UNLABELLED We have previously reported the results of a 1-y double-blind, placebo-controlled study of embryonic dopamine cell implantation for Parkinson's disease. At the end of the blinded phase, we found a significant increase in putamen uptake on (18)F-fluorodopa ((18)F-FDOPA) PET reflecting the viability of the grafts. Nonetheless, clinical improvement was significant only in younger (age < or = 60 y) transplant recipients, as indicated by a reduction in Unified Parkinson's Disease Rating Scale (UPDRS) motor scores. METHODS We now report long-term clinical and PET outcomes from 33 of the original trial participants who were followed for 2 y after transplantation and 15 of these subjects who were followed for 2 additional years. Longitudinal changes in UPDRS motor ratings and caudate and putamen (18)F-FDOPA uptake were assessed with repeated-measures ANOVA. Relationships between these changes over time were evaluated by the analysis of within-subject correlations. RESULTS We found that UPDRS motor ratings declined over time after transplantation (P < 0.001). Clinical improvement at 1 y was relatively better for the younger transplant recipients and for men, but these age and sex differences were not evident at longer-term follow-up. Significant increases in putamen (18)F-FDOPA uptake were evident at all posttransplantation time points (P < 0.001) and were not influenced by either age or sex. Posttransplantation changes in putamen PET signal and clinical outcome were significantly intercorrelated (P < 0.02) over the course of the study. Image analysis at the voxel level revealed significant bilateral increases in (18)F-FDOPA uptake at 1 y (P < 0.001) in the posterior putamen engraftment sites. PET signal in this region increased further at 2 and 4 y after engraftment. Concurrently, this analysis disclosed progressive declines in radiotracer uptake in the nonengrafted caudate and ventrorostral putamen. Clinical improvement after transplantation correlated with the retention of PET signal in this region at the preoperative baseline. CONCLUSION These results suggest that clinical benefit and graft viability are sustained up to 4 y after transplantation. Moreover, the dependence of clinical (but not imaging) outcomes on subject age and sex at 1 y may not persist over the long term. Last, the imaging changes reliably correlate with clinical outcome over the entire posttransplantation time course.