted by your editorial2, that "transplanting less fetal tissue in the future" would be the simplest way to increase the clinical efficacy and safety of neural transplantation in Parkinson disease patients. We question both this interpretation and the remedy offered.

This study shows modest symptomatic improvements in younger patients at one year after transplantation. The necessity to improve the safety of their transplantation method was based on the observation that 5 of 33 transplanted patients developed dyskinesias and facial dystonias, even in the absence of medication. These were interpreted as resulting from “a relative excess of dopamine” derived from “too many” grafted neurons, in direct analogy to peak-dose dyskinesias seen with L-dopa medication. However, a more careful analysis of the post mortem data obtained from two of their patients reveals a number of facts that may point in another direction. Each transplant patient received the equivalent amount of tissue derived from only two embryonic ventral mesencephali per putamen. In the two post mortem autopsy cases, survival of only modest numbers of dopaminergic neurons were reported (7,000–38,000 per putamen). This stands in contrast to the substantially higher numbers (82,000–138,000 per putamen, and up to 208,000 in total in one patient)3, alongside a generally more pronounced clinical improvement3,5, achieved using conventional neurosurgical techniques and fresh tissue from embryos that had been cool-stored for at most a few days3. More generally, the extent of symptomatic improvement over several years after transplantation is closely related to the degree of dopaminergic-cell survival and reinnervation as assessed by positron emission tomography4, which is in turn dependent on the amount of tissue implanted as well as the methods of preparation5. From this perspective, the numbers reported by Freed et al. are at the lower rather than upper limits of survival for any significant functional benefit to be seen. Most importantly, previous studies5,6,7 have not reported dyskinesia or dystonia in any way comparable in either frequency or magnitude to those reported by Freed et al. Therefore, the lack of more substantial clinical improvement and the occurrence of major side effects are unlikely to be due to the implantation of too much tissue, but rather to a number of major variations of tissue preparation, implantation technique and, possibly, the lack of immunosuppression employed by Freed and colleagues (manuscript submitted and http://www.nesu.mphy.lu.se/nectar). Unfortunately, too little information is provided to resolve the source of dyskinesias in these patients as the authors have not used standardized assessment protocols for neural transplantation to permit comparability with other studies8,9. Moreover, there is no receptor–ligand neuroimaging4, which might resolve the relationship between cell survival, reinnervation and receptor sensitivity on which the dyskinetic side effects presumably depend.

In conclusion, the study by Freed et al. provides further evidence that neural transplants can ameliorate parkinsonian symptoms (and that sham surgery has little, or no, lasting effect), but it has opened up more, and not fewer, questions that should foster basic science and clinical research in this exciting field of neural repair.

See Reply to “Transplanted dopaminergic neurons: More or less?” by Freed et al.