Neuromodulation in the Treatment of Alzheimer’s Disease: Current and Emerging Approaches

INTRODUCTION

With an aging global population, disorders such as Alzheimer’s disease (AD) are becoming increasingly prevalent [1]. AD affects about 3% of the population between 65–74 years of age and >50% of the population over the age of 85 [2]. Symptoms typically begin with episodic memory difficulties, and as the disease progresses, other cognitive domains, such as language and executive function, are affected [3–5]. Pathologically, AD is characterized by the presence of extracellular amyloid-β (Aβ) plaques and intraneuronal tau neurofibrillary tangles (NFT). The most prevalent theory, the amyloid cascade hypothesis, proposes that Aβ plaques precede and influence the formation of NFTs, which in turn leads to downstream neurodegeneration and cognitive impairments [6–9]. Additional theories of AD pathogenesis share many features, but attempt to better integrate vascular and inflammatory pathology or immune-pathogen interactions as core features of the disease. These theories arose in response to abundant evidence of chronic inflammation and cerebrovascular pathology, as well as evidence of infections in many AD patients and results demonstrating seeding of Aβ pathology by microbial proteins [10–12].

In 2018, the NIA-AA updated their criteria for AD, emphasizing the importance of biomarkers in a biological definition of AD [8]. The new classification changed the definition of AD from a set of clinical symptoms to the presence of Aβ plaques and NFT [8]. Recognizing that the pathogenesis of AD begins prior to the development of clinical symptoms, a phase termed preclinical AD, will facilitate identification and preventative treatment [8]. It has been suggested that commencing treatment at earlier stages of AD has higher potential to slow or prevent downstream neurodegeneration [8, 13]. As existing neurodegeneration may preclude clinically significant efficacy for many therapies, the push to develop early detection and intervention strategies, prior to the onset of cognitive impairment, brings new promise [8].

Currently, treatment for AD is symptomatic and largely pharmacological, including cholinesterase inhibitors and the NMDA receptor antagonist mem-antine [14]. Although these are mildly effective at reducing cognitive impairment in the short term, long-term efficacy and their effect on disease course is uncertain and minor at best [15–17]. Immunological strategies such as monoclonal antibodies or active immunization targeting Aβ and to a lesser extent tau are in active clinical trials, though results have been mixed on pathology-related endpoints and largely negative on clinical scores [18, 19], with some recent exceptions [20, 21]. Nonetheless, given the lack of approved disease-modifying therapies for AD, and the growing burden the disease will place on global healthcare infrastructure [1], the need for therapies with the potential to slow or prevent progression and deterioration is great.

Development of AD therapies is also complicated by trial designs necessary for producing strong evidence. Power analyses accounting for the expected rate of decline in early AD patients on cognitive scales including the Clinical Dementia Rating Scale (CDR), Mini-Mental State Examination (MMSE), and cognitive subscore of the AD Assessment Scale (ADAS-cog) suggest a sample of hundreds to thousands of patients would be necessary to detect reduced decline [22, 23]. For example, increased observation period and frequency of measurements might reduce required sample size from tens of thousands to several hundred [22–24]. Certain biological endpoints, such as neuroimaging measures of hippocampal volume, might require only a few dozen patients, although their clinical relevance is uncer{-}tain [22, 24]. Regulatory agencies have recently begun moving toward accepting biomarkers as evidence of efficacy in AD clinical trials, although they are unlikely to be accepted as primary outcomes [25]. Although benefit on well-validated disease-specific scales has been the gold standard, improvement in other domains such as mood, aggressive behavior, and apathy can also lead to substantial improvements in quality of life [26]. The benefit of neuromodulation, particularly non-invasive modalities, as adjunct therapies in AD alongside existing pharmacological or behavioral interventions may be powerful, particularly given the role psychiatric symptoms play in functional impairment.

Neuromodulation involves altering neuronal act-ivities through the delivery of a stimulus, including electrical, magnetic, or optogenetic, to a discrete target, which in turn may yield a therapeutic effect [27, 28]. Neuromodulation techniques vary widely in many aspects including their invasiveness and spatial extent of the brain that can be effectively treated (Table 1), and thus potential applications will differ by modalities. In some cases, neuromod-ulation may simultaneously address cognitive and psychiatric complications of AD by targeting structures implicated in both cognitive and emotional dysfunction [29, 30]. Existing neuromodulatory techniques trialed in AD range from invasive procedures such as deep brain stimulation (DBS; Fig. 1a), to non-invasive approaches such as transcranial mag-netic stimulation (TMS; Fig. 1b), transcranial electric stimulation (tES; Fig. 1c), and focused ultrasound (FUS; Fig. 1d). Additionally, some emerging app-roaches achieve neuromodulation without directly altering electrical activity in the brain, such as photo-biomodulation (PBM; Fig. 1e) as well as visual and auditory stimulation (VAS; Fig. 1f). This literature review will provide an overview of studies in both animal models and humans to evaluate the potential of current and emerging neuromodulation modalities for AD.

OPEN IN VIEWER

Fig. 1

Emerging neuromodulation modalities for the treatment of Alzheimer’s disease. From left to right: a) Deep brain stimulation (DBS) directly modulates neuronal activity via an intracranial electrode driven by a programmable implanted pulse generator (IPG). b) Transcranial magnetic stimulation (TMS) induces currents via a magnetic field generated by passing current through a coil positioned over a patient’s head. c) Transcranial electric stimulation (tES) involves passing current between electrodes attached to a patient’s scalp, a portion of which passes through the skull. d) Focused ultrasound (FUS) involves transcranial mechanotransduction via ultrasound waves, inducing downstream biological effects. e) Photobiomodulation (PBM) involves transcranial transmission of near infrared wavelength light and absorption by intracellular chromophores. f) Visual and auditory stimulation (VAS) employs phasic stimuli that induce entrainment of neuronal activity in both sensory and higher-order brain regions.

DEEP BRAIN STIMULATION

DBS is a neurosurgical procedure involving the direct electrical stimulation of a neuronal population through insertion of an intracranial electrode into a target brain region [31]. Various stimulation parameters such as pulse amplitude, width, and frequency can be modified, and have been shown to elicit a var-iety of biological effects [31]. DBS is standard of care in patients with Parkinson’s disease, dystonia, and essential tremor [32], and has been investigated in clinical trials for a broad range of neurological and psychiatric illnesses, such as treatment-resistant depression [33], anorexia nervosa [34], and addiction [35]. Although DBS is an invasive procedure, acute improvements in the cognitive or psychiatric symptoms of AD could reduce disability and improve quality of life [26], as would longitudinal changes in the rate of neurodegeneration [36].

Proposed mechanisms for the circuit effects of DBS include direct inhibition or excitation of neural activity, indirect effects on neurotransmitters, glial activity, and inhibitory and excitatory balance, as well as functional disruption of connectivity between brain regions [37–39]. An example of acute effects of DBS is the disruption of pathologic oscillations in the subthalamic nucleus that results in immediate improvement in motor features of Parkinson’s disease [37, 41], possibly a result of synaptic de-pression [42]. More delayed effects of DBS suggest changes in plasticity and adaptations in the underlying brain networks [38, 39], or neuroprotective effects, potentially by recruiting activity-dependent signaling mechanisms such as BDNF release [43]. The precise outcomes associated with DBS will de-pend on the brain region targeted and the stimulus parameters, with stimulation frequency and pattern both impacting on the efficacy of DBS [42]. Indeed, high frequency stimulation (e.g., >100 Hz) is more consistently associated with inhibitory effects than low frequency stimulation (e.g., 10–15 Hz) [37, 42].

DBS has also been proposed as a strategy for AD, and potential targets assessed in animal and/or hu-mans include the fornix, entorhinal cortex (EC), nu-cleus basalis of Meynert (NBM), anterior thalamic nuclei, mammillothalamic tract, hippocampus, and ventral capsule [29, 45]. EC stimulation in the TgCRND8 and 3×Tg mouse models of AD attenuated deficits in a range of tasks assessing spatial and recognition memory in both young and aged mice [46–48]. Biological outcomes also included enha-nced neurogenesis as well as reduced plaque load and Aβ peptide concentration, although such effects on amyloid pathology may be age-dependent [47, 48]. Similar results have been observed with stimulation of the midline thalamic nuclei in a TgCRND8 model [46]. In a rat intrahippocampal Aβ injection model, DBS of several targets, namely the fornix, EC, and anterior thalamic nuclei all improved spatial memory, although only forniceal and EC DBS significantly improved recognition memory [49]. Forniceal DBS in a scopolamine-induced rat model of dementia also improved recognition memory [50]. Recently, work in transgenic mouse and rat AD models demonstrated reduced soluble amyloid peptides, reduced hippocampal and cortical apoptosis, and enhanced spatial memory following NBM DBS [45, 51].

Interest in DBS as a treatment for AD in humans followed a case report detailing the effects of stimulation near fiber tracts of the fornix, a component of circuits critical for declarative memory [52]. Acute stimulation induced recall of autobiographical memories and chronic stimulation improved performance on tests of memory and spatial associative learning [52]. Electroencephalographic localization suggested that stimulation influenced structures of the medial temporal lobe [52]. Phase I (n = 6 mild AD) and II (n = 42 mild AD) trials of bilateral fornicial stimulation showed mixed effects on cognitive scores including the MMSE and ADAS-Cog in the phase I, and age-specific effects in the phase II trial [53–56]. Subgroup analyses suggested that DBS may be associated with better outcomes on ADAS-cog and an index of cognitive and functional impairment in older patients, and worse outcomes in younger patients [53, 57]. Exploratory analyses also suggest DBS might have slowed atrophy of the fornix and hippocampus [53–56], and prevented or attenuated reductions in metabolism, particularly within the temporal and parietal cortices [53, 58]. This is consistent with evidence in animal models demonstrating reduced amyloid pathology and neurodegeneration in the medial temporal lobe [59], although the precise mechanism is uncertain and may include complex network effects [50, 60]. A small (n = 3) DBS trial targeting the ventral capsule, a common target for psychiatric indications, suggested reduced progression on cognitive measures relative to a matched control group that received no treatment, with similar effects on brain metabolism [29].

Neurodegeneration affecting cholinergic nuclei, particularly those within the basal forebrain such as the NBM, is an early event in AD thought to underlie some cognitive impairments [14, 61]. Cholinergic neurotransmission from the basal forebrain is critical for attention, learning, and memory, both acutely coordinating network activity supporting these functions as well as regulating plasticity, neuronal sur-vival, and immune response [61–64]. Turnbull et al. stimulated the NBM for AD as early as 1985 in a single case report, although without observing significant clinical benefit [65]. NBM stimulation in an initial group of six AD patients in a phase I trial followed by an additional four patients did not show strong or significant effect on cognitive or functional impairment, however the authors suggested a lower rate of disease progression than expected clinically [66–68]. Here, it appeared that patients in the earlier stages of AD received greater benefit [66, 67]. Similarly, chronic stimulation of the NBM also appeared to prevent expected declines in glucose metabolism as assessed by FDG-PET within frontal, temporal and parietal cortices [66]. These findings are consistent with the proposed neuroprotective mechanisms of NBM DBS, which are thought to be associated with preserving cholinergic transmission [45], but more robust evidence of an altered disease course or neuropathology in larger groups will be needed.

To summarize, DBS findings in animals are more promising than results from human trials (Table 2). This discrepancy may be related to the failure of animal models to recapitulate the full spectrum of pathological features seen in human AD, particularly NFT formation and neurodegeneration [69, 70]. Although clinical outcomes in humans were mixed, DBS was well-tolerated in all patients across both targets with no treatment-limiting, stimulation-related adverse events [53, 66]. However, studies to date were small and there is limited information regarding the long-term effects [53, 66]. Large randomized studies are currently underway and will allow us to evaluate the extent to which DBS can provide clinically meaningful benefits for AD patients.

TRANSCRANIAL MAGNETIC STIMULATION

TMS is a non-invasive means of influencing brain circuitry, wherein a surface coil generates a mag-netic field which induces intracranial electrical currents sufficient to yield suprathreshold effects on neuronal membrane potentials [71]. TMS can produce involuntary movements when stimulating the motor cortex [71, 72]. Circuit effects of TMS tend to be frequency-dependent. Repetitive TMS (rTMS) delivered at low-frequency (<1 Hz) produces functional inhibition and high-frequency (>1 Hz, usually 10–20 Hz) TMS produces excitation [73]. TMS might have therapeutic benefit in AD, but also has potential as a biomarker. For instance, changes in cortical excitability (e.g., reduced motor threshold, motor-evoked potentials, short-latency afferent inhibition) assessed by TMS may be associated with deficits in cholinergic neurotransmission [74–82].

Several studies employing rTMS and related magnetic stimulation techniques in animal AD models have yielded promising results on many outcome measures. rTMS at 1 Hz significantly improved spatial memory and attenuated deficits in hippocampal long-term potentiation in experimental and control groups in a study with Aβ seeding model of AD [83]. Here, rTMS enhanced NDMA receptor subunits and BDNF expression, consistent with another study in a rat model of vascular dementia [84]. After rTMS in healthy rats, CSF levels of sAβPP- were elevated, which was found to attenuate Aβ-induced toxicity in vitro. Similar improvements in behavioral out-comes were observed in transgenic models of AD (e.g., 3×Tg, 5×FAD) after TMS [86–88]. In these studies, improvements were associated with, and in some cases dependent on, normalization of electro-physiological properties and synaptic protein expression in memory relevant brain regions [86–88]. These results suggest that rTMS may have beneficial effects on hippocampal function in both healthy and diseased animals, and could impact multiple aspects of AD.

In clinical studies, the dorsolateral prefrontal cortex (dlPFC) is a common rTMS target for AD, likely because high frequency rTMS of the dlPFC is an FDA-approved treatment for major depression [30, 89–92]. The dlPFC is involved in working memory, and stimulation of this target has been shown to be-nefit relevant measures in randomized, controlled studies [93–96]. In AD patients, high frequency unilateral rTMS improved performance across a variety of language measures both acutely and at longer timepoints (e.g., 12 weeks) [90, 98]. Ahmed et  al. evaluated 1 Hz, 20 Hz, or sham stimulation of bilateral dlPFC in 45 AD patients and found 20 Hz stimulation was generally superior to 1 Hz across measures of global cognition, instrumental activities of daily living, and depressive symptoms [89]. A subgroup analysis suggested that those with milder AD dementia experienced greater benefit than more severe patients.

Additional regions including the precuneus, inferior frontal gyrus, parietal cortex, temporal cortex, Broca’s area, and Wernicke’s area have also been explored [30, 100]. Two randomized controlled trials employed 6 weeks of 10 Hz rTMS of the bilateral dlPFC, bilateral parietal somatosensory cortex, and Wernicke and Broca’s areas alongside cognitive training in 27 AD patients and 15 mild AD patients [30, 92]. The latter study showed significantly improved in ADAS-cog immediately and at 6 weeks, an effect that persisted to 4.5 months. Notably, a subgroup analysis found greater improvement in subjects with mild AD. In another study, improvement in primary outcome measures was observed after 20 Hz rTMS of the parietal and posterior temporal cortex in patients with mild disease only [101]. Lastly, two crossover studies identified selective improvements in delayed recall following 2 weeks of 20 Hz precuneus rTMS in 14 patients with MCI, and in part A and B of the trail making test acutely following 10 Hz inferior frontal gyrus rTMS in 10 patients with early AD [99, 100].

To summarize, rTMS in animal studies corroborate findings in human subjects, and suggest TMS has the potential to both improve cognition and reduce AD pathology. Clinical studies demonstrate significant benefit on various cognitive scales, although often in subgroup analysis (Table 2). Moreover, in some cases benefits appear to persist for weeks or months after treatment. Together, these results suggest that rTMS is promising strategy for treatment of AD. As a modality, rTMS is attractive given its non-invasiveness and potentially lasting neuromo-dulatory effects. Because rTMS is clinically approved for other indications, it is available with the expertise and infrastructure essential for continued rigorous, high quality research. Nonetheless, many studies employ daily treatments for weeks. Future studies should focus on larger studies, as well as optimizing patient selection and stimulation protocols.

TRANSCRANIAL ELECTRIC STIMULATION

In tES, an electrical current is applied across the skull between two or more electrodes overlying a region of interests [102]. Although the majority of the current passes via the exterior soft tissues, a portion passes through a transcranial route. In contrast to the frequently suprathreshold effects of TMS, tES exerts subthreshold electrophysiological effects on neuronal membrane excitability [71, 102]. In transcranial direct current stimulation (tDCS), a constant electrode current alters subthreshold membrane potential and therefore neuronal excitability [102]. In general, anodal tDCS, which produces a positive current, enhances neuronal excitability, while cathodal tDCS produces a negative current and reduces excitability [102]. It is worth noting that other tES protocols are common, and frequently defined by the temporal characteristics of the stimulation. These include transcranial alternating current stimulation which uses a current of alternating polarity, transcranial random noise stimulation wherein the intensity and frequency of an alternating current are randomly varied [103].

Although different electric stimulation paradigms by temporal characteristics exist, tDCS is the most common paradigm investigated in animal AD models [102, 104]. Indeed, relative to TMS there is a general paucity of tES studies in animal models of neurological and psychiatric illnesses [105]. A recent study evaluating tDCS in 3×Tg mice with cathode placement over the anterior cingulate and anode placement over the retrosplenial cortex failed to find improvements in spatial or recognition memory with 50μA stimulation, nor differences in Aβ or tau pat-hology [106]. In a rat model of AD induced by hippocampal injection of Aβ, anodal tDCS with current ≥100μA over the right frontal cortex improved spatial memory, attenuated neuroinflammation, and reduced deficits in acetylcholine synthesis [104]. Despite a lack of further evidence in AD models, tDCS has repeatedly been demonstrated to modulate plasticity, memory, and hippocampal neurogenesis in rodents [105, 107–111].

Despite limited animal evidence, randomized controlled clinical trials in tES are relatively abundant. Khedr et al. employed 10 days of left dlPFC tDCS in a sham-controlled trial (n = 34) and found significant polarity-independent improvement of cognitive scores versus sham, though with significant benefit was seen on more measures following cathodal stimulation [112]. Similarly, a small (n = 10) trial of anodal, cathodal, and sham tDCS of the bilateral temporoparietal cortex found word recognition was improved with anodal stimulation, worsened with cathodal, and unaffected by sham [113]. Another randomized, sham-controlled trial comparing 25 AD patients to 22 untreated healthy controls found only a trend toward improvement on the delayed recall after 6 sessions of left temporal tDCS, but no effect on broader cognitive scores [114]. Consistent with this, a randomized, sham-controlled crossover trial (n = 15) of bilateral anodal tDCS of the temporal cortex for one week found improved visual recognition but no effect on other cognitive scores [115]. However, a sham-controlled trial of 10 sessions of left temporal-parietal cortex anodal tDCS in a mixed cohort of 3 AD and 7 frontotemporal dementia patients found significantly improved picture naming and digit span performance versus sham at 2 weeks [116]. Conversely, two other randomized, sham-controlled trials of 2 weeks of left dlPFC tDCS with cohorts of similar size failed to find an effect on several cognitive assessments [117, 118].

Overall, although there is evidence for a potential benefit of tDCS in AD, outcomes across trials are highly variable and more work must be done to identify optimal target(s) for consistent, clinically meaningful improvements. More studies in animal models may be beneficial in refining these variables, as well as in clarifying the mechanisms by which tES might confers benefits to patients. Additionally, these results should be interpreted with caution due to the small study cohorts and potential active effects and perceivable sensation from sham stimulation paradigms [119]. A potentially more appropriate control is the flipping of anode and cathode. Nonetheless, tES may be more accessible and cost-effective than TMS, and may offer a superior safety profile due to subthreshold effects on neuronal activity (Table 1).

FOCUSED ULTRASOUND

Low-intensity focused ultrasound stimulation (FUS) is an emerging modality that employs low energy, pulsed, or continuous sonication of discrete brain regions [120]. This technique has been shown to precisely target both shallow cortical and deep brain regions with high accuracy to alter electrophysiological activity [120]. FUS can also be used to disrupt the blood-brain barrier (BBB) through interactions with intravenously injected microbubbles [121]. In AD mouse models, FUS-induced BBB opening has been shown to decrease Aβ pathology, enhance delivery of endogenous antibodies, increase glial cell activation, and improve behavioral deficits [121–123].

A study in a 5×FAD transgenic mouse model showed FUS without BBB opening may reduce Aβ pathology, reduce cerebral blood flow deficits, upregulate eNOS, enhance expression of neurotrophic factors, and reverse behavioral deficits [124]. Similarly, in an aluminum-induced rat model of AD, a similar protocol reduced Aβ peptide concentrations, attenuated acetylcholinesterase activity, reduced neuronal damage in the hippocampus, and reversed behavioral deficits [125]. Recent work has demonstrated that FUS BBB opening in the PFC of AD patients can transiently (<1 day) alter functional connectivity, although no effect on Aβ plaques was demonstrated by PET imaging [126, 127]. Additional phase II trials of FUS BBB opening in AD patients are currently underway.

Although evidence for ultrasound as a treatment of AD is limited and primarily focused on BBB opening in humans, more trials are underway. Future studies are likely to investigate this approach as a means of Aβ reduction with or without drug administration, and in the process the ability of FUS to induce neuromodulatory effects will be investigated. More work needs to be done to clarify the effects of FUS in AD both alongside and in the absence of BBB opening. However, FUS BBB opening has already shown promise for drug delivery and the ability to confer added benefits via neuromodulation could facilitate clinical translation and proliferation (Table 1).

PHOTOBIOMODULATION AND SENSORY STIMULATION

Photobiomodulation (PBM) represent a family of light-based biomodulatory therapies that employ red (620–750 nm), near infrared (NIR; 750–1400 nm), or infrared (IR) light to exert direct effects on cellular substrates, particularly mitochondrial cytochrome C at NIR wavelengths which best penetrate the human skull, stimulating cellular metabolism and producing downstream functional effects [128, 129]. Indeed, promising work with PBM has been done in the context of stroke and traumatic brain injury, with recent growing interest as a treatment for AD [130–132]. In vitro studies show PBM can reduce Aβ-induced inflammation and oxidative stress as well as modulate gene expression and prevent dendritic atrophy [133–135]. Data supporting PBM has also been generated in several murine models demonstrating a range of benefits on Aβ and tau pathology, oxidative stress and antioxidant capacity, mitochondrial respiration and mitophagy, apoptosis, and inflammation [136–141]. Importantly, these biochemical outcomes were associated with behavioral improvements in spatial and recognition memory [137, 141]. Interestingly, similar biochemical and behavioral results have also been observed when PBM was applied to the bone marrow of 5×FAD mice [142]. Such findings led to a small (n = 11) randomized, double blind, placebo-controlled trial in patients with dementia employing PBM for 6 minutes daily over 28 days [130]. Three of 6 treated individuals improved on a subset of cognitive measures including delayed word recall whereas none receiving placebo showed benefit [130]. Additionally, a case series of 5 dementia patients reported improved MMSE and ADAS-Cog scores during 12 weeks of weekly transcranial PBM plus daily intranasal PBM, as well as loss of these benefits in the weeks following cessation of treatment [143].

The clinical evidence available for sensory stimulation in AD is currently limited and poor in quality. The rationale for this based on changes in hippo-campal gamma frequency (particularly 40 Hz) power spectrum during sharp wave ripples, which is associated with memory replay and consolidation [144, 145]. Studies in several transgenic models show gamma frequency-specific effects of sensory stimulation including induction of a phagocytic phenotype in microglia, reduced Aβ and tau pathology, reduced neuroinflammation and vascular deficits, and attenuated electrophysiological changes in sensory and higher-order brain regions such as the PFC and hip-pocampus [146–148]. This was accompanied by im-provements on several measures of recognition and spatial memory following seven daily 1-hour stimulation sessions [146–148]. Although most VAS research in AD has been performed in animal models, some studies have been conducted in human patients. A trial predating the above animal results investigated the effect of 10 Hz (alpha band) visual stimulation on word recognition in 30 healthy older adults found frequency-specific enhancement of alpha band activity and recognition memory [149]. Similarly, a further case series in 3 healthy volunteers replicated the ability of 40 Hz sensory stimulation to drive neural activity at the same frequency, including the spread of this entrainment to widespread brain regions [150]. Nonetheless, a study assessing the effect of 40 Hz visual stimulation in AD patients failed to identify changes in Aβ burden as assessed by PET [151]. Still, new trials are underway in an attempt to translate the recent animal evidence into human results [152, 153].

Given the limited clinical evidence available for PBM or sensory stimulation in AD, further work should be done to characterize the effects of such app-roaches on brain activity in human patients and larger studies with standardized protocols should be conducted. Still, there is convincing evidence of beneficial effects on multiple aspects of AD pathology in diverse animal models (Table 2). The potential accessibility and cost-effectiveness of these technique as well as their effects on large regions of the brain (Table 1) make them particularly promising and suggest the potential for combination with other therapies should they demonstrate benefit in clinical trials. Nonetheless, more work should be done to find minimum necessary doses and better characterize the duration of benefits to increase likelihood of clinical uptake and patient adherence. These techniques are, after all, much less well-validated than approaches such as DBS and rTMS for which clinical approvals exist and large numbers of patients have been treated in other indications.

CONCLUSION

Neuromodulation for the treatment of AD is an emerging field experiencing rapid growth. New methods of neuromodulation have shown promising results in animal models of AD, and to date, human trials have been shown to be safe and to have potential benefits at diverse brain targets. Larger, long-term studies are now needed to ensure the safety and effectiveness of these procedures. More work will also be needed to optimize treatment protocols and identify clinical subpopulations that will benefit from a given approach. Nonetheless, benefits have been seen on a range of pathological and clinical endpoints, some persisting long after treatment. Such outcomes make neuromodulation strategies attractive alternative approaches to the traditional methods of managing AD.

DISCLOSURE STATEMENT

Authors’ disclosures available online (https://www.j-alz.com/manuscript-disclosures/20-0913r1).