Can transcranial magnetic stimulation rescue dopaminergic signalling in Alzheimer's disease?

The precuneus and other areas of the posteromedial parietal cortex (PMC), synergizing with the frontal cortex, are primarily engaged during tasks that involve episodic memory, visuospatial working memory and visuomotor abilities. Early deficits, including episodic memory, are widely recognized in patients who suffer from mild cognitive impairment (MCI) due to Alzheimer disease (AD) and, in particular, the amnestic variant, (a-MCI).

Recent research has shown that intrinsic brain activity, as observed by functional MRI (fMRI), manifests as coherent signal changes in networks, including brain regions that spread extended neuronal pathways. One such network is the default mode network (DMN), a brain system that is anchored in the posteromedial cortex. The DMN has become the primary target in recent studies that have linked intrinsic activity to cognition and examined how changes in intrinsic signals are altered in disease. Based on these results, hypometabolism in the posteromedial parietal cortex, with reference to the posterior cingulate cortex and precuneus, has been proposed as a valid biomarker of progression to dementia in patients with MCI.¹

We have read with great interest the randomized phase II clinical trial by Koch and colleagues² that determined the efficacy of repetitive transcranial magnetic stimulation (rTMS) on the precuneus in patients with mild-to-moderate AD. The authors recruited 50 AD patients and randomly assigned them 1:1 to sham or real rTMS.

Their stimulation protocol consisted of 2 weeks of intensive treatment (5 days/week), followed by a once-per-week session for an additional 22 consecutive weeks, with real or sham rTMS pulses applied at 20 Hz in 2-s trains with 28 s intertrain intervals, resulting in 1600 pulses per session. The primary outcome was the change in Clinical Dementia Rating Scale-Sum of Boxes (CDR-SB) scores at 24 weeks from baseline; the secondary outcomes were Alzheimer’s Disease Assessment Scale-Cognitive Subscale (ADAS-Cog), Mini-Mental State Examination (MMSE), Alzheimer’s Disease Cooperative Study-Activities of Daily Living (ADCS-ADL), Frontal Assessment Battery (FAB), and Neuropsychiatric Inventory scores.

Notably, the intensity and positioning of rTMS delivery were established using single-pulse TMS, in combination with a 64-channel EEG (TMS-EEG) system, based on the evaluation of TMS-evoked potentials, to monitor the effect of rTMS treatment on cortical activity.

The efficacy of the treatment was rated by comparing the baseline evaluation with those at Weeks 12 and 24 (end of the clinical study). The results showed a significant difference between the real and sham rTMS groups at Week 24, with a power of approximately 90%.

Beyond the meaningful clinical changes that were observed in the neuropsychological assessments, the authors found that rTMS induced significant changes in cortical activity, based on TMS-EEG recordings. Specifically, cortical excitability remained unchanged after 24 weeks in the rTMS group but declined dramatically in sham rTMS recipients. Further, local gamma band oscillations were boosted only in patients who received real rTMS.

Several previous studies have evaluated the therapeutic effects of rTMS in AD (for a review, see Di Lazzaro et al.³), in which the most extensively examined target was the dorsolateral prefrontal cortex, due to its relevance in cognitive functions, such as attention, executive functions, and working memory. The results of these studies were promising, but the efficacy and durability were limited. The recent study by Koch and colleagues² provides the first convincing evidence of the possibility of slowing cognitive decline in AD with prolonged rTMS treatment.

The rationale of the approach that was adopted by Koch and colleagues² is based on the decision to target posterior brain areas where the neurodegeneration appears to develop and, due to decreased functional connectivity, damage several structures of the default mode network. A previous study by the same group showed that precuneus stimulation modulates long-term memory and strengthens the connectivity between this area and the temporal cortex in normal subjects. Thus, the aim is to promote changes in posterior brain areas that enhance the connectivity with interconnected structures, improving cognitive function and, in particular, memory. There is a large body of evidence, including functional magnetic resonance imaging studies, that show that rTMS induces neuroplastic effects in cortical areas that are directly stimulated and in remote brain regions that are functionally connected to the stimulated area. However, because the effects of rTMS are transient, it is unknown how effective rTMS can be in slowing the progression of cognitive decline in AD.

Experimental in vitro and in vivo studies have reported that with these temporary changes, repetitive magnetic stimulation can elicit long-term after-effects that are related to the: (i) induction of immediate early genes (c-fos, zif268, Arc, BDNF); (ii) modulation of AMPA receptor trafficking through NMDA receptor activation; (iii) remodelling of GABA receptors; and (iv) modulation of neurotransmitter synthesis and release (for a review, see Cirillo et al.⁴). But, what are the long-lasting effects of precuneus rTMS that are relevant to its therapeutic potential in AD?

Previous studies⁵ have demonstrated that protocols for stimulation that enhance cortical excitability [high-frequency (HF) rTMS or intermittent theta burst stimulation], delivered over the left dorsolateral prefrontal cortex, are highly effective in treating depression; promising results have been also been reported for such psychiatric conditions as schizophrenia. The authors of these studies hypothesize that rTMS is likely to be an effective treatment modality for negative symptoms, because rTMS stimulation can activate the prefrontal cortex by modulating dopamine activity directly in the prefrontal cortex and through interconnections in remote brain areas.

Today, we wonder whether the results of Koch and colleagues² depend on modulation of the mesocorticolimbic dopaminergic circuitry, as mediated by prolonged rTMS treatment. Our consideration stems from recent experimental observations in a validated mouse model of AD⁶ that, in conjunction with subsequent studies in humans, demonstrated a crucial function of dopamine and the dopaminergic midbrain in the prodromal phase of AD.

Actually, histological studies have proven that early anatomical changes arise in the ventral tegmental area (VTA), a subcortical brain region that is rich in dopaminergic neurons. Specifically, neuronal loss occurs in this area prior to any deposition of amyloid-β plaques or alterations in hippocampal synaptic plasticity. Moreover, the VTA degeneration is accompanied by reduced dopaminergic innervation to the mesocorticolimbic pathway and deteriorations in memory performance.

Anatomically, the ascending dopamine system is distributed widely across brain regions, including core areas of the default mode network. The VTA and substantia nigra pars compacta (SNc)—the primary brain centres that are involved in dopamine synthesis—are the respective origins of the mesocorticolimbic dopaminergic and nigrostriatal systems. The mesocorticolimbic dopaminergic system projects from the VTA to the nucleus accumbens (NAc) and cortical and limbic regions, including the medial prefrontal cortex, hippocampus and amygdala.

Consistent with analyses in AD mouse brains, subsequent studies confirmed the involvement of the VTA in the early stages of AD. By resting-state functional MRI, patients with MCI presented with early and progressive functional disconnection between the VTA and its projecting areas, which worsens in AD.⁷ Further, a structural MRI study demonstrated a strict correlation between the volume of the VTA and memory performance, suggesting that changes in the integrity of the VTA constitute an early marker of future neurodegeneration.⁸ More recently, two studies focused on the function of the dopaminergic system in early AD. An in vivo SPECT imaging study⁹ measured the binding of ¹²³I-FP-CIT to dopamine transporter (DAT) in patients with MCI due to AD and in AD patients. As a result, there was more widespread loss of interconnections between subcortical and cortical targets of the mesocorticolimbic pathway in AD patients and MCI patients, whereas the nigrostriatal connections were preserved. Most importantly, alterations in the mesocorticolimbic dopaminergic circuitry were already present in the prodromal disease phase.

Finally, a notable fMRI study¹⁰ tested the hypothesis that VTA functional disconnection in aMCI patients is associated with a higher risk of conversion to AD. Incidentally, MCI converters experienced a meaningful reduction in functional connectivity in the posterior cingulate cortex (BA31), which includes the precuneus, for which rTMS stimulation might slow cognitive and functional declines in AD, as Koch and colleagues have demonstrated.²

As discussed, the most widely accepted mechanism for the long-term neural effects of HF-rTMS is that it alters synaptic plasticity—primarily the long-term potentiation of excitatory synaptic transmission—through various molecular pathways.

A growing body of experimental and clinical evidence has demonstrated that dopamine signals modulate many, if not all, of these neuronal activities, rendering rTMS a potentially valuable therapeutic tool for mitigating the impairments in DA release in the mesolimbic dopaminergic system during disease and decreasing the severity of accompanying clinical symptoms. In conclusion, previous findings are useful in partially explaining the results of Koch and colleagues,² collectively forming a springboard for further research that establishes the precision of clinical diagnosis and the efficacy of dementia therapeutics; plans clinical trials that aim to restore dopaminergic tones that could be altered in the prodromal phase of AD; and guides the development of therapeutic interventions that protect dopaminergic function and thus slow disease progression.

 

Data availability

Data sharing is not applicable to this article as no new data were created or analysed in this study.

Funding

M.D.A. and V.D.L. were supported by Fondazione Roma (Rome, Italy). M.D.A. was supported by the American Alzheimer’s Association [AARG-18-566270; AARG-21-851219], by the Italian Ministry of Health [Research Grant: RF-2018-12365527].

Competing interests

The authors report no competing interests.