Gene Therapy in Alzheimer Disease-It May Be Feasible, but Will It Be Beneficial?

In this issue of JAMA Neurology, Rafii et al1 report the results of a randomized clinical trial of intracerebral gene delivery in patients with Alzheimer disease (AD). Neurodegenerative diseases, of which AD is the most common, remain a challenging frontier for medical treatment. There are no preventive measures, and for the millions of individuals who already have AD, there are no disease-modifying agents to slow, stop, or reverse progression. The primary efforts of the field include a diverse assortment of ongoing drug trials attempting to alleviate or reduce various AD pathophysiologic processes: amyloidogenesis, neurofibrillary change, synaptic losses, and neuronal loss.2 While these efforts may be getting closer to success, they have not yet reached fruition. Meanwhile, there is a less advanced, secondary, intercurrent interest in restorative or regenerative strategies. Degenerative or destructive disorders of systemic organs, such as the heart, liver, or kidney, can be successfully treated with transplantation strategies. Such options are not available for the brain. Thus, there is a burgeoning interest in regenerative strategies on a more granular scale—via cell transfers, such as stem cells, or via genetic material, such as viral vectors, DNA, RNA, or antisense agents.3 The article by Rafii et al1 represents an effort to determine feasibility of helping patients with AD through intracerebral gene delivery. An earlier open-label study demonstrated the potential ability to perform such delivery.4 The current study extended this and allowed preliminary evaluation of safety and efficacy through performance of placebocontrolled treatment (sham surgery) in 49 patients with mild to moderate AD, using a viral vector to deliver DNA that codes for nerve growth factor (NGF) into the brain’s nucleus basalis of Meynert (NBM) through stereotactic neurosurgery. This trial did not show evidence of benefit, but the investigators should be commended for a well-planned and well-executed study. The rationale for this trial1 was to treat the cholinergic deficit in AD. In AD, the NBM, a small set of cells in the basal forebrain, experiences atrophy and destruction. The degeneration of this nucleus is responsible for cortical cholinergic deficits in AD, because the NBM magnocellular neurons project widely throughout the frontal, parietal, and temporal neocortical regions, as well to the hippocampus, an important region for memory, and supply cholinergic input into these diverse regions. In fact, it is this deficit in cholinergic tone that 3 of the widely used US Food and Drug Administration–approved drugs (donepezil, galantamine, and rivastigmine)5 partially ameliorate through central inhibition of synaptic acetylcholinesterase activity. Furthermore, the cells of the NBM are known to be sensitive to the neurotrophic molecule NGF, with culture and animal experiments showing that NGF deprivation can result in the degeneration and death of NBM cells and that NGF exposure can rescue such deprived NBM neurons or cause hypertrophy and increased cholinergic function in these cells.6-8 Thus, the reasoning was that increasing NGF in the basal forebrain might improve NBM neuronal function in patients with AD, causing increased cholinergic function and better cognition. Such a treatment would only be a small part of the total, because brain degeneration in AD is much more widespread, involving many more structures than the NBM and many more features than simply cholinergic dysfunction. Synaptic losses, neuronal dysfunction, and death afflict the hippocampal region and nearly all neocortical regions, independent of cholinergic status. Thus, at best, a successful intervention at the level of NBM or cholinergic function would likely provide only a modest symptomatic benefit, perhaps similar to that of the approved anti-acetylcholinesterase drugs. But demonstration of success with such a strategy might indicate potential for development of further strategies devoted to neuronal survival and restoration. The trial by Rafii et al1 enrolled 49 people with AD, who ranged in age from 55 to 80 years and had disease severity ranging from mild to moderate (with Mini-Mental State Evaluation scores of 17 to 26 on a 30-point scale). The trial occurred at 10 academic medical centers over a 3-year period and had 86% participant retention. The intervention was bilateral, stereotactically guided injection into the NBM of 2x1011 viral genomes (in approximately 20 μL of volume) of adenoassociated-virus-2 (AAV2) vector, which codes for expression of full-length human nerve growth factor (NGF). The AAV2 vector is useful for delivering the genetic material of small molecules because it has a small single-stranded DNA genome of approximately 4700 nucleotide bases;9 this can accommodate a small strand of complementary DNA, such as the one encoding NGF (which has about 1000 nucleotide bases).10 The AAV2 vector can infect nondividing cells (such as neurons), does not kill these cells, and is not thought to engender an immune response, making it a reasonable vector.9,11 This study involved randomization to a comparison control arm: half the research participants underwent the same processes as those delivered drugs, but had sham surgery, receiving neither agent (AAV2-NGF) nor placebo injections, but had neurosurgical preparations, stereotactic frame, and partial burr holes without injections. Because surgical interventions are accompaRelated article Opinion