Combining Low-Dose Radiation Therapy and Magnetic Resonance Guided Focused Ultrasound to Reduce Amyloid-β Deposition in Alzheimer’s Disease

Alzheimer’s disease (AD) is among the leading causes of death in the United States and Europe, and there is an urgent need of disease-modifying treatments. Amyloid-β deposition is one of the neuropathological hallmarks of AD and many drugs in the therapeutic pipeline are aimed to its reduction [1], but only aducanumab has been controversially approved in the United States, as FDA calls for federal investigation into this drug [2].

Two physical health technology approaches, namely low dose radiation therapy (LD-RT) and magnetic resonance guided focused ultrasound (MRgFUS) have been proposed as potential treatments in AD.

LD-RT by whole brain irradiation (WBI) with 10– 20 Gy, which is less than half of the dose typically delivered in brain cancer, has been demonstrated to reduce amyloid-β and tau and improve cognitive function in AD animal models [3, 4]. Despite the mechanisms of LD-RT in AD are not completely understood [5], these preclinical studies paved the way to ongoing clinical trials in AD patients [6].

Converging evidence in animal models and proof-of-concept preliminary data in humans support the view that MRgFUS in combination with microbubbles can safely and reversibly induce blood-brain barrier (BBB) opening [7], facilitate the delivery of targeted therapeutics to the hippocampus, reduce amyloid, and promote neurogenesis in AD [8, 9].

Limitations and drawbacks of these two health technologies in AD are not yet fully understood, as their investigation is still in a preliminary phase. Table 1 summarizes the physical features, side effects, and experimental evidence of MRgFUS-induced BBB opening and LD-RT. We will briefly discuss open questions and the potential combination of MRgFUS and LD-RT to reduce amyloid-β.

Any potential benefit of LD-RT in patients with AD should be weighed against possible RT-induced neurotoxicity, which could induce further cognitive deficits. Evidence on cognitive side effects of RT derives from patients with brain tumors [10], who may behave differently from AD patients, who in turn, may be more sensitive to the radiation-induced damage. Single nucleotide polymorphisms in the APOE gene, and in genes involved in aging, inflammation, DNA repair and response to oxidative stress may influence cognitive outcomes to RT [11]. Whether genetic variants might increase cognitive side effects of RT in AD is unknown. Cerebral microbleeds are common in AD [12], particularly in patients carrying mutations for dominantly inherited AD [13], and represent a marker of RT-induced vascular brain damage [14]. The brain of AD patients might thus have increased sensitivity to RT vascular damage. At the higher dose delivered for cancer, radiation can cause microangiopathy, with consequent cognitive deterioration. More studies are needed to understand the possible damage to vascular supply and to verify the safety of low-dose RT protocols in AD.

Hippocampal damage and reduced neurogenesis play a key role in the cognitive decline induced by brain RT [15]. Retrospective and prospective studies in brain cancer patients documented that 7.3– 10 Gy RT doses administered to the hippocampus correlated to cognitive impairment [10, 16]. These values would be largely exceeded in ongoing AD clinical trials, which will deliver 10– 20 Gy through WBI and might not represent the best trade-off between adequate amyloid reduction in the hippocampus and absence of RT-related cognitive side effects [6]. Moreover, white matter (WM) lesions are common in the aging brain and are often associated with cerebral small vessel disease. WM is a radiosensitive tissue involved in cognitive decline following brain RT [17], and its irradiation may be avoided in AD, since pathological amyloid-β deposition mainly involves the cortical region, while WM abnormalities are common [18, 19], and may represent an early neuropathological event in AD [20].

MRgFUS preclinical work has paved the way to human trials exploring BBB opening and focal drug delivery in AD, but there are concerns on the induction of neuro-inflammation, which may potentially worsen AD, following treatment [21]. Ongoing clinical trials are exploring MRgFUS for BBB opening for local delivery of targeted drugs to the hippocampus. Considering that abnormal deposition of amyloid-β plaques is not limited to the hippocampal regions, focal treatment of the hippocampus may not be sufficient to reduce amyloid-β in other cortical regions. Applying MRgFUS to open the BBB to large brain regions affected by amyloid deposition might be complex [22], time consuming, expensive, and yield neuroinflammation-related side effects [21]. Moreover, adjacency to the skull and no pass zones related to skull sinuses and calcifications may hamper adequate sonication of some cortical areas.

To overcome the limitations of LD-RT and MRgFUS, these health technologies may be combined. Multidisciplinary management is common in other medical fields, e.g., radiation oncology, where radiotherapy may be delivered in combination with surgery, chemotherapy, and hyperthermia by means of focused ultrasound [23]. In analogy, combined treatment might be a reasonable approach to AD treatment. In details, the neurotoxicity of RT may be more marked for WBI protocols, but the application of advanced radiation techniques might reduce the dose to critical brain structures involved in the cognitive deficits, such as the hippocampus and the WM. Hippocampal sparing WBI techniques may reduce cognitive deficits with a dose– response relationship [24] in patients with brain cancer and might be a reasonable approach in AD. At the same time, hippocampal sparing RT techniques may be difficult to apply in AD, because AD neuropathology starts from the transentorhinal and entorhinal regions in the first stages of the disease and then involves the adjacent and functionally related hippocampus [25]. WM sparing might be a complex task even with the most advanced radiation techniques and only a moderate sparing might be obtained without reducing the dose to the adjacent cortical tissue.

LD-RT focused on cortical areas more positive to amyloid PET tracer, excluding hippocampal and other deep brain regions, would allow direct sparing of the cortical tissue with a low amyloid PET uptake and indirect sparing of WM, because of the reduced extension of the irradiated volume. Complementarily, BBB opening through MRgFUS should be focused to the entorhinal, hippocampal, and deep brain regions combined with drugs to reduce amyloid deposition, tau pathology and neuronal damage [26], thus reducing side effects related to systemic delivery. The increasing availability of MRgFUS in a larger number of centers worldwide [8] might pave the way for future clinical trials on combined LD-RT and MRgFUS protocols.

Finally, amyloid deposition is not the only neuropathological feature of AD, and strategies aimed to reduce tau pathology might represent a more effective strategy. Amyloid deposition and tau pathology are hypothesized to occur early in AD [26] and decreasing amyloid and tau concentration may not change the clinical course. Combined therapeutics strategies additionally targeting neuroinflammation [6] and neuronal damage, according to the ATN pathological framework of AD [26] might offer added value for future clinical trials, which are needed to support the hypotheses discussed in this paper.