First Successful Surgical Trial for Gene Therapy in Parkinson's Disease

Abstract & Commentary

By Claire Henchcliffe, MD, Associate Professor of Neurology and Neuroscience, Weill Cornell Medical Center. Dr. Henchcliffe reports she is on the speakers bureau and advisory board for Allergan and Teva; speakers bureau for Boehringer-Ingelheim, GlaxoSmithKline, and Novartis; advisory board for Merz; and is a consultant for Gerson Lehman Group and Guidepoint Global.

Synopsis: A controlled surgical trial of glutamic acid decarboxylase gene therapy, targeting the subthalamic nucleus in advanced Parkinsons's disease, demonstrated statistically significant improvement in motor symptoms off medication in those receiving gene therapy compared with controls.

Source: LeWitt PA, et al. AAV2-GAD gene therapy for advanced Parkinson's disease: A double-blind, sham-surgery controlled, randomised trial. Lancet Neurol 2011;10:309-319.

This multicenter, double-blind, randomized, sham-surgery controlled trial examined the effects of introducing the glutamic acid decarboxylase (GAD) gene into the subthalamic nucleus (STN), bilaterally, of 45 individuals with advanced Parkinson's disease (PD), by means of the adeno-associated viral vector 2 (AAV2). This approach has well-described scientific underpinnings. GAD catalyzes the rate-limiting step in GABA synthesis, so by increasing GABA-ergic "tone," it would "convert" STN output from the abnormally excitatory state seen in PD, thus alleviating motor symptoms. In the present study, participants were randomized to one of two scenarios: 1) AAV-GAD infusion or 2) sham surgery. Mean participant age was 61.8 ± 7 and 60.6 ± 7.4 years (AAV-GAD and sham groups, respectively), and mean disease duration was 10.6 ± 4.3 years and 12.0 ± 5 years (AAV-GAD and sham groups, respectively). Of the 22 subjects assigned to receive AAV-GAD, 16 were eligible for analysis, and 21/23 subjects assigned to sham-surgery were eligible (exclusions were based on improper targeting revealed by subsequent imaging, infusion failure, pump malfunction). Over a 6-month duration, investigators were encouraged to keep medications stable. Notably, at this time point, Unified Parkinson's Disease Rating Scale (UPDRS) motor subscores, with patients in the "off" state, improved from 34.8 ± 1.6 to 26.6 ± 2.0 in the AAV-GAD group, compared with 39.0 ± 1.9 to 34.3 ± 2.5 in the sham-surgery group (P = 0.04). In a responder analysis of these same scores (cutoff: 9 point improvement), 50% in the AAV-GAD group were assessed as responders, compared with just 14.3% in the sham surgery group (P = 0.03). Changes in wearing off response and freezing of gait also were superior in the AAV-GAD group vs the sham-surgery group.


The investigators have designed an exceedingly rigorous approach to test effects of somatic gene therapy in advanced PD, following a previous open-label trial of unilateral GAD-AAV gene that demonstrated safety and tolerability. Surgical introduction of the GAD gene into the bilateral STN resulted in achievement of the study's primary endpoint, that is, a significant improvement in UPDRS motor scores (off medication) over the sham-surgery "placebo." This is encouraging and provides hope for a gene therapy intervention to provide an efficacious and safe treatment for PD. There are a number of features in the study's design that likely contributed to its success. First, patients were selected on rigorous clinical criteria (including PD diagnosis of > 5 years and a sustained levodopa response of > 12 months) plus a requirement that 18F-flurodeoxyglucose positron emission tomography imaging was consistent with a diagnosis of PD according to published criteria. This actually resulted in 11 screening failures, quite surprising in a highly selected population. Second, the investigators chose an elaborately simulated sham-surgery procedure, with OR personnel making extensive efforts to reproduce all features of the AAV-GAD infusion, including stereotactic frame placement, sounds of microelectrode recordings played during the sham procedure, similar external catheter and pump arrangements, and a similar time frame for the procedure. Subjects were asked to guess assignment on post-operative day 3, and in both groups the majority thought they had undergone AAV-GAD infusion rather than sham-surgery. Third, the investigators examined results separately from subjects whose catheter tip placement (therefore gene transfer) was inaccurate, and 4/5 of these subjects had no improvement during the study. In summary, the methodology employed will provide insight for trials to come, and it will be fascinating to see more long-term data from this first successful sham-surgery controlled gene therapy trial.