Composition for the prevention or treatment of cognitive impairment and method for manufacturing the same

Abstract

To provide a composition useful for preventing or treating cognitive impairment due to reduced hippocampal neurogenesis, and a manufacturing method therefor.SOLUTION: A composition for preventing or treating cognitive impairment is combined with high air pressure oxygen therapy and includes memantine.SELECTED DRAWING: None

Description

Technical Field

[0001] The present invention relates to a composition for preventing or treating cognitive dysfunction, which is used together with hyperbaric oxygen therapy, and a method for producing the same.

Background Art

[0002] Conventionally, it has been known that the toxicity to the central nervous system caused by cancer treatments such as radiation therapy and chemotherapy induces cognitive dysfunction by causing hippocampal toxicity, thereby reducing the quality of life of patients. This is considered to be mainly due to the decrease in hippocampal neurogenic ability caused by toxicity to the central nervous system.

[0003] For example, Patent Documents 1 and 2 describe compositions for hippocampal neurogenesis containing nucleosides, deoxyribonucleotides, amino acids, and the like.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Patent Document 3

Non-Patent Documents

[0005]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, to date, no effective preventive or therapeutic methods have been established for cognitive impairment caused by a decline in hippocampal neurogenesis.

[0007] This invention has been made in view of the above circumstances, and aims to provide a composition effective for preventing and treating cognitive impairment caused by a decrease in hippocampal neurogenesis, a method for producing the same, and a new treatment method. [Means for solving the problem]

[0008] To solve the above problems, the present invention provides a composition for the prevention or treatment of cognitive impairment, to be used in conjunction with hyperbaric oxygen therapy, and is characterized by containing memantine.

[0009] The present invention relates to a method for producing a composition for the prevention or treatment of cognitive impairment, to be used in conjunction with hyperbaric oxygen therapy, and is characterized by comprising the step of incorporating memantine. [Effects of the Invention]

[0010] According to the composition of the present invention, cognitive impairment caused by a decrease in hippocampal neurogenesis can be prevented or treated. According to the method for producing the composition of the present invention, the said composition can be produced. [Brief explanation of the drawing]

[0011] [Figure 1] This figure shows examples of HBO and radiotherapy devices applied to humans and rodents. [Figure 2] This figure shows the immunohistochemical analysis of DCX in mouse DG in the following groups: control, RT (Radiation Therapy), HBO, RT + memantine, and RT + memantine + HBO. [Figure 3] This figure shows the bromodeoxyuridine staining of the hippocampus of mice from each group (n=4). [Figure 4] Golgi-stained cells in the subgranular layer of hippocampal DG in the control, RT, RT+memantine (M), and RT+HBO groups. The scale bars indicate 2,50, and 200 μm, and the images are shown at high, medium, and low magnification. The lower right column shows the number of shin-, stubby-, and mushroom-type spines per 10 μm for each group (total: "Number of thin + stubby + mushroom-type spine / 10 μm". *p<0.05, **p<0.01.) [Figure 5] (A) is a figure showing the immunohistochemical expression of NF-200kDa in DG for each group: control, RT, RT+memantine, and RT+HBO. (B) is a figure showing the immunohistochemical expression of MBP in mouse brain sections indicated by the white squares (A) for each group. (C) is a figure showing a high-magnification image of the area indicated by the white frame in panel a. [Figure 6] The figure on the right shows a semi-quantitative analysis of MBP expression in each treatment group, and the figure in the center shows morphological measurements of axonal length. "Axonal length (b / a)" is defined as "length of the axon (b) / maximum radius of the cell body (a)". [Figure 7] The upper figure shows the staining of the Golgi apparatus in the posterior splenic dysplastic cortex of mice from each treatment group. The area enclosed in the red box is shown magnified from left to right for each treatment group. The lower figure shows the number of thin-, stubby-, and mushroom-type spines / 10μm in each group (total: “Number of thin+stubby+mushroom-type spine / 10μm” *** p<0.001, **p<0.01, *p<0.05). [Figure 8] This figure shows the results of evaluating anxiety-like behavior using the elevated cruciform maze test (n=6). [Figure 9] This figure shows the results of a fear conditioning test. The freezing response was measured one day after pre-exposure (control group, RT group, RT+memantine group (n=12), RT+HBO group, RT+HBO+memantine group (n=6)). [Figure 10]This is a diagram showing an example in which the angle between the sample arm and the reward arm was changed using the DNMP protocol of a radial arm maze to verify pattern separation ability (S: start-arm, R: reward-arm). The sample arm and the reward arm were separated at 45°, 90°, and 135°. [Figure 11] This is a diagram showing the results of verifying pattern separation ability using the DNMP protocol (correct response rate of selective responses at each angle (n = 12)). [Figure 12] This is a diagram showing the exploration times of novel objects and familiar objects for each group. [Figure 13] This is a diagram showing the average correct response rates for the memory tasks of the hippocampus with the presentation of New, Lure, and Same. From left to right, it shows healthy volunteers, brain tumor patients: non-irradiated, before intracranial irradiation, and after intracranial irradiation. The black circles represent individual data. ** p < 0.01. [Figure 14] This is a diagram showing the relationship between the correct response rate in the Lure task and the radiation dose of the right hippocampus (left figure) and the left hippocampus (right figure). [Figure 15] Each panel shows, in order from the left, the BOLD response curve, activation map, %BOLD signal change, and correlation between %BOLD signals of brain tumor patients after brain irradiation. Also, in order from the top, it shows healthy volunteers (n = 36), brain tumor patients in the non-irradiated group, brain tumor patients before brain irradiation, and brain tumor patients after brain irradiation. The correlation coefficient (r) between the %BOLD signal and the correct response rate in the Lure task was: healthy volunteers: 0.55 (p = 0.0006), brain tumor patients in the non-irradiated group: 0.49 (p = 0.01), brain tumor patients before brain irradiation: 0.22 (p > 0.05 = 0.54), brain tumor patients after brain irradiation: -0.05 p = 0.86). [Figure 16] This is a diagram of axial, coronal, and sagittal sections showing activated or inactivated spots (the voxel-level threshold was p < 0.001 uncorrected, and the cluster-level threshold was corrected by FWE to p < 0.05). The right side of the image shows the left hemisphere of the participant. The colors indicate the following small regions of the hippocampus: DG, red; CA3, yellow; CA1, green; hippocampus, blue; EC, light blue; perirhinal cortex, color bar: T value. [Figure 17] This is a figure showing the neuropsychological test scores (MMSE, TMT B-A, DS, DST) based on four scales. The patients were divided into three groups according to the presence or absence of HBO or memantine treatment. *p<0.05. DS: digit span, DST: digit symbol test, HBO: hyperbaric oxygen, MMSE: mini mental state examination, TMT B-A: trail-making test A and B. [Figure 18] This is a graph showing the time course of scores related to the Lure task and neuropsychological tests. The closed circles indicate the correct response rates of the behavioral fMRI tasks before radiotherapy (Pre), during radiotherapy (During), and after radiotherapy (Post). The open circles indicate the time course of the following scores. 3MS (case 3, case 1), HDS-R (case 3), MMSE (case 2, case 8, case 3), TMT A (case 3, case 8), TMT B (case 9, case 8), TMT B-A (case 9, case 4), stroop reading (I), (case 4, case 3), stroop naming (II) (case 2, case 9), stroop interference (III) (case 3, case 6), stroop III-II (case 1, case 3), digit span (case 1, case 3), and DST (case 2, case 1). [Figure 19]This figure shows changes in the BOLD response in the dentate gyrus and memory task scores in the hippocampus during chemoradiotherapy. Panel (A) shows details of chemoradiotherapy in a 62-year-old female patient with cerebral glioma, the corresponding time course of the BOLD response in the dentate gyrus (DG), and memory task scores in the hippocampus. The patient received cranial radiotherapy (total dose 56 Gy), chemotherapy, and hyperbaric oxygen therapy after surgery to remove a brain tumor. The graph shows the percentage change in the signal of the BOLD response in the DG during treatment. The values ​​in parentheses are the radiation dose to the hippocampus. At 14 Gy, the percentage change in the signal of the BOLD response in the DG decreased, and the Lure task score became zero. At 14 Gy, memantine (5 mg / day) was started. A positive peak appeared in the BOLD response curve, and at 56 Gy, the Lure task score recovered to normal. Chemotherapy with temozolomide was administered monthly. Following the initiation of this treatment, the BOLD response showed a negative pattern, and the Lure score decreased to less than 2 standard deviations (SD) of the normal range. atm, atmospheric pressure; Fr, fraction; HBO, hyperbaric oxygen; TMZ, temozolomide. (B) Panel shows details of radiotherapy and HBO therapy with the time course of the DG BOLD response and hippocampal memory task score in a 61-year-old female patient with atypical meningioma. The patient underwent surgery to remove the brain tumor after receiving cranial radiotherapy (total dose 46 Gy) and hyperbaric oxygen therapy. At 14 Gy, the percentage of signal change in the DG BOLD response decreased, and the Lure task score decreased to zero. Post-irradiation, the BOLD response curve improved, and the Lure task score returned to normal. [Figure 20] This figure shows the neuropsychological test scores of patients undergoing radiation therapy. Panels A and B represent the same cases as those shown in Figures 19A and 19B, respectively. [Figure 21]This figure shows four indicators (FA, MD, AD, RD) extracted from pre-operative (left) and post-irradiation (right) DTI data of patients classified into three groups based on whether or not they received HBO or memantine therapy. Each light bar represents pre-irradiation data, and each dark bar represents post-irradiation data. Group that received neither HBO nor memantine (n=3). Group that received only HBO therapy (n=9). Group that received both HBO and memantine therapy (n=8). *p<0.05. AD: Axial diffusion coefficient, ATR: Anterior thalamic radiation, CNG: Cingulate gyrus, CST: Corticospinal tract, DTI: Diffusion tensor image, FA: Fractional anisotropy, IFOF: Inferior fronto-occipital fasciculus, ILF: Inferior longitudinal fasciculus, MD: Mean diffusion coefficient, RD: Radial diffusion coefficient, SLF: Superior longitudinal fasciculus, UNC: Uncinate fasciculus. [Figure 22] This figure shows the effects of memantine and hyperbaric oxygen therapy on the integrity of the ultrastructure of white matter bundles in a representative irradiated patient. A: Right lateral view of the white matter bundles before treatment in case No. 13. The colored DTTs correspond to seven specific bundles as follows: ATR: orange, CST: green, CNG: vermilion, IFOF: light yellow, ILF: dark blue, SLF: yellow, UNC: light blue. B: The panel shows the white matter bundles after irradiation. ATR appears after treatment (indicated by the white arrowhead). C: Enlarged view of the dashed white circle area in A (left) and B (right). A dense bundle representing the IFOF after irradiation is shown (right). D: Fusion of the dose distribution map of the irradiation therapy and T1-weighted gadolinium-weighted coronary MRI. The irradiated right cerebral hemisphere shows values ​​from 27.5 Gy (light blue) to 50.0 Gy (yellow), and the white arrow indicates the IFOF region in C. E: This figure shows the mean FA values ​​and fiber counts for each of the seven bundles before (left) and after (right) treatment. FA values ​​remained constant or increased, with the exception of ATR values. The fiber counts for ATR and IFOF increased significantly (ATR: 2 to 558, IFOF: 71 to 381). ATR: Anterior thalamic radiation; CNG: Cingulate gyrus; CST: Corticospinal tract; DTT: Diffusion tensor tractography; HBO: Hyperbaric oxygen; IFOF: Inferior fronto-occipital fasciculus; ILF: Inferior longitudinal fasciculus; SLF: Superior longitudinal fasciculus; UNC: Uncinate fasciculus. [Figure 23]This figure shows the effects of memantine and hyperbaric oxygen therapy on the main direction of diffusion within each voxel in irradiated cases. A: This is a right sagittal view of the main scanning direction of the preoperative color FA image of case No. 13. The colored FA image shows the main direction of diffusion within different white matter pathways. Red, green, and blue indicate the horizontal, vertical, and perpendicular directions, respectively. The inset is a T1-weighted sagittal section, showing the same slice section as the color FA image. B: The panel is a postoperative color FA image. The white arrow indicates irradiation to the anterior thalamus. C, D: These are enlarged views of the areas enclosed by white rectangles in panels A (left) and B (right). While the green-colored areas were sparsely visible in the preoperative view, the postoperative views (B and D) show green-colored FA images from anterior to posterior direction. E, F: These are bar graphs showing the mean and SD of each tensor in the areas shown in C and D. xx, yy, and zz in the graphs represent the horizontal, anterior-posterior, and vertical tensors, respectively. The labels xy, yz, and xz represent the tensors of each orthogonal plane. After irradiation, the tensor values ​​in the horizontal and vertical directions decreased, while those in the front-to-back direction increased. [Figure 24] These are images of the hippocampus of a Thy1-YFPH transgenic wild-type mouse (male, 17 weeks old) stained with YFP, Rhod4 pseudocolor fluorescence, HE staining, and NR2A AB staining. [Figure 25] This figure shows the changes in NMDA dependence of normalized F525 in the hippocampal CA1 and DG regions in the control, RT, RT+HBO, RT+memantine (Mema), and RT+HBO+memantine (Mema) treatment groups. Three square regions (10×10 μm) were monitored from the CA1 and DG regions (black, blue, and light blue traces). [Figure 26] This figure shows the mean maximum value (n=5) of Normalized F555 after NMDA application using different slice samples. Raw data are indicated by white dots above the bar graph. **p<0.01; *: p<0.05, two-tailed t-test. [Modes for carrying out the invention]

[0012] The inventors, in their search for a method to effectively promote the enhancement of endogenous neurogenesis as a way to treat various memory impairments caused by cancer treatment, 1) The cognitive decline in patients due to central nervous system toxicity caused by radiation is due to impaired hippocampal neurogenesis and damage to the cerebral white matter, which is the anatomical basis of its network function. 2) Identification of effective therapies for activating hippocampal neurogenesis, 3) Establish an effective method for repairing white matter myelin and axons, and demonstrate that the combination of the NMDA receptor antagonist memantine and hyperbaric oxygen therapy (HBO) is superior to monotherapy. This discovery provided a new insight, leading to the idea that memantine could be a preventative and therapeutic agent for cognitive impairment caused by cancer treatment and other factors, ultimately resulting in the completion of the present invention.

[0013] The following describes one embodiment of the composition, manufacturing method, and treatment method of the present invention.

[0014] The composition of the present invention is a composition for the prevention or treatment of cognitive impairment, used in conjunction with hyperbaric oxygen therapy, and contains memantine. The method for producing the composition of the present invention also includes a step of incorporating memantine.

[0015] The cognitive impairments targeted by the compositions of the present invention are not particularly limited, but examples include cognitive impairments resulting from a decrease in hippocampal neurogenesis. Specifically, examples include cognitive impairments caused by central nervous system toxicity from cancer treatments such as radiotherapy and chemotherapy.

[0016] Hyperbaric oxygen therapy (HBO) is a treatment that aims to improve a patient's condition by placing them in an environment with a pressure higher than atmospheric pressure and having them inhale high-concentration oxygen. The pressure inside the device and the treatment time can be set as appropriate according to the patient's condition.

[0017] The memantine (generic name: memantine hydrochloride) contained in the composition of the present invention is a therapeutic agent for Alzheimer's disease that acts by antagonizing N-methyl-D-aspartate (NMDA) receptors, which are one of the glutamate receptor subtypes. It is known to exert neuroprotective effects by suppressing the activation of NMDA receptors by excessive glutamate.

[0018] Furthermore, the conventionally known effects of memantine on Alzheimer's disease are thought to be that it suppresses the hyperactivity of extrasynaptic NMDA receptors caused by spill-over glutamate, while its inhibitory effect on intrasynaptic NMDA receptors involved in memory and learning is weak.

[0019] The composition of the present invention is based on the novel finding that, when used in combination with hyperbaric oxygen therapy (HBO), it can improve impaired hippocampal neurogenesis caused by central nervous system toxicity such as radiation, and thus differs in its use and function from the conventional use of memantine for the treatment of Alzheimer's disease as described above.

[0020] Furthermore, the composition of the present invention may consist of memantine alone, or it may contain one or more other compounds.

[0021] In the treatment method of the present invention, by administering hyperbaric oxygen therapy (HBO) to a patient and simultaneously administering the composition of the present invention (memantine), hippocampal neurogenesis can be activated, thereby preventing or treating (improving) cognitive impairment. This is an effect that could not be predicted from the conventionally known mechanism of action and effects of memantine on Alzheimer's disease, or from the effects of hyperbaric oxygen therapy (HBO).

[0022] Furthermore, in the treatment method of the present invention, the timing of hyperbaric oxygen therapy (HBO) and the administration of the composition (memantine) is not particularly limited.

[0023] Furthermore, the composition of the present invention can be used by incorporating it into pharmaceuticals or food and beverages.

[0024] In the case of pharmaceuticals or food products, excipients such as excipients, lubricants, binders, and disintegrants may be included, and the product may be manufactured as a pharmaceutical or food product having a dosage form or shape suitable for the intended route of administration or method of intake.

[0025] Examples of excipients include lactose, sucrose, D-mannitol, D-sorbitol, starch, pregelatinized starch, dextrin, glucose, corn starch, crystalline cellulose, low-substituted hydroxypropyl cellulose, sodium carboxymethylcellulose, gum arabic, pullulan, light anhydrous silicic acid, synthetic aluminum silicate, and magnesium aluminometasilicate.

[0026] Examples of lubricants include sugar esters such as sucrose fatty acid esters and glycerin fatty acid esters, hydrogenated oils such as calcium stearate, magnesium stearate, stearic acid, stearyl alcohol, and powdered vegetable oils, waxes such as bleached beeswax, talc, silicic acid, and silicon.

[0027] Examples of binders include pregelatinized starch, sucrose, gelatin, gum arabic, methylcellulose, carboxymethylcellulose, sodium carboxymethylcellulose, crystalline cellulose, sucrose, D-mannitol, trehalose, dextrin, pullulan, hydroxypropylcellulose, hydroxypropylmethylcellulose, and polyvinylpyrrolidone.

[0028] Examples of disintegrants include lactose, sucrose, starch, carboxymethylcellulose, carboxymethylcellulose calcium, croscarmellose sodium, carboxymethyl starch sodium, light anhydrous silicic acid, and low-substituted hydroxypropylcellulose.

[0029] Furthermore, examples of additives commonly used in the manufacture of pharmaceuticals and food and beverages include various oils and fats (e.g., vegetable oils such as soybean oil, corn oil, safflower oil, and olive oil; animal oils such as beef tallow and sardine oil), herbal medicines (e.g., royal jelly, ginseng, etc.), amino acids (e.g., glutamine, cysteine, leucine, arginine, etc.), polyhydric alcohols (e.g., ethylene glycol, polyethylene glycol, propylene glycol, glycerin, sugar alcohols, e.g., sorbitol, erythritol, xylitol, maltitol, mannitol, etc.), and natural polymers (e.g., gum arabic, agar, water-soluble corn). Examples of ingredients include fiber (e.g., gelatin, xanthan gum, casein, gluten or gluten hydrolysate, lecithin, starch, dextrin, etc.), vitamins (e.g., vitamin C, B vitamins, etc.), minerals (e.g., calcium, magnesium, zinc, iron, etc.), dietary fiber (e.g., mannan, pectin, hemicellulose, etc.), surfactants (e.g., glycerin fatty acid ester, sorbitan fatty acid ester, etc.), purified water, diluents, stabilizers, isotonic agents, pH adjusters, buffers, humectants, solubilizers, suspending agents, colorants, flavoring agents, odorants, fragrances, antioxidants, sweeteners, flavor components, acidulants, etc.

[0030] Furthermore, there are no particular restrictions on the dosage form or shape of pharmaceuticals or food and beverages.

[0031] Examples of pharmaceutical formulations include tablets, capsules, granules, powders, syrups, dry syrups, liquids, suspensions, inhalants, and suppositories, but oral formulations are preferred.

[0032] Examples of food and beverages include health foods and beverages in the form of candies, tablets, chewable tablets, powders, capsules, granules, and drinks (supplements, nutritional supplements, health supplements, and foods with health claims (foods for specified health uses, foods with nutritional function claims, foods with functional claims)).

[0033] The dosage or intake of pharmaceuticals and food products can be adjusted as appropriate, taking into consideration the age and weight of the subject, the route of administration, the frequency of administration or intake, etc. Furthermore, there are no limitations on the duration of administration or intake of pharmaceuticals and food products; for example, pharmaceuticals and food products can be administered or taken continuously for a period of one week or more, two weeks or more, one month or more, two months or more, six months or more, one year or more, or longer, according to the above-mentioned dosage and administration instructions.

[0034] The composition and method for producing the present invention are not limited to the embodiments described above. [Examples]

[0035] The present invention will be described below with reference to examples, but the present invention is not limited in any way to the following examples.

[0036] 1. Method (subject) For the acquisition of diffusion tensor images (DTIs), 20 brain tumor patients who had completed initial treatment, including surgery and radiotherapy, between June 2011 and November 2013 participated. Initial treatment was performed according to the neuropathological diagnosis and included temozolomide administration (TMZ 15 mg / kg daily for 42 days). Subsequently, chemoradiotherapy (30 fractions, totaling 60 Gy over 6 weeks) and HBO therapy were used in combination. Memantine therapy was administered to randomly selected participants. HBO therapy was administered at an absolute pressure of 2.8 atmospheres for 40 minutes immediately before each radiotherapy session.

[0037] Memantine (5 mg daily) was administered to selected participants during radiation therapy. The 20 participants were divided into three groups: a control group (no HBO or memantine treatment) (n=3), an HBO group (n=9), and an HBO + memantine group (n=8).

[0038] The fMRI analysis included 10 brain tumor patients before cranial irradiation (cIR) (mean age 28.7 ± 11.3 years, range 13–47 years, 3 males, 7 females), 13 brain tumor patients after cIR (mean age 29.0 ± 10.0 years, range 13–47 years, 4 males, 9 females), 31 benign brain tumor patients (mean age 26.2 ± 6.6 years, range 15–38 years, 13 males, 18 females), and 36 healthy volunteers (mean age 25.2 ± 2.9 years, range 22–35 years, 23 males, 13 females). The healthy volunteers who participated in the fMRI analysis were treated as a control group, distinct from those included in the neuropsychological analysis. None of these patients had any signs or history of neurological or psychological disorders.

[0039] (Drug administration to mice) Memantine hydrochloride (Sigma-Aldrich, MO, USA) was dissolved in sterile saline solution. For chronic treatment, ALZET pumps (Model 2004, ALZET, CA, USA) were filled with either saline solution or memantine solution and subcutaneously implanted in the scapula of mice for 4 weeks. The pumps were filled with a concentration that allowed 5 mg / kg / day of memantine to diffuse.

[0040] (immunohistochemistry) Mice were deeply anesthetized with isoflurane and transcardially perfused with 4% paraformaldehyde in phosphate-buffered saline (PBS). The brains were stored at 4°C for 3 or 5 days in 4% paraformaldehyde in PBS before being transferred to 70% ethanol. The coronary artery segmented brains were embedded in paraffin and sectioned with a microtome. After deparaffinization and blocking, they were incubated overnight with primary antibody at 4°C.

[0041] (Golgi apparatus staining) Golgi apparatus staining was performed using FD Rapid Golgistain Kits (FD Neuro Technologies, USA) according to the manufacturer's protocol. Briefly, mouse brains were collected after perfusion with 4% PFA, transferred to Golgi staining solution, and 100 μm thick sections were obtained using a cryostat.

[0042] (Hyperbaric oxygen treatment using mice) Two to three 8-week-old mice were placed in cages (width x height x depth; 19 x 10.5 x 12 cm) and housed in a hyperbaric chamber (Valortec Hanuda Co., Ltd., Tokyo, Japan) (Figure 1). Hyperbaric oxygen therapy (HBO) was performed at 2.5 atmospheres for 40 minutes. Oxygen was supplied to the chamber at 5 L / min, then discharged at 2 L / min for 10 minutes, and then pressurized to 2.5 atm. After the pressure reached 2.5 atm, the supply and discharge rates were balanced at 2 L / min and maintained for 40 minutes, after which the pressure was reduced for 10 minutes. Radiation therapy was performed within 30 minutes after the end of HBO.

[0043] (Radiation on mice) Each mouse was anesthetized with isoflurane, and its limbs were secured to a lead sheet with adhesive tape. The entire body of the mouse, excluding the head, was covered with a lead sheet. The mice were placed in an X-ray irradiation device (Figure 1) (Faxitron RX-650, FAXITRON BIOPTICS, Arizona, USA), and their whole brains were irradiated with 2 Gy or 5 Gy (5 Gy for immunohistochemistry, 2 Gy for behavioral experiments). The non-irradiated group of mice were also secured to a lead sheet and kept in the radiation box for the same amount of time as the irradiated mice, but were not irradiated. For behavioral experiments, they were irradiated with 2 Gy for 5 days, for a total of 10 Gy.

[0044] (Elevated cross maze test) The system was set up 50cm above the floor and employed an elevated cross-shaped maze consisting of two open arms, two closed arms (30 x 6cm each), and a neutral zone. The mouse was placed in the center of the neutral zone, facing the closed arms, and allowed to move freely for 3 minutes. The time spent on the open and closed arms, and the number of times the mouse moved to a different arm were recorded and scored.

[0045] (Fear conditioning test) The fear conditioning test procedure was as follows: On day 1, mice were placed in a chamber (175 × 165 × 300 mm) for 150 seconds and given a foot shock (0.5 mA × 2 seconds). After 28 seconds, they were returned to their home cage. The test was performed on day 2. Mice were re-exposed to a chamber without foot shock for 3 minutes, and the animals' freezing response was monitored for 3 minutes. Each chamber was washed with distilled water, and then with 70% ethanol.

[0046] (Delayed Non-Matching to Place (DNMP) task) The DNMP task was used to test pattern separation ability. An eight-arm radial maze consisting of eight arms (centers of octagons measuring 7 cm wide, 35 cm long, and 20 cm wide) was employed. The experiment was conducted after irradiation treatment, and all mice underwent food restriction for one week prior to the experiment, maintaining their body weight at 85%–90% during the experiment. Water was supplied with Adlibitat.

[0047] To acclimate the mice to the 8-arm radial maze, on the first day, mice in the same cage explored the maze together for one hour. On the second day, they explored the maze individually for 5 minutes. Subsequently, they performed the DNMP task three times a day for 10 days. This task evaluated their ability to retrieve samples from the correct arms within the maze.

[0048] In the sampling phase, the doors to the starting arm and the sample arm were opened. Mice were kept in the starting position for 5 seconds, after which they were allowed to move to the sample arm and collect a food pellet. In the selection phase, the starting arm, the sample arm (non-rewarded), and another correct arm (rewarded) were opened. In trials, the correct arm and the sample arm were separated by 45, 90, and 135 degrees. The starting arm position was changed pseudo-randomly in each trial. Mice that selected the correct arm were considered to have made the correct choice, and mice that selected the sample arm were considered to have made the wrong choice and were allowed to enter the correct arm and collect a food pellet before being removed from the maze.

[0049] (New object search test) After irradiation, the mice were handled for 5 minutes each day for 5 days prior to the start of the experiment. On day 1, they were acclimatized for 10 minutes in an open-field box (25 x 25 cm), and on day 2, they were offered two identical objects for 10 minutes. On day 3, one of the objects was replaced with a new one. The mice were observed with a video camera for 5 minutes. Object recognition was measured by the time the mouse's nose was pointed towards the object and both of its forelegs were simultaneously touching the object.

[0050] (Event-related fMRI) (Experimental paradigm) To evaluate the pattern separation function of the hippocampus, an fMRI behavioral task was used. The experimental paradigm followed the methods described in Non-Patent Document 1 and Patent Document 3. The behavioral condition was a three-choice forced selection task consisting of novel, same, and similar stimuli made up of color photographs of common objects. Fully randomized functional performance consisted of a total of 108 trials: 16 Lure sets, 16 Same (repeated) sets, and 44 unrelated novel items (foils).

[0051] A new task was presented consisting of 44 foils trials, 16 Same (repeated) sets (with the first trial presented), and 16 Lure sets (with the first trial presented). The number of trials separating similar and identical pairs was randomly varied from 10 to 40. The Same task and Lure task were presented again, using the Same (repeated) set and Lure set, respectively. In each trial, images were presented on a goggle display (Resonance Technologies, Inc., Salem, Massachusetts) at 2,500 ms with an inter-stimulus interval of 0–1,000 ms.

[0052] Participants were instructed to identify whether a picture was new, the same, or similar, and to press a button indicating their response. Responses were recorded using a button box (Current Designs, Inc., Philadelphia, Pennsylvania). Presentation(R) software (Neurobehavioral Systems, Inc., Austin, Texas) was used for presenting visual stimuli and collecting behavioral data.

[0053] (Acquisition of high-resolution MRI data) Functional and structural data were collected using imaging techniques. Participants were scanned using a 3-T MRI scanner (Discovery MR750; GE Medical Systems, Waukesha, Wisconsin, USA) with a 32-channel head coil and high-order manual shimming of the whole brain. Array spatial sensitivity encoding (parallel imaging) was employed to reduce geometric distortion in echo-planar imaging (EPI) and acquire imaging data.

[0054] High-resolution EPI sequences were collected to measure bold contrast. Twenty-three oblique coronal slices were fitted to the main longitudinal axis of the hippocampus, covering the entire bilateral medial temporal lobe. A total of 303 volumes were collected sequentially in ascending order in each session during the experimental period.

[0055] Whole-brain spoiled gradient entrench echo (SPGR) structural sequences were obtained with isotropic resolution of 1 mm x 1 mm x 1 mm.

[0056] High-resolution T2 fast spin echo (T2 FSE) scans were obtained for the coordinated registration of SPGR and EPI data. Functional and T2 FSE structural images were acquired in the same slice oriented along the long axis of the hippocampus. The entire hippocampus (head, torso, and tail) was imaged in 23 slices.

[0057] (fMRI data analysis) Data analysis was performed using SPM8 software (Wellcome Trust Centre for Neuroimaging, University College London, London, UK). The first five volumes of each EPI dataset were removed to confirm that the signal had reached a steady state. Next, EPI functional images were co-registered and motion artifacts were corrected by a readjustment process. Slice timing correction processing was performed to correct the EPI data to account for differences in slice acquisition time.

[0058] First, EPI data were co-registered with T2 FSE structural images. Next, they were co-referenced with T2 FSE structural images. Spatial normalization parameters were generated from structural SPGR images using a segmentation routine. Individual spatial normalization parameters were adopted, and structural SPGR and EPI functional images were spatially normalized using the Montreal Neurological Institute space (1×1×1 mm). Finally, EPI data were spatially smoothed using a Gaussian kernel with a full width at half maximum of 3 mm. To detect brain activity related to specific tasks while simultaneously reducing noise, the smoothing kernel size was set to a recommended size (2-3 times the voxel size). A high-pass filter (200 seconds) was used to remove slow signal drift within the session. The effect of rigid body motion was examined to identify correct activation spots in the brain. For this purpose, head motion parameters (3 translations and 3 rotations) were used. For each subject, novel, same, and lure contrasts were estimated using a generalized linear model calculated by combining time and variance derivatives with the canonical hemodynamic response function.

[0059] (Group-level fMRI analysis) Second-level contrast was calculated for each group (healthy volunteers, benign brain tumor patients, brain tumor patients before cIR, and brain tumor patients after cIR) using a one-sample t-test for New, Lure, and Same tasks. Differences in activation intensity between task conditions were confirmed at the voxel level threshold p<0.001 (uncorrected) and at the cluster level threshold FWE (Family Wise Error) corrected p<0.05. Activation and inactivation spots were identified on an anatomically defined hippocampal atlas.

[0060] (Analysis of blood oxygen level-dependent (BOLD) response) BOLD responses were extracted from DG data using Lure task data. The latency (seconds) and amplitude (rate of change of BOLD signal) of the BOLD curve were measured at the peak of the BOLD response (N1), the second negative peak (N2), the initial positive peak (P1), and the second positive peak (P2), respectively. The mean and standard deviation of the BOLD curve were calculated for each group (normal healthy individuals, patients with benign brain tumors, patients with brain tumors before cIR, and patients with brain tumors after cIR).

[0061] (Functional analysis of the hippocampus) Reaction times and accuracy rates for the New, Lure, and Same tasks were evaluated for statistical significance between groups using the Mann-Whitney U test (p < 0.05). Comparisons were made between healthy controls, patients with benign brain tumors, patients with brain tumors before cIR, and patients with brain tumors after cIR. The correlation coefficient between the proportion of bold signals and accuracy rates in the Lure task was calculated using Spearman's rank correlation.

[0062] (Analysis of neurocognitive function) Neurocognitive function assessments were performed based on the following eight tests: (I) Mini-Mental State Examination (MMSE) and Modified MMSE (3MS) for global cognitive screening; (II) Trailmaking Test (TMT) and Stroop Test (ST); (III) Wechsler Adult Intelligence Scale-Revised Working Memory (WAIS-R) Digit Span Subtest (DS); (IV) WAIS-R Digit Symbol Test (DST) for psychomotor speed; and (V) WAIS-R Block Design Subtests (items 5 and 9) and Cube Copy Test for visuospatial ability.

[0063] These tests were performed on 119 preoperative patients (27 with malignant brain tumors (20 with gliomas, 2 with metastatic tumors, and other malignant tumors), 66 with benign brain tumors (17 with schwannomas, 30 with meningiomas, 9 with pituitary adenomas, 3 with epidermal tumors, 3 with craniopharyngiomas, 2 with choroid plexus papillomas, 1 with hemangioemyocytoma, 1 with neurocytoma, and 1 with lipoma), 12 with cerebrovascular disorders (4 with cerebral infarction, 3 with arteriovenous malformations, 1 with chronic encapsulated hematoma, 1 with cavernous hemangioma, and other diseases), and 14 with other conditions (3 with normal pressure hydrocephalus, and other diseases)). Scores were analyzed using the Wilcoxon signed-rank test (* p<0.05).

[0064] (Immunohistochemical analysis) Mouse brains were perfused and fixed with 4% paraformaldehyde, embedded in paraffin, and then sectioned into 4 μm thick sections, which were stored until immunohistochemical analysis. For BrdU staining, 50 mg / kg of BrdU (Merck, Germany) was intraperitoneally injected into each mouse 5 hours before perfusion and fixation. Pre-treatment by autoclaving (10 mM citrate buffer (pH=6), 121°C, 10 minutes) was performed to denature the DNA; this was necessary to promote the antibody-BrdU reaction. Anti-doublecortin antibody (E-6; monoclonal, 1:50, Santa Cruz Biotechnology, Inc., CA, USA), anti-neurofilament H 200kD antibody (AHP2259GA; polyclonal, 1:1000, Bio-Rad, CA, USA), anti-myelin base protein (MBP) antibody (A0623; polyclonal, 1:500, DakoCytomation, Glostrup, Denmark), and anti-BrdU antibody (BU-1: monoclonal, 1:1000, ThemoFisher Science, Waltham, USA) were used. Positively stained cells were counted using Zeiss Axio Observer 7 (Zeiss, Germany).

[0065] (DTI(diffusion tensor images)) To quantitatively evaluate the state of white matter bundles, diffusion tensor imaging (DTI) was acquired preoperatively and after the initial treatment. Images were acquired using a Discovery MR750 (3.0 Tesla model, General Electric (GE) Healthcare). The imaging conditions were: Slice 2mm, Dir 29, Echo time (TE) 81.8 msec, Repetition time (TR) 9,500 msec, b-value 1,000 s / mm2. Neuropsychological tests were also performed simultaneously to evaluate the impact on neurocognitive function.

[0066] (Quantitative analysis of DTI measurements (FA, MD, AD, RD) in white matter bundles) To minimize the effects of surgery and brain swelling, white matter bundles on the contralateral side of the tumor were measured. The white matter fiber bundles measured were defined as projection fibers and association fibers. Commissural fibers connecting both cerebral hemispheres were excluded. The seven fiber bundles analyzed were the anterior thalamic radiation (ATR), corticospinal tract (CST), cingulate gyrus (CNG), inferior fronto-occipital fasciculus (IFOF), inferior longitudinal fasciculus (ILF), superior longitudinal fasciculus (SLF), and uncinate fasciculus (UNC).

[0067] All analyses were performed using a Mac Pro (Late 2013) system (Processor: 3.5 GHz 6-Core Intel Xeon E5; memory: 32 GB; 1866 MHz DDR3 ECC; graphics card: AMD FirePro D500 3072 MB). After whole-brain analysis, additional quantitative analysis using DTI was performed on each white matter bundle. TBSS was used for whole-brain analysis. For the additional quantitative analysis, regions of interest (ROIs) were defined using the atlas included with the FMRI software library (FSL).

[0068] For this analysis, we selected the white matter tractography atlas from Johns Hopkins University (JHU). First, we launched the FSL viewer to determine the white matter fiber bundles to be analyzed, and then used fffmaths to search for the target structures in the atlas. Next, we used the fslsplit command to separate the skeletal images of each case. Finally, we applied the extracted structural images as masks and executed the fslstat command. For each case, we extracted the mean values ​​of DTI-derived indices such as FA. For each group, we calculated the mean values ​​of DTI-derived indices for each white matter bundle, and analyzed the statistical significance of pre-operative and post-initial treatment differences using the Wilcoxon t-test (*: p<0.05).

[0069] (Brain slice calcium imaging) Hippocampal brain sections (thickness 300-400 μm, obtained using a Linear Slicer; PRO 7N, Dosaka EM, Japan) were prepared from 17-18 week old male Thy1-YFPH transgenic wild-type mice (The Jackson Laboratory, stock number: 003709, strain name: B6.Cg-Tg.(Thy1-YFP)16Jrs / J) (Feng et al. 2000).

[0070] The slices were incubated in a standard Krebslinger solution (125 mM NaCl, 2.5 mM KCl, 10 mM D-glucose, 1.25 mM NaH2PO4, 26 mM NaHCO3, 2 mM CaCl2, 1 mM MgCl2, continuous bubbling of a mixed gas [95% O2; 5% CO2]) at room temperature for 60 minutes to allow them to recover from damage during the cutting process. The Ca indicator Rhod-4 AM (excitation wavelength: 530 nm, emission wavelength: 555 nm, Kd, ​​0.525 μmol / L) (5 μM) (Dojindo, Kumamoto, Japan) was immobilized on the brain slices for 90 minutes. YFP fluorescence was monitored to confirm the morphology of the hippocampus. To monitor intracellular Ca2+(Ca2+)i, Rhod-4 fluorescence intensity (F555) was monitored using a photomultiplier tube-equipped confocal microscope (excitation wavelength 543 nm; LSM5 PASCAL, Carl Zeiss, Germany). In this experiment, Ca imaging of the entire hippocampal slice was performed using a low-magnification (2.5x) objective lens (FLUAR 2.5×, NA=0.12, Carl Zeiss, Germany). Rhod-4 fluorescence images (512×512 pixels) were digitized and stored on a computer every 10 seconds for 50 minutes.

[0071] The F555 values ​​for the CA1 and DG regions were normalized to the mean F555 value (from 0 to 5 minutes before NMDA application) in offline analysis (Microsoft Excel, Microsoft Corporation, WA) (normalized F555). During Ca imaging, brain sections were placed in a perfusion chamber (35 mm μ-dish, Ibidi GMBH, Grafelfing, Germany) and continuously perfused with extracellular fluid [125 mM NaCl, 2.5 mL / min] (2 mL). Perfusion was performed along with continuous bubbling of a mixed gas (95% O2; 5% CO2) of 5 mM KCl, 10 mM D-glucose, 1.25 mM NaH2PO4, 26 mM NaHCO3, 2 mM CaCl2, 1 mM MgCl2, and 1 μM tetrodotoxin. For NMDA, the MgCl2 concentration was set to 0 mM to prevent Mg-dependent inhibition of NMDARs. The normalized F555 value was estimated as the maximum value within 5 minutes after application of NMDA (Tocris, Bristol, UK) (50 μM). Glycine (Tocris, Bristol, UK) (10 μM) was used to activate the NMDAR.

[0072] 2.Results (A model of radioactive toxicity in the brain of rodents) In primary brain tumors, localized wide-area irradiation with intensity-modulated radiotherapy

[25] is common. In this experiment, a whole-brain irradiation field was used instead of the hippocampal target field commonly used in animal models.

[0073] To suppress neurogenesis in the hippocampus, we developed a rodent model using fractionated whole-head irradiation that reflects the toxicity of human radiation therapy, rather than local irradiation of the hippocampus.

[0074] HBO is an established treatment for decompression sickness, air embolism, carbon monoxide poisoning, and radiation damage in various parts of the body. HBO has a strong radiosensitizing effect and is applied to radiation-resistant gliomas to enhance the effects of ionizing radiation. However, its neuroprotective effects against radiation-induced intracranial toxicity remain unclear.

[0075] The inventors developed a novel neuroprotective method that considers hippocampal neurogenesis and white matter preservation, and analyzed the effectiveness of HBO administration before cranial radiation therapy and neuromodulation with memantine, an NMDAR antagonist that promotes neurogenesis in the hippocampal dentate gyrus (DG).

[0076] (Effects of memantine and HBO) To detect the initial phenomena of hippocampal neurogenesis after exposure to ionizing radiation, histological examination was first performed on bromodeoxyuridine (BrdU)-labeled specimens. S-phase DNA was analyzed in animals that underwent fractionated whole-brain irradiation (2 Gy / day / fraction, 5 consecutive days, total dose). S-phase DNA was analyzed in animals that received 10 Gy / 5 fractions (n=12, 4 independent experiments), or HBO preconditioning before irradiation (n=12), and / or memantine (5 mg / kg, 5 consecutive days), and compared with data from unirradiated mice (n=12). Specimens were prepared 2 days after the final treatment. The irradiation group had significantly fewer BrdU-positive cells compared to the control group (p<0.001). DNA synthesis increased in the irradiation + memantine group, reaching a level similar to the control group (reference value) (n=12).

[0077] Memantine also induced migratory or mitotic morphologies in doublecortin-positive neural progenitor cells (Figure 2). These cells migrated horizontally in the subgranular layer and vertically throughout the granular layer. Oxygen affects the differentiation and proliferation of neural stem cells (NSCs). In vitro, hypoxia (less than 5%) induces self-renewal of type B NSCs, while hyperxia (greater than 20%) induces neuronal maturation. As evident from BrdU labeling experiments, HBO alone did not stimulate DNA synthesis compared to controls (n=12) (Figure 3). Doublecortin (DCX)-positive cells were located in the subgranular layer (Figures 2 and 3). Furthermore, in the HBO + memantine group, bipolar migratory cells frequently induced by memantine monotherapy disappeared, as did migratory and mitotic cells observed in memantine-treated irradiated mice (Figure 2). The BrdU labeling index in the radiotherapy (RT) + HBO + memantine group was significantly increased compared to the RT alone group (p<0.001), but there was no change compared to the RT + memantine group.

[0078] Memantine administration suppressed the reduction of dendritic spines in hippocampal neurons of irradiated mice. While RT reduced the number of Golgi-stained cells in the subgranular layer, memantine administration restored this number to non-irradiated control levels (n=4). Furthermore, spine retraction and contraction induced by RT were reversed by memantine treatment (Figure 4).

[0079] Next, we investigated whether the administration of HBO or memantine improved radiation-induced white matter damage.

[0080] Since excessive glutamate signaling via NMDARs in oligodendrocytes mediates signal transduction from axons to myelin, it was hypothesized that NMDAR activation caused by excessive glutamate induced by brain irradiation would induce acute excitotoxicity in myelin-forming oligodendrocytes. Indeed, intracranial radiation exerted neurotoxicity on the white matter tract (Figure 5A). HBO remodeled myelin more effectively than memantine, as inferred from the expression of myelin basic proteins (myelin sheath markers in oligodendrocytes) (Figure 5B).

[0081] Furthermore, immune responses to high molecular weight neurofilament proteins (NFP-200kDa) revealed that radiation causes severe morphological changes in pyramidal cells. Axonal degeneration and deformation were also observed, but memantine and HBO improved axonal degeneration (HBO was more effective than memantine) (Figure 5C, Figure 6). Therefore, it was confirmed that HBO repairs white matter damage and restores brain nerve cells, and that memantine effectively addresses the radiotoxicity to hippocampal neurogenesis.

[0082] The morphology of dendrites was examined by Golgi staining. In the irradiated group, pyramidal cells were completely absent from the subgranular layer of the hippocampus, and a small number of Golgi-stained cells were observed, along with a decrease in branching, dendritic length, and spine density (n=4) (Figure 7). In memantine-administered mice (n=4), the morphology of degenerated and receding spines and the sparse spine density of dendrites were found to be significantly restored to the same extent as in non-irradiated mice (n=4). This indicates that memantine maintains the morphology of nerve cells and the density of dendritic spines, promotes neurogenesis, and exhibits neurotrophic effects (Figure 4).

[0083] (Behavioral responses in radiotoxic mice due to increased neurogenesis) An elevated cross maze experiment was used to evaluate whether ionizing radiation affects anxiety-like behavior (Figure 8). Irradiated mice spent more time and more frequently going to the open arm, and also spent more time in the open arm.

[0084] Next, we investigated whether the regulation of hippocampal neurogenesis by memantine, HBO therapy, or a combination thereof could improve behavioral responses in radiotoxic mice. For this purpose, pattern separation ability was evaluated using a delayed non-matching to place protocol while varying the angle between the sample arm and the reward arm in a radial arm maze (Figure 10). In separations 1-3, the sample arm and reward arm were separated at angles of 45°, 90°, and 135°, respectively (Figure 10). As a result, we found that memantine administration increased correct selection only in separation 3, but the difference was not statistically significant (n=12) (Figure 11).

[0085] To assess cognitive memory, a novel object search test was used, consisting of a 10-minute exposure period and a 5-minute test period.

[0086] Mice that received brain irradiation showed a decrease in exploratory behavior (n=6). The control group and the RT+memantine group showed a significant increase in the time spent exploring novel objects compared to familiar objects (Figure 12). Memantine has been reported to induce neurogenesis in the hippocampal DG region. These results suggest that neurogenesis in the hippocampal DG region plays an important role in protecting against RT-induced NMDAR inactivation.

[0087] (Non-invasive neuroimaging of neurogenesis in the human hippocampus) The inventors used previously reported fMRI behavioral paradigms (Non-Patent Document 1, Patent Document 3). This task involves an event-related fMRI design based on an explicit three-choice forced-choice task, including New, Same, and Lure tasks consisting of color photographs of common objects.

[0088] The average radiation dose to the hippocampus was 22.7 ± 18.3 Gy (0.5-60 Gy), and 13 patients underwent cranial irradiation (cIR), resulting in a significant decrease in accuracy. The accuracy rate on the Luer task was 38.0 ± 19.0% (n=13, age: 28.7 ± 11.3 years; 13-41 years) before CIR, but decreased to 25.0 ± 27% (n=13, age: 29.0 ± 10.5 years; 13-41 years) after CIR. On the other hand, the correct response rate for the lure task among healthy individuals was 47.0±19.0% (n=36, 25.2±2.9 years, 23-35 years), and among the non-irradiated group it was 43.0±21.0% (n=31, 24.6±8.1 years, 7-38 years). In healthy individuals, the correct response rate for the lure task was 0.9% in the non-irradiated group and 0.5% in the irradiated group.

[0089] Importantly, analysis of the accuracy rates for the New and Same tasks (excluding the Lure task) showed no significant difference between the groups (Figure 13). Figure 14 shows the relationship between accuracy rates and hippocampal dose in individual patients.

[0090] In the Lure task, the rate of incorrect answers decreased when the average dose to the right hippocampus was 8.5±5.0 Gy (0.5-6.2 Gy; n=10) and the average dose to the left hippocampus was 7.9±5.4 Gy (0.5-16.0 Gy; n=10) (Tables S1, S6). Four out of ten patients who received radiation to the left hippocampus and four out of ten patients who received radiation to the right hippocampus had a 0% correct answer rate on the Lure task, indicating complete inhibition of neurogenesis (Figure 14). Additionally, five patients received ionizing radiation to both the right and left hippocampuses and were similarly included in the right and left hippocampal exposure group. In this overlapping group, the correct answer rate increased by 56-81% in one patient and a slight increase of 6-13% in another patient. Two other patients in the group exposed to radiation in both the right and left hippocampi showed reduced scores on the Lure task with radiation doses of 2 Gy or less, demonstrating high sensitivity to radiotoxicity (Figure 14). While hippocampal sensitivity to toxicity is not uniform among individuals, there was no difference in laterality or dominance regarding the site of radiation exposure between the right and left hippocampi. Furthermore, radiation exposure caused a decrease in pattern separation ability, but not a decrease in correct responses to New or Same tasks (Figure 13). No significant differences were observed between groups in the analysis of reaction times for New, Lure, and Same tasks.

[0091] Next, we analyzed the BOLD response patterns in the left and right DG during the Lure task. BOLD is an established brain function imaging signal that reflects neural responses that lead to changes in the oxyhemoglobin to deoxyhemoglobin ratio to support the energy demands associated with functional responses.

[0092] Neurophysiologically, this hemodynamic response correlates more with local potentials, incident input, and local processing of a given region than with action potentials (MUV; multi-unit activity). In healthy subjects (n=36, 25±15.0 years), the first dip in the BOLD response peak (first negative-peak; N1) occurred at 1.7±1.5 seconds, followed by a fractional increase in the first positive peak (P1) at 4.1±1.1 seconds. Subsequently, the signal decreased 8.4±1.6 seconds later (second negative peak; N2), with a %BOLD signal change from resting signal of -0.19±0.27, followed by a slope that plateaued or peaked with a long pulse (>20s) (Figure 15, first row). On the other hand, in patients with non-irradiated brain tumors (n=31, 25±15.0 years old), a characteristic feature was a slope reaching a low-amplitude plateau, followed by a low-amplitude N1, P2 (second positive peak), and then a low-amplitude plateau (Figure 15, second row). Furthermore, in patients with cIR (n=13, 25±15.0 years old, Figure 14, fourth row), a significant decrease in %BOLD signal changes was observed. In cIR patients, a delay in N1 (4.7±0.59 s) and a small amplitude of %BOLD signal change (-0.08±0.24) were observed, followed by a delay in N2.

[0093] In the group before RT (n=10, 25±15.0 years old), a small P1 (0.02±0.13s) without an initial dip was followed by N1 and P2 with large SD values ​​(amplitudes -0.11±0.29 and 0.08±0.31, respectively), indicating that the signal in this group was fluctuating (Figure 15, third panel). In the non-IR group and healthy individuals, a positive correlation was observed between the bold response and the accuracy rate on the Luer task (Figure 15, first and second panels). However, no such correlation was observed in the cIR group (Figure 15, third and fourth panels).

[0094] Similar to the mouse hippocampal X-ray ablation model, analysis of human subjects using the Lure task-based fMRI paradigm revealed that brain irradiation affects hippocampal function and partially induces DG dysfunction. Furthermore, in the metabolic map (Figure 15, fourth row), hippocampal DG was inactivated, and a high error rate was observed in the Lure task, but not in the New task (Figure 14). These findings represent the first successful imaging of cranial radiation-induced toxicity in relation to human adult neurogenesis.

[0095] Next, we analyzed the hippocampal pathways using continuous metabolic maps obtained during the fMRI task. In healthy subjects, both hippocampi showed similar activation in the DG, CA3, CA1, and hypocerebellar pathways (n=36; Figure 16A). In the cIR- group, the signal was limited to the right hippocampus, the activation spots were limited to the right spinal cortex, and the inactivation spots were observed in the DG, CA3, CA1, and subiculum (n=13) (Figure 16B). In the New task, healthy subjects showed activation in both the bilateral perispinal cortex, DG, CA3, CA1, and trabeculae, while the cIR- group showed activation in both bilateral perispinal cortex and trabeculae, and inactivation in both CA1 pathways (Figures 16C, 16D). Finally, in healthy subjects, similar to the Lure task, activation signals were detected bilaterally, and in addition to direct activation of CA1 in both hippocampi, activation of the DG, CA3, CA1, and endoplasmic reticulum pathways was also observed (Figure 16E). In the irradiated patient group, hot spots were observed in the right EC, bilateral CA1, and subpyramidal regions, but not in the DG or CA3 (Figure 16F).

[0096] (Mechanisms of neural regulation in the human brain by memantine and HBO) Neurogenesis in adult humans plays a crucial role in learning and memory; therefore, restoring neurogenesis is important for improving patients' cognitive function. Based on previous clinical trials and our own results from rodent models, the inventors selected memantine, a non-competitive NMDAR inhibitor, and investigated its improvement in neurogenesis when used in combination with HBO. In an initial exploratory randomized phase II clinical trial, we inferred whether the correct response rate to established tasks reflected hippocampal neurotoxicity (i.e., whether it suggested neuromodulatory neurogenesis). To this end, we analyzed hippocampal function using task-based fMRI (behavioral paradigm) in patients undergoing radiotherapy.

[0097] Twenty-two patients received 5 mg of memantine 9.0 ± 10.0 days after the start of radiotherapy (1.0 ± 5.0 Gy). The correct answer rate for the Lure task before radiotherapy was 17.9 ± 16.9% (n=14). Memantine was administered during radiotherapy (45.0 ± 15.0 Gy). After the completion of radiotherapy, the correct answer rate for the Lure task increased to 29.5 ± 14.1% (n=14). In the non-memantine group, a significant deterioration was observed, from 19.6 ± 20.3% before radiotherapy to 11.5 ± 12.7% after radiotherapy (n=8). The Lure task score reflects the destruction and regeneration of hippocampal neurogenesis, suggesting that neuromodulation with memantine in combination with HBO is effective in this respect.

[0098] (Improvement of higher-order cognitive brain functions through promotion of neurogenesis) In mice, increased neurogenesis in the adult hippocampus induced enhancements to normal object recognition, spatial learning, contextual fear conditioning, and extinction learning. Increased efficiency in distinguishing overlapping contextual representations was also observed, indicating enhanced pattern separation. Furthermore, stimulating adult hippocampal neurogenesis, combined with interventions such as spontaneous movement, significantly increased exploratory behavior and improved cognition and mood. To determine whether increased neurogenesis improves higher-order cognitive brain functions, neuropsychological assessments, analysis of pattern separation ability, and changes in white matter integration were examined. Preoperative analysis revealed that higher-order cognitive functions are influenced at least by hippocampal neurogenesis (Figure 13).

[0099] The fMRI pattern separation ability test was administered to the same subjects as the neuropsychological tests. The group with a higher accuracy rate on the Luer task than the median (25%) (high score group: 47.0±16.8%, range: 25-88%, low score group: 6.5±6.6%, range: 0-23.0%) was included. Furthermore, scores on 3MS, HDS-R, DS, DST, ST-naming, and ST-interference were also high. This indicates that hippocampal memory function is important for general cognitive function (3MS & HDS-R), psychomotor speed (DST), and executive function (TMT & ST) (Figure 17).

[0100] As shown in Figure 13, neurogenesis was significantly reduced in patients who received intracranial radiation. Furthermore, impaired hippocampal neurogenesis worsened neurocognitive dysfunction in patients who received cranial radiation (Figure 18). Interesting representative cases are shown in Figures 19 and 20. The relationship between hippocampal pattern separation ability and neurocognitive domains is shown, indicating that hippocampal DG decreased in the early stages of cranial radiation, and hippocampal function and high cognitive function recovered by the end of chemoradiotherapy. Of the four areas of neuropsychological testing (MMSE, TMT, DS, DST), patients who received memantine + HBO showed a significant increase in DST scores, but no significant results were observed in the other areas (Figure 17). Since DST can estimate psychomotor velocity, it is presumed that white matter health is an important factor in signal transmission in neural networks.

[0101] Next, we estimated the contralateral hemisphere's cranial radiation-induced toxicity by applying diffusion tensor imaging data to estimate the integrity of the white matter's ultrastructure. Fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) were examined. Whole-brain analysis was performed using TBSS, which is highly reproducible, objective, and observer-independent. In rodent brain toxicity models, a broad irradiation field affected the hippocampus and caused white matter damage. In humans, when radiation therapy was performed on an area 2 cm larger than the area of ​​T2-weighted images (MR images) representing infiltration and tumor size, not only direct tumor infiltration but also significant edema often occurred simultaneously, causing deformation of the local hemisphere.

[0102] Therefore, we examined seven association fibers and projection fibers from the non-affected hemisphere: the anterior thalamic radiation (ATR), corticospinal tract (CST), cingulate gyrus (CNG), inferior fronto-occipital fasciculus (IFOF), inferior longitudinal fasciculus (ILF), superior longitudinal fasciculus (SLF), and uncinate fasciculus (UNC).

[0103] In cases where radiotherapy and hyperbaric oxygen therapy were used in combination, seven white matter functional fibers—ATR, CST, CNG, IFOF, ILF, SLF, and UNC—were measured using FA, MD, AD, and RD values, and comparisons were made before and after the combination therapy (Figure 21, n=9).

[0104] Following treatment, FA values ​​significantly increased in CST, IFOF, and UNC, suggesting improved white matter integrity. MD values ​​significantly decreased in IFOF, ILF, and UNC, and these fibers simultaneously showed a significant decrease in RD values, suggesting enhanced myelination. AD values ​​significantly decreased in SLF after radiotherapy, and RD values ​​decreased in CNG. These results suggest that hyperbaric oxygen therapy has multifaceted effects on neural networks.

[0105] In the HBO group, complete recovery of the microstructure of five fibers—CST, IFOF, ILF, SLF, and UNC—was observed. In representative cases of the HBO + memantine combination group, complete recovery of the microstructure of two fibers—ATR and ILF—was observed (Figures 22A-E, 23).

[0106] (Mossy fiber (MF) axonal rearrangement, negative bold signal, and pattern separation) Recent imaging studies in humans suggest that pattern separation is mediated by a circuit consisting of EC, DG, and CA3. Rearrangement of MF axons is necessary for pattern separation. Morphological maturation of hippocampal MF synapses affects cognitive abilities, and its absence leads to intellectual disability. Furthermore, changes in dendritic spines are associated with long-term potentiation.

[0107] The inventors applied immunohistochemical analysis of DCX to detect neural progenitor cells and found that a significant number of DCX+ cells were observed in the HBO+ memantine group.

[0108] Patients who received cranial radiation showed a significant decrease in BOLD signal changes, suggesting failure of neurogenesis in the DG. Analysis of two representative cases of brain tumors (cerebral glioma and atypical meningioma) revealed a negative BOLD response associated with treatment. This negative BOLD phenomenon indicates suppression of local synaptic function. Furthermore, this negative BOLD response was shown to be restored by HBO (Figure 19B) or HBO + memantine (Figure 19A) treatment. Therefore, normalization of the BOLD response was achieved by treatment that improved pattern discrimination ability (Figure 20).

[0109] (Protective effects of HBO and memantine against NMDA receptor inactivation by RT) Using mouse hippocampi, NMDAR activity was monitored in the dendritic region of CA1 pyramidal cells, rather than in the somatic cell region. As shown in Figure 24, the expression of the NR2A subunit was confirmed in the hippocampal CA1, DG, CA3, and CA2 regions.

[0110] Furthermore, as shown in Figures 25 and 26, the combined use of HBO and memantine (RT+HBO+Mema) was confirmed to have a significant improvement effect on RT-induced NMDAR inactivation in the hippocampal CA1 region (Figures 25B-F upper and 26 left). Similarly, the combined use of HBO and memantine (RT+HBO+Mema) was confirmed to have a similar effect on RT-induced NMDAR inactivation in the hippocampal DG region (Figures 25B-F lower and 26 right). These results indicate that neurogenesis in the hippocampal DG region plays an important role in protecting against RT-induced NMDAR inactivation.

[0111] Thus, it was confirmed that memantine promotes neurogenesis and, when used in combination with HBO, repairs neural networks and synaptic function, thereby efficiently improving cognitive function. [Industrial applicability]

[0112] The inventors of this invention have confirmed, using animal models, that the combined use of memantine and hyperbaric oxygen therapy (HBO) shows a remarkable effect on the radiation damage repair process. Furthermore, functional magnetic resonance imaging (fMRI) in humans showed that the combined use of memantine and HBO significantly accelerated the reconstruction of functional circuits, a feat not achieved with single-agent therapy, and a significant improvement was observed in the WAIS-R digit symbol test (DST), a neuropsychological index of white matter nerve conduction velocity. Therefore, it is believed that the composition and treatment method of this invention can be applied to cognitive impairment in various brain diseases.

Claims

1. A composition for the prevention or treatment of cognitive impairment caused by a decrease in hippocampal neurogenesis due to central nervous system toxicity caused by radiotherapy or chemotherapy, A composition characterized by containing memantine and being used in conjunction with hyperbaric oxygen therapy.

2. A method for producing a composition for the prevention or treatment of cognitive impairment resulting from a decrease in hippocampal neurogenesis caused by central nervous system toxicity due to radiotherapy or chemotherapy, and for use in conjunction with hyperbaric oxygen therapy, A method for producing a composition, characterized by including a step of incorporating memantine.

Images