Clinical Approach of Low-Dose Whole-Brain Ionizing Radiation Treatment in Alzheimer’s Disease Dementia Patients¹

INTRODUCTON

For a century, ionizing radiation has played an important role in treating not only solid cancers but also hematologic cancers. Recently, the possibility of deploying radiation treatment for benign diseases has been suggested and the range of radiation therapy indications is expected to be expanded [1]. Due to the growing aging population, the number of patients with degenerative brain diseases has significantly increased; however, a cure remains unavailable [2]. Recently, the results of preclinical studies have shown that low-dose irradiation (LDIR), a nonpharmacological treatment method, may be a new treatment option for patients with Alzheimer’s disease dementia (AD) [3, 4].

Generally, high-dose radiation causes inflamma-tion and cancer in the body. In particular, palliative whole-brain radiation therapy in patients with meta-static brain tumors reportedly results in diminished memory and cognitive function [5]. Consequently, the application of brain radiation therapy in patients with AD was not considered as a viable treatment option. However, if the radiation dose administered to the hippocampus during whole-brain radiotherapy for patients with brain metastasis were to be reduced below a certain level (median dose <9 Gy), the decline in the patient’s cognitive function and memory would be less significant [6]. In 2016, a decrease in the amyloid-β (Aβ) protein concentration after hemibrain irradiation was reported at Beaumont Hospital [7]. Further, brain radiation therapy was shown to reduce tau-associated neurofibrillary tangles as well as Aβ in 3xTg-AD mice [8].

Based on these preclinical results, our hospital research team is conducting a phase II study (NCT04203121) to examine the efficacy of low-dose brain irradiation in patients with early or mid-stage AD, which is also being investigated by other insti-tutions including the Virginia Commonwealth Uni-versity (NCT02769000), University of Geneva (NCT03352258), and Beaumont Hospital (NCT02359864).

The first two reports about the mode of action for the neuroprotective effects of LDIR in our research laboratory have been made available to date [9, 10] and suggest this method could be a nonpharmacological treatment option for AD patients. In addition, analyzing changes over time after LDIR would help to establish a protocol for confirming treatment re-sponse in future clinical studies.

DETERMINING WHETHER LDIR DIRECTLY OR INDIRECTLY REDUCES THE Aβ PROTEIN CONCENTRATION IN AD

The classical effect of high-dose low linear energy transfer X-ray irradiation is the formation of hydroxyl radicals through radiolysis of water. The hydroxyl radical is primarily responsible for causing double strand breaks in DNA leading to cell death. One study has reported no effect of high dose radiation on Aβ_1–42 fibrils or the formation of these fibrils in vitro [11].

In the first experiment, LDIR [9 Gy/5 fractions (fx); biologically effective dose; BED₂ = 17.1 Gy₂ ] was administered to whole brains of 4-month-old mice, who were euthanized after four days to examine the tissue; at this point, a reduction in the Aβ protein concentration was not observed [9]. The absence of effect of LDIR four days after treatment suggests on the contrary that the radiations do not directly impact amyloid but that it is an indirect effect which may take time to be effective. In addition, synaptic degeneration, neuronal loss, and neuroinflammation were significantly suppressed in the hippocampus and cerebral cortex compared with the group not treated with LDIR. To determine the cytoprotective effects against Aβ toxic substances, SH-SY5Y neu-ronal cells were treated with Aβ_1–42 (2μM), followed by LDIR (1 Gy). Based on the MTT assay and DNA-fragmentation test results, LDIR successfully protected nerve cells. Moreover, the production of proinflammatory molecules and the activation of the nuclear factor-kappa B (NF-κB) pathway was suppressed in BV-2 microglial cells after LDIR (Fig. 1).

Fig. 1

Schematic diagram of intra cranial change according to timing after LDIR.

In our second paper, whole-brain LDIR was administered to 6-month-old 5xFAD mice [10]. After eight weeks, cognitive tests were performed, while the brain tissues were examined at 10 weeks after the mice were sacrificed (Fig. 2).

Fig. 2

Effect of LDIR on Aβ deposition in the subiculum and cerebral cortex of 5XFAD mice. A) Schematic diagram of the experimental procedure. 5XFAD mice were exposed to sham or LMDIR (9 Gy in five fractions), and then sacrificed 4 days after the final irradiation. B) Histochemical staining for thioflavin S was performed to identify amyloid plaques in the cerebral cortex and hippocampal dentate gyrus (DG) of the sham- and LDIR-treated 5XFAD mice (scale bar = 250 m). C) Aβ-positive areas in the cerebral cortex and dentate gyrus of the sham- and LDIR-treated 5XFAD mice were quantified and plotted as a percentage. D) Representative of swimming paths was recorded on videotape during the test session of the Morris water maze. E) Latency to escape time of the mice was measured. Repeated-measures two-way ANOVA with Fisher’s post hoc test compared to 5XFAD+Sham mice. F) Measurement of quadrant time to spend in the hidden platform. Data are presented as mean SD (n = 10 mice in sham-treated 5XFAD mice and n = 14 mice in LDIR-treated 5XFAD mice).

In cognitive tests, the LDIR-treated mice showed significantly improved learning and memory skills compared with the sham group. In addition, Aβ was significantly decreased in brain tissue. Furthermore, the proinflammatory cytokine tumor necrosis factor-α (TNF-α) was decreased and the anti-inflammatory cytokine transforming growth factor-β (TGF-β) was increased in the brain tissue of 5xFAD mice treated with LDIR. After LPS-stimulated BV-2 microglial cells were treated with LDIR (1 Gy), the phenotype of these cells converted to the M2 phenotype by M1. Moreover, the proinflammatory cytokine markers [TNF-α, interleukin (IL)-1, and IL-6] were decreased and anti-inflammatory markers such as M2 cytokine markers (TGF-β, TGF-α, and IL-10) were increased. Furthermore, LDIR significantly increased TREM2 expression in the membranes of BV-2 microglial cells (Fig. 1).

In summary, the therapeutic effect is concerned by the indirect impact of immune cell phenotype regulation along with inhibition of the NF-κB pathway and anti-inflammatory cytokines rather than the direct effect of LDIR reducing Aβ protein levels and tau-associated neurofibrillary tangles [12].

LDIR REGULATES THE PHENOTYPE OF MICROGLIAL CELLS

Microglial cells play an important role in the phagocytosis of neuronal debris and synapse pruning to maintain homeostasis in the brain microenvironment [13, 14]. In a recent report, the regulation of microglial cells was suggested as a potential treatment method for AD [15]. Based on results obtained by our research team, immediately after LDIR, microglial cells in the Tg mouse brain showed red-uced microgliosis and neuronal loss, indicating a neuroprotective effect of LDIR. When the M2 type was replaced, Aβ plaque levels were decreased due to an increase in Aβ phagocytosis, while synaptic pruning and synaptic plasticity activation may eventually lead to increased cognitive function [10].

Meanwhile, the long-term effects of LDIR have been reported in another study; specifically, the number of microglial cells was increased at four weeks after irradiation with 10 Gy/5 fx and further increased after eight weeks [3].

In this experiment, neurogenesis was not observed in the brain tissue immediately after LDIR in a 4-month-old dementia mice model, while neurogenesis remained unconfirmed even after nine weeks. However, the possibility of neurogenesis in 6-month-old dementia mice could not be excluded.

Summarizing the results of the above mentioned experiments, rather than decreasing Aβ directly with LDIR, inhibition for the production of pro-inflam-matory molecules and the activation of the NF-κB pathway were associated with reduction of Aβ toxicity immediately after LDIR. These results indicate the neuroprotective effects are due to the suppression of Aβ and, by controlling the phagocytosis of abnormal proteins in brain tissue, synapse formation, and phenotype of microglia involved in synaptic pruning, an improvement in cognitive function and a decrease in the Aβ protein concentration after at least eight weeks were observed.

IDENTIFYING THE BEST STAGE OF AD IN WHICH TO PURSUE LOW-DOSE RADIATION THERAPY

The research on LDIR has shown the triggering of indirect effects is the main mechanism of action, such as the inhibition of proinflammatory cytokines, increased anti-inflammatory cytokine release, and microglia phenotype regulation caused by the inhibition of the NF-kβ pathway, rather than the direct degradation of Aβ proteins. Therefore, the stage of AD in which neurodegeneration has progressed only minimally may be considered the best time to perform LDIR [16] (Fig. 3).

Fig. 3

Schematic representation of changes in cognitive, metabolic, structural and molecular pathogenetic parameters in relation to estimated years to symptomatic onset of dominantly inherited AD [17].

As shown in Table 1, the four study subjects were primarily in the early stages of dementia. However, based on the study results, LDIR was performed in a prodromal stage in high-risk patients before serious neurological damage, during asymptomatic periods, and during mild cognitive impairment, which is expected to delay the onset of symptoms in patients with AD and dementia and contribute to breakthrough development in the treatment of AD.

Unfortunately, we could not confirm neurogenesis after LDIR in two experiments, but it is difficult to completely eliminate the possibility of neurogenesis when we can clearly observe an increase in cognitive function at eight weeks after low-dose brain irradiation. Since clinical research is ongoing, it is not possible to confirm the improvement of cognitive function, and LDIR has been reported to improve cognitive function in the animal model (5XFAD model).

Therefore, it can be suggested that it is even effective to administer radiation therapy in the prodromal stage or early stage where irreversible neurodegeneration is relatively minor due to nerve damage caused by Aβ proteins and tau-associated neurofibrillary tangles.

DETERMINATION OF TOTAL RADIATION DOSE, FRACTIONATION SIZE, AND INTERFRACTIONAL INTERVAL

CONCLUSIONS

To date, four clinical studies on the role of LDIR in whole-brain radiation therapy for patients with early or mid-stage AD have been reported; however, definitive clinical results have not yet been published. As such, despite the promising results of these preclinical studies, radiotherapy can be applied to patients with AD only when clinical results are made available. In addition, in the case of low-dose radiation therapy, additional large-scale clinical studies are needed to determine the severity of AD, indications for radiation therapy, the total dose to be irradiated, fraction size, and intervals of radiation therapy.

If low-dose radiation therapy is used to treat AD, it is expected to become a breakthrough treatment method at a relatively low cost and short treatment period compared to the cost of drug development and treatment and treatment period. This treatment could play a significant role in reducing social and national medical costs and improving the lives of AD patients and their caregivers as well as of patients with other diseases.