Background

Deep Brain Stimulation (DBS) is an invasive treatment. Pairs of electrodes are surgically implanted in the brain and connected to a pulse generator placed under the skin on the chest. DBS-nbM delivers low-frequency stimulation to the nucleus basalis of Meynert in the basal forebrain. The nbM is a major source of acetylcholine in the brain, with projections to the hippocampus and cortex. Its degeneration in Alzheimer’s, Parkinson’s, and Lewy body diseases results in loss of cholinergic transmission and subsequent cognitive decline. In Parkinson’s patients, acetylcholine deficits are also linked to delusions and visual hallucinations. Using DBS to activate these neurons and stimulate acetylcholine release thus targets the same pathway as cholinesterase inhibitors.

In addition to enhanced acetylcholine release, potential mechanisms of DBS-nbM include increased cerebral blood flow, neurotrophic factor release, neuroplasticity, and hippocampal enlargement. DBS-nbM improved cognitive measures in rats with basal forebrain cholinergic neuron lesions, and promoted neurogenesis, synaptic plasticity, and cognitive performance in rats with scopolamine-induced dementia (Lee et al., 2016; Liu et al., 2022). Healthy adult rats, but not aged rats, increased secretion of nerve growth factor in the cortex in response to DBS-nbM (Hotta et al., 2009). In the APP/PS1 model of AD, high-frequency, bilateral nbM stimulation improved animals’ performance in the Morris water maze (Huang et al., 2019). In a transgenic rat model of AD, bilateral intermittent DBS-nbM led to supernormal performance on a spatial memory task (Koulousakis et al., 2020). Monkeys exposed to intermittent stimulation improved performance in a working memory task, and DBS was as effective as an acetylcholinesterase inhibitor (Liu et al., 2017). In this study, continuous stimulation impaired performance (for review of primate studies, see Bava et al., 2023). 

Interest in DBS-nbM for Parkinson’s disease was spurred by a case report that stimulation of the nbM in a PD patient with dementia led to a marked improvement in several cognitive tests, which reversed to baseline when the electrodes were shut off (see Jun 2009 news on Freund et al., 2009). Another case report supported the feasibility of dual stimulation, e.g., using a single lead with multiple electrodes to simultaneously stimulate the nbM and the subthalamic nuclei (STN, Nombela et al., 2019). The latter is a standard clinical target for DBS to control motor symptoms of Parkinson’s disease.

Other clinical studies evaluating DBS for AD are targeting the fornix and the ventral capsule/ventral striatum, a modulator of frontal lobe networks (Scharre et al., 2018).

Findings

The first clinical trial of DBS-nbM for Alzheimer’s disease began in January 2010 in Germany, where six mild to moderate AD patients were implanted with bilateral electrodes, and received continuous, low-frequency stimulation for two weeks. This was followed by a two-week off period, and then stimulation for 11 months. The primary endpoint was the ADAS-Cog. According to published results, the intervention was safe. After one year, four patients were stable or improved on the ADAS-Cog, and three had increased glucose uptake on FDG-PET (May 2014 news on Kuhn et al., 2014). Additional analyses reported improvements in nutritional status during treatment, and associated treatment effects with preserved cortical thickness (Noreik et al., 2015; Baldermann et al., 2017; Hardenacke et al., 2016). A separate study by the same group described cognitive improvements in two younger patients treated at an earlier stage of AD (Kuhn et al., 2015).

In 2017, a study began in Spain to test stimulation of the nbM or fornix in six people with AD. The study ended after treating just one patient, who received a fornix implant (Barcia et al., 2022).

In April 2020, the Beijing Pins Medical Company registered a study using its implantable neurostimulator for DBS-nbM. The status of the study, involving 30 AD patients at one hospital in Beijing, is unknown. In another trial of the same device, eight patients with advanced AD received DBS-nbM for 12 months. The study reported transient cognitive improvement on the MMSE at one month (Jiang et al., 2022; Zhang et al., 2021).

A study in Parkinson’s disease dementia ran at the University College London from 2012-2015. The crossover, sham-controlled trial in six patients found six weeks of low-frequency DBS-nbM was safe but elicited no improvement on a cognitive battery, the primary outcome (Gratwicke et al., 2018). Some improvement on psychiatric scores was reported. A three-year follow-up found varying rates of cognitive decline among the patients, but reached no conclusions about effects of stimulation (Cappon et al., 2022).

A 2014-2016 trial of DBS-nbM for Lewy body disease at University College London treated six patients with mild to moderate dementia. After implantation, participants received six weeks of electrodes on and six weeks off in the crossover design, with primary outcomes assessing cognitive, psychiatric, and motor symptoms. The treatment was safe. It did not improve cognition, but reduced neuropsychiatric symptoms in three of five patients. Functional MRI revealed stimulation-associated functional connectivity changes in brain networks involved in cognition (Gratwicke et al., 2020; also see comment by Liu and Yu). A different trial in six French patients compared active stimulation to sham for three months each in a crossover design. The primary outcome of selective recall did not differ between groups, but the authors noted significant decreases in some cognitive test scores after implantation, and after starting stimulation (Maltête et al., 2021).

Several PD trials have tested, or are testing, combined stimulation paradigms that target both the STN and the nbM. One, a 2017 sham-controlled study in Canada found no cognitive improvement from one year of combined stimulation in six patients (Sasikumar et al., 2022). A similar study was conducted in 10 patients in Germany starting in 2016 (Daniels et al., 2020); its results have not been posted to date. A dual stimulation study is ongoing at the University of São Paulo General Hospital in Brazil. Enrolling 10 patients with PD and mild cognitive impairment, its primary outcome is safety. The trial will also assess a long list of secondary cognitive and motor outcomes after 16 weeks of dual stimulation. The study is slated to complete in January 2024.

Another study is planned to begin in October 2023 testing the safety of implanting nbM leads as an add-on to STN leads in 10 PD patients. Thus, investigators will place a total of four leads during surgery. The trial will compare two surgical approaches for targeting the nbM, and assess novel stimulation patterns. Secondary outcomes are motor and cognitive tests over two years. Funded by the NINDS, the trial will run at Stanford University in California until August 2027.

Also in October 2023, a new AD study is set to begin testing intermittent nbM stimulation in people with mild cognitive impairment or early dementia. The study, at Vanderbilt University, will enroll eight people, who will receive one hour of low-frequency stimulation per day, or sham stimulation, for one year. The primary outcome is Clinical Dementia Rating score. The expected finish date is October 2028.

For details on DBS-nbM trials, see clinicaltrials.gov.