Background: In an open-label phase 1 trial, gene delivery of the trophic factor neurturin via an adeno-associated type-2 vector (AAV2) was well tolerated and seemed to improve motor function in patients with advanced Parkinson's disease. We aimed to assess the safety and efficacy of AAV2-neurturin in a double-blind, phase 2 randomised trial. Methods: We did a multicentre, double-blind, sham-surgery controlled trial in patients with advanced Parkinson's disease. Patients were randomly assigned (2:1) by a central, computer generated, randomisation code to receive either AAV2-neurturin (5·4×1011 vector genomes) injected bilaterally into the putamen or sham surgery. All patients and study personnel with the exception of the neurosurgical team were masked to treatment assignment. The primary endpoint was change from baseline to 12 months in the motor subscore of the unified Parkinson's disease rating scale in the practically-defined off state. All randomly assigned patients who had at least one assessment after baseline were included in the primary analyses. This trial is registered at ClinicalTrials.gov, NCT00400634. Results: Between December, 2006, and November, 2008, 58 patients from nine sites in the USA participated in the trial. There was no significant difference in the primary endpoint in patients treated with AAV2-neurturin compared with control individuals (difference -0·31 [SE 2·63], 95% CI -5·58 to 4·97; p=0·91). Serious adverse events occurred in 13 of 38 patients treated with AAV2-neurturin and four of 20 control individuals. Three patients in the AAV2-neurturin group and two in the sham surgery group developed tumours. Interpretation: Intraputaminal AAV2-neurturin is not superior to sham surgery when assessed using the UPDRS motor score at 12 months. However, the possibility of a benefit with additional targeting of the substantia nigra and longer term follow-up should be investigated in further studies. Funding: Ceregene and Michael J Fox Foundation for Parkinson's Research. © 2010 Elsevier Ltd.