METHOD FOR DETERMINING ALZHEIMER'S DISEASE RISK

Abstract

With the present invention, glycoalbumin and hemoglobin A1c are measured in a blood sample, and the incidence of or the existence or non-existence of a development risk of Alzheimer's disease can then be determined based on the calculated glycoalbumin/hemoglobin A1c ratio. Compounds for treating or preventing Alzheimer's disease can also be selected using this glycoalbumin/hemoglobin A1c ratio.

Claims

1. A method for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease, comprising a step of determining that a subject is affected with Alzheimer's disease or at risk of developing Alzheimer's disease when a glycoalbumin/hemoglobin A1c ratio in a blood sample from the subject is at least 2.6.

2. The method according to claim 1, wherein the blood sample is subjected to the following steps (1) to (3) prior to the step of determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease: 1) measuring glycoalbumin and hemoglobin A 1 c in the blood sample from the subject; 2) using the measurement values obtained in step 1) to calculate the glycoalbumin/hemoglobin A1c ratio; 3) comparing the glycoalbumin/hemoglobin A1c ratio obtained in step 2) with a cut-off value for predicting the incidence of or the existence or non-existence of a development risk of Alzheimer's disease.

3. A method according to claim 1, wherein the glycoalbumin/hemoglobin A1c ratio is at least 2.85.

4. A method according to claim 1, wherein the subject does not suffer from mild cognitive impairment (MCI).

5. A method according to claim 1, wherein the subject does not suffer from vascular dementia.

6. A biomarker for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease, wherein a glycoalbumin/hemoglobin A1c ratio is used as an indicator.

7. A device for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease, having: (a) a calculation unit for calculating a glycoalbumin/hemoglobin A1c ratio based on measurement values for glycoalbumin and hemoglobin A1c; and (b) an evaluation part for evaluating the incidence of or the existence or non-existence of a development risk of Alzheimer's disease based on the calculated glycoalbumin/hemoglobin A1c ratio.

8. The device according to claim 7, wherein the device further has a measuring part for glycoalbumin and/or a measuring part for hemoglobin A1c.

9. A kit for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease, comprising a glycoalbumin measurement reagent and a hemoglobin A1c measurement reagent, wherein the kit comprises a manual stating to the effect that a subject is determined to be affected with Alzheimer's disease or at risk of developing Alzheimer's disease when a glycoalbumin/hemoglobin A1c ratio is at least 2.6.

10. A method for screening substances for use in treating or preventing Alzheimer's disease, comprising a step of selecting a substance that reduces a glycoalbumin/hemoglobin level in a blood sample from a subject as a candidate substance.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0097] FIG. 1 shows the hazard ratios for Alzheimer's disease onset according to the levels of blood glucose-associated indicators based on Example 1 of the present invention.

[0098] FIG. 2 shows hazard ratios of blood glucose-associated indicator levels for Alzheimer's disease onset according to the presence or absence of abnormal glucose metabolism (diabetes +prediabetes) based on Example 2 of the present invention.

[0099] FIG. 3 shows the relationship between Alzheimer's disease and other diseases, and the relationships between these diseases and various measurement methods. Mild cognitive impairment (MCI) here means a condition in which multiple cognitive functions having reached normalcy are then lowered by some cause, but without becoming an obstacle to everyday life and social life. The disease concept of MCI and the disease concepts of mild neurocognitive decline (MNCD) according to the DSM-4 and mild neurocognitive disorder (MNCD) according to the DSM-5 are all similar. On the other hand, dementia is a condition in which either mild cognitive impairment or more serious cognitive decline is present and becomes an obstacle everyday life or social life. Types of dementia include Alzheimer's disease and vascular dementia. The diagnostic criteria for dementia used in the present study are the DSM-IIIR criteria for dementia, the NINCDS-ADRDA criteria for Alzheimer's disease and the NINDS-AIREN criteria for vascular dementia. Moreover, the MMSE, MoCA and revised Hasegawa's Dementia Scale (HDS-R) are tests of cognitive function, and are all three equivalent (MMSE MoCA HDS-R). The range of cognitive decline that can be confirmed with these tests is limited to MCI and dementia, and no one test can identify all diseases. However, the results of the present study can specifically identify Alzheimer's disease.

[0100] FIG. 4 is a block diagram showing the functional configuration of a device for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease in the present invention.

DESCRIPTION OF EMBODIMENTS

[0101] The present invention is explained in more detail below.

(1) Glycoalbumin

[0102] Glycoalbumin in the present invention is a form of albumin comprising glucose bound non-enzymatically to albumin, and may be called glycoalbumin or glycated albumin. Glycoalbumin can be measured by HPLC or by immune methods or enzymatic methods for example. Currently, enzymatic methods are widely used for measuring glycoalbumin. Glycoalbumin measurement in the present invention includes not only measuring the glycoalbumin concentration in a biological sample, but also measuring the glycoalbumin concentration as a percentage of the total albumin in a biological sample.

[0103] In enzymatic methods, for example the glycoalbumin in a biological sample is first digested with a protease to produce glycated amino acids. These glycated amino acids are then reacted with ketoamine oxidase, and the resulting hydrogen peroxide is reacted with a peroxidase in the presence of a hydrogen donor and a dye to quantify the glycated amino acids. The glycoalbumin concentration in a biological sample can also be determined from the quantified glycated amino acids by using a known calibrator or the like. The albumin concentration in the same biological sample is determined by an albumin quantification method such as an improved BCP assay, and the glycoalbumin concentration is divided by the albumin concentration to determine the percentage (%) of glycoalbumin to total albumin. In Japan, standard methods for measuring glycoalbumin are recommended by the Japan Society of Clinical Chemistry (Non-Patent Document 4), and it is desirable to use sufficiently accurate and precise measurement methods conforming to standard methods of glycoalbumin measurement.

[0104] In this Description, glycoalbumin is represented as an HPLC-traceable value (%), but when using another unit or a value traceable to another method, these can be converted appropriately using Formula 6 above or the like. Moreover, glycoalbumin is sometimes abbreviated as GA in the present invention.

(2) Hemoglobin A1c

[0105] In the present invention, hemoglobin A1c is a form of hemoglobin comprising glucose non-enzymatically bound to the N-terminal valine of the hemoglobin chain, and may be called hemoglobin A1c, HbA1c, glycohemoglobin, glycohemoglobin A1c, A1C or the like. Hemoglobin A1c can be measured by HPLC or by an immune method or enzymatic method for example. At present, HPLC methods, immune methods and enzymatic methods are all widely used for measuring hemoglobin A1c. Hemoglobin A1c measurement in the present invention includes not only measuring the hemoglobin A1c in a biological sample, but also measuring the hemoglobin A1c as a percentage of the total hemoglobin in the biological sample.

[0106] In HPLC methods, hemoglobin A1c in which the N-terminal valine of the chain is glycated can be separated and measured with a cation exchange column for example, and the percentage (%) of hemoglobin Aic relative to total hemoglobin in the biological sample can be determined. In immunological methods, a peptide bound to the glucose of the chain N-terminal valine is used as an antigen, and measurement is performed with the resulting antibody. In enzymatic methods, a glycated peptide excised using a protease that cleaves the peptide at the N-terminal of the chain is conducted to a color development system using fructosyl peptide oxidase to quantify the hemoglobin A1c. Hemoglobin A1c measurement values have been standardized by the Japan Diabetes Society, the Japan Society of Clinical Chemistry and the US National Glycohemoglobin Standardization Program (NGSP) and the like, and it is desirable to use sufficiently accurate and precise measurement methods conforming to standard methods. In this Description, glycoalbumin is represented as an NGSP-traceable value (%), but when using another unit or a value traceable to another method, these can be converted appropriately using Formula 1 to 5 or the like above. Moreover, hemoglobin A1c is sometimes abbreviated as HbA1c or A1C in the present invention.

[0107] The glycoalbumin/hemoglobin A1c ratio in the present invention is a ratio obtained by dividing the measured glycoalbumin value by the measured hemoglobin A1c value.

(3) Determining the Incidence of or Risk of Developing Alzheimer's Disease

[0108] Determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease means the act of evaluating the risk (likelihood) that a subject currently suffers from Alzheimer's disease and the risk (likelihood) of developing Alzheimer's disease when the subject does not suffer from Alzheimer's disease but may develop it in the future. The incidence of or the existence or non-existence of a development risk of Alzheimer's disease is normally determined by statistical methods as an odds ratio or hazard ratio showing increased frequency given 1 as the risk when a certain item is normal.

[0109] The Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA) and revised Hasegawa's Dementia Scale (HDS-R) can only identify mild cognitive impairment (MCI) and some forms of dementia. On the other hand, the present invention can identify Alzheimer's disease specifically. That is, the Alzheimer's disease targeted by the present invention is distinguished from mild cognitive impairment (MCI) or vascular dementia (FIG. 3). The disease concept of mild cognitive impairment and the disease concepts of mild neurocognitive decline (MNCD) according to the DSM-4 and mild neurocognitive disorder (MNCD) according to the DSM-5 are all almost identical.

[0110] The present invention provides a method for predicting the incidence of or the existence or non-existence of a development risk of Alzheimer's disease using the glycoalbumin/hemoglobin A1c ratio in a biological sample as an indicator. The present invention may be implemented alone or in combination with another method. For example, using the method of the present invention together with a commonly-used method of determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease is desirable from the standpoint of increasing the prediction accuracy.

[0111] In commonly-used methods, the incidence of or the existence or non-existence of a development risk of Alzheimer's disease is determined by considering low amyloid protein concentrations and elevated tau protein concentration in cerebrospinal fluid (CSF), genetic risk factors (apolipoprotein EE4), vascular risk factors (high blood pressure, diabetes, hypercholesterolemia) and smoking and the like, and also after considering such protective factors as regular exercise, dietary factors, leisure activities, social participation, active spiritual activity, cognitive training and moderate alcohol consumption.

[0112] A cut-off value for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease is explained in detail below.

[0113] It is generally known that a glycoalbumin/hemoglobin A1c (JDS) ratio of about 3 in blood from a diabetes patient is a characteristic value when blood sugar is stabilized (Non-Patent Document 5).

[0114] In the present invention, glycoalbumin/hemoglobin A1c (NGSP) ratios were determined from measured values for glycoalbumin and hemoglobin A1c (NGSP), and these ratios were compared with separately performed Alzheimer's disease diagnoses, and the relationship was investigated. The characteristic values of about 2 to 4 for glycoalbumin/hemoglobin A1c (NGSP) ratio may vary depending on race, gender, age and target population, and the measurement units and the like may also change in the future due to international standardization.

[0115] A cut-off value for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease is a reference value for glycoalbumin/hemoglobin A1c ratio used to assess the (percentage) likelihood that a subject currently suffers from Alzheimer's disease, or will suffer in the future if they do not at present, or will develop Alzheimer's disease in the future. There may be single cut-off value or multiple cut-off values.

[0116] Any number of glycoalbumin/hemoglobin A1c (NGSP) ratio between 2 and 4, or preferably 2.6 or above such as between 2.6 and 3.2 or preferably between 2.7 and 3.1 can be used as a cut-off value. The values of 2.85 or 2.83 shown in the examples may be used, but since these values may change slightly under examination, any numbers near these numbers may be used.

[0117] The glycoalbumin/hemoglobin A1c (JDS) ratios of healthy subjects and diabetes patients with stable blood glucose are characteristically about 3.0 (or 2.83 or 2.85 using the NGSP values in the examples of this application), and since the glycoalbumin/hemoglobin A1c (NGSP) ratio diverges from the characteristic values as the likelihood of having Alzheimer's disease or the existence or non-existence of a development risk of it in the future increases, a cut-off value may be 1 or more numbers selected from 2.6, 2.7, 2.8, 2.83, 2.85, 2.9, 3.0, 3.1 and 3.2, or another value may be used.

[0118] Alternatively, in at least 20 (or preferably at least 30, or more preferably at least 50, or most preferably at least 70) patients, a cut-off value can be obtained by measuring glycoalbumin and hemoglobin A1c (NGSP), determining the glycoalbumin/hemoglobin A1c (NGSP) ratios, and comparing these with the results for healthy subjects not suffering from Alzheimer's disease and cognitive syndrome according to the NINCDS-ADRDA diagnostic criteria for Alzheimer's disease. A cut-off value can also be obtained by known statistical methods (ROC analysis, etc.) (Examples). Since a cut-off value of 2.85 for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease was obtained by such methods in the Examples of this Description, a cut-off value of 2.85 is preferably selected for evaluating the the incidence of or the existence or non-existence of a development risk of Alzheimer's disease.

[0119] The comparison of measured glycoalbumin/hemoglobin A1c (NGSP) ratios with a cut-off value for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease, and the determination of the incidence of or the existence or non-existence of a development risk of Alzheimer's disease based on this comparison, are explained in detail below.

[0120] When the cut-off value for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease is a single number, it is possible to predict that a subject is affected with Alzheimer's disease or at high risk of developing Alzheimer's disease in the future even if they do not have it at present if the ratio of the measured values for glycoalbumin and hemoglobin A1c in a biological sample from that subject is at or above the cut-off value, and to predict a low likelihood of Alzheimer's disease when the ratio is below the cut-off value.

[0121] For example, if the cut-off value for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease is 2.85, there is a high risk that the subject has Alzheimer's disease or may develop it in the future even if they do not have it now if the glycoalbumin/hemoglobin A1c (NGSP) ratio measured in a biological sample is 2.85 or more, while if the ratio is less than 2.85 a low risk of having Alzheimer's disease or developing it in the future if it is not currently present can be predicted.

[0122] When there are multiple cut-off values for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease, a multistage determination of the incidence of or the existence or non-existence of a development risk of Alzheimer's disease may be made. For example, if the cut-off values are set at A1, A2, A3 . . . An, the ranges of glycoalbumin/hemoglobin A1c (NGSP) ratios can be set at less than A1, A1 to A2, A2 to A3 . . . An1 to An and above An, and the likelihood that Alzheimer's disease is currently present or that the subject will develop Alzheimer's disease in the future can be predicted to be higher the higher the numerical range. For example, if the cut-off values for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease are set at 2.60, 2.85 and 3.00, it can be predicted that when the glycoalbumin/hemoglobin A1c (NGSP) ratio measured in a biological sample is at least 3.00, there is a extremely high likelihood that Alzheimer's disease is currently present, and an extremely high likelihood of developing Alzheimer's disease in the future, while when it is at least 2.85 and less than 3.00, there is a high likelihood of Alzheimer's disease, or a high likelihood of developing it in the future, when it is at least 2.60 and less than 2.85, there is a possibility that Alzheimer's disease is currently present, and a possibility of developing it in the future, and when it is less than 2.60, there is a low likelihood that Alzheimer's disease is currently present, and a low likelihood of developing it in the future.

[0123] The method for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease based on the results of a comparison of the glycoalbumin/hemoglobin A1c (NGSP) ratio measured in a biological sample with a cut-off value for the incidence of or the existence or non-existence of a development risk of Alzheimer's disease is not limited to the method described above, and any method for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease based on the results of a comparison of the glycoalbumin/hemoglobin A1c ratio measured in a biological sample with a cut-off value for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease is acceptable.

[0124] (5) Device of Invention

[0125] The device of the present invention for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease may be any device provided with a glycoalbumin measurement part, a hemoglobin A1c measurement part, an operation unit for calculating a measured glycoalbumin/hemoglobin A1c ratio, and an evaluation part for making predictions based on the calculated glycoalbumin/hemoglobin A1c ratio. The measurement parts, operation unit, evaluation part and the like constituting this device do not all have to be installed in the same device, and may each be separate units.

[0126] The glycoalbumin measurement part is a site for measuring glycoalbumin in a biological sample introduced into the device. The hemoglobin A1c measurement part is a site for measuring hemoglobin A1c in a biological sample introduced into the device. The sizes and compositions of these sites are not particularly limited, and the measurements are not particularly limited as long as the respective proteins can be measured in a biological sample, and for example the common glycoalbumin and hemoglobin A1c measurement methods described above may be used.

[0127] In addition to HPLC methods, immune methods and enzymatic methods, electrophoresis methods, electrode methods, column methods and the like may also be used for measuring glycoalbumin, and other methods may also be used. For example, glycoalbumin may be measured by a widely-used enzymatic method. In enzymatic methods, for example the glycoalbumin in a target sample is first hydrolyzed with a protease, after which the hydrogen peroxide produced by reacting ketoamine oxidase with glycated lysine is reacted with a dye and quantified to determine the glycoalbumin concentration, while the albumin concentration in the same target sample is measured, and the glycoalbumin concentration is divided by the albumin concentration to obtain the glycoalbumin value (% or mmol/mol), but other commercial reagents and known methods and reagents may be used. The measured values for glycoalbumin in the present invention are based on HPLC methods, but values standardized by the ID/MS method or other values may be used, and the cut-off values for these can be converted to the values of the present invention using the relational formulae for the respective values.

[0128] A widely used enzymatic method, HPLC method, immune method, column method or electrode method may also be used to measure hemoglobin A1c. In the HPLC method, hemoglobin A1c can be separated and fractionated with an ion-exchange column for example, and the concentration (%) calculated. NGSP values are used for the hemoglobin A1c measurements in the present invention, but IFCC or other values may also be used, and the cut-off values and the like can be converted to the values of the present invention using the relational formulae for the respective values.

[0129] The operation unit is a site for dividing the glycoalbumin value obtained by measurement in the glycoalbumin measurement part by the hemoglobin A1c value obtained by measurement in the hemoglobin A1c measurement part, and may be a central processing unit (CPU) provided as part of the device for example. After measurement and prior to the operation, the glycoalbumin value and hemoglobin A1c value may also be stored in a storage device such as a memory or hard disk provided as part of the device, or the divided value from the operation unit may be stored temporarily in such a storage device.

[0130] The evaluation part is a site for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease based on the glycoalbumin/hemoglobin A1c ratio obtained by the calculation part.

[0131] In the present device, a determination of the incidence of or the existence or non-existence of a development risk of Alzheimer's disease can be made by comparing the glycoalbumin/hemoglobin A1c ratio obtained by the calculation part with a cut-off value for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease. The cut-off value for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease may have already been stored in the device, or may be input through an input part of the device when making a prediction. If for example the comparison shows that the glycoalbumin/hemoglobin A1c level obtained by the calculation part is at or above the cut-off value for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease, it can be predicted that the individual who provided the biological sample is likely to be affected with Alzheimer's disease or at high risk of developing Alzheimer's disease in the future, while conversely if the level is below the cut-off value for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease, it can be predicted that the individual who provided the biological sample is unlikely to be affected with Alzheimer's disease or at low risk of developing Alzheimer's disease in the future. One cut-off value or multiple values for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease may be set in the device, and when multiple cut-off values have been set a multi-stage prediction can be made as described above.

[0132] The device may also have an output part. The output part performs processing such as displaying or outputting the results of the determination of incidence of or the existence or non-existence of a development risk of Alzheimer's disease on a display device such as a screen or on a printing device such as a printer.

[0133] From the standpoint of convenience, the device is preferably a POC testing device (in general, a device for testing immediately near the patient in a clinical setting, such as at the bedside). It may also be a handy measurement device which the patient can use at home.

[0134] FIG. 4 is a block diagram showing the functional configuration of a device 10 for determining incidence of or the existence or non-existence of a development risk of Alzheimer's disease in the present invention. As shown in the diagram, the device 10 is provided with an input device 11, an input part 12, a memory part (first memory part, second memory part) 13, a calculation part (first calculation part, second calculation part) 14, an evaluation part 15, an output part 16 and an output device 17.

[0135] The device 10 of the present invention may for example be a specific program executed on a general-purpose personal computer. The input part 12, calculation part 14, evaluation part 15 and output part 16 thus represent operation modules executed by the computer's processor according to the program, and in fact these together constitute the processor of the device 10 of the present invention.

[0136] The memory part 13 is a memory device such as a hard disk of the device 10 of the present invention. The input device 11 is an input means such as a keyboard, mouse, touch panel or the like whereby the user sends processing instructions to the device 10 of the present invention, or inputs data and parameters. Data can also be read from a memory medium or the like via a USB (universal serial bus) interface. The operations performed by the user via the input device 11 are controlled by the input part 12. The output device 17 is a display device, printer or the like. Output processing to the output device 17 is controlled by the output part 16.

[0137] The user uses the input device 11 to input the glycoalbumin values of blood samples collected from subjects and the hemoglobin A1c values of blood samples collected from subjects.

[0138] The memory part 13 records the input glycoalbumin values and hemoglobin A1c values, but may also record the arithmetic expressions and algorithms used in evaluation by the operations part and the calculation results and the like.

[0139] The calculation part 14 calculates the glycoalbumin/hemoglobin A1c ratio.

[0140] The glycoalbumin/hemoglobin A1c ratio calculated by the calculation part 14 is output to the output device 17 via the output part 16.

[0141] The evaluation part 15 evaluates whether or not the glycoalbumin/hemoglobin A1c ratio calculated by the calculation part 14 exceeds a specific threshold. When the evaluation part 15 judges that the glycoalbumin/hemoglobin A1c ratio is at or above a specific threshold, an evaluation result to the effect that the subject may be affected with or at risk of developing Alzheimer's disease is output to the output device 17 via the output part 16.

[0142] The glycoalbumin measurement reagent contained in the kit for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease of the present invention is not particularly limited as to the measurement method (immune method, enzymatic method, chemical method, etc.) or constitution of the reagent as long as the reagent can measure glycoalbumin in a biological sample. An enzymatic method or immune method is desirable as the measurement method, and for example a Lucica GA-L (Asahi Kasei Pharma Corporation) reagent using enzymatic methods is a preferred embodiment of the glycoalbumin measurement reagent.

[0143] The hemoglobin A1c measurement reagent contained in the kit for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease of the present invention is not particularly limited as to the measurement method (immune method, enzymatic method, etc.) or constitution of the reagent as long as the reagent can measure hemoglobin A1c in a biological sample. Hemoglobin A1c measurement has been standardized by the Japan Diabetes Society and related societies, and measurements confirming to these standards are preferred. For example, CinQ HbA1c (Arkray, Inc.) is an example using an enzymatic method.

[0144] The kit of the present invention for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease is used to determine the the incidence of or the existence or non-existence of a development risk of Alzheimer's disease using as an indicator the ratio (glycoalbumin/hemoglobin A1c ratio) of the glycoalbumin measured in a biological sample with the glycoalbumin measurement reagent contained in the kit to the hemoglobin A1c measured in a biological sample with the hemoglobin A1c measurement reagent contained in the kit. As discussed above, the risk is determined by comparing the ratio to a cut-off value for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease.

[0145] The biological sample to be measured in the present invention may be any test liquid containing at least hemoglobin and albumin. Preferred examples of test liquids include blood components such as serum, plasma, blood cells, whole blood, or isolated red blood cells or the like. Serum or plasma that has been separated by ordinary methods may also be used. The biological sample may be easily obtained by well-known methods or the like, and may also be a biological sample that has been pre-treated by ordinary methods.

[0146] The present invention provides a preventative and/or treatment agent for dementia or Alzheimer's disease, containing as an active ingredient a substance selected from the group consisting of the diabetes treatment drugs (hypoglycemic drugs, insulin resistance improving drugs, etc.) that has the effect of at least suppressing a rise in the glycoalbumin/hemoglobin A1c ratio.

[0147] The present invention also provides an agent that inhibits oxidative stress, containing as an active ingredient a substance that has the effect of suppressing production of at least one substance selected from the group consisting of glycoalbumin, hemoglobin A1c and the glycoalbumin/hemoglobin A1c ratio.

[0148] The present invention also provides an agent that lowers blood glucose, containing as an active ingredient a substance that has the effect of suppressing the expression of at least one selected from the group consisting of glycoalbumin, hemoglobin A1c and the glycoalbumin/hemoglobin A1c ratio.

[0149] These agents may be used as drugs or as reagents for experimental purposes.

[0150] Examples of the substance that has the effect of suppressing the expression of at least one selected from the group consisting of glycoalbumin, hemoglobin A1c and the glycoalbumin/hemoglobin A1c ratio include uric acid, uric acid lowering drugs such as xanthine oxidase inhibitors, hypoglycemic drugs such as Metformin, SGLT2 inhibitors, Pioglitazone and DPP4 inhibitors, and insulin resistance improvers. Various compounds that promote pancreatic 13 cell proliferation, and insulin and insulin derivatives, may also be used in the same way.

[0151] One kind of substance that has the effect of suppressing the expression of at least one selected from the group consisting of glycoalbumin, hemoglobin A1c and the glycoalbumin/hemoglobin A1c ratio may be used, or multiple kinds may be combined.

[0152] A substance having the effect of suppressing the expression of at least one selected from the group consisting of glycoalbumin, hemoglobin A1c and the glycoalbumin/hemoglobin A1c ratio may be administered orally or non-orally to a mammal (such as a human, rabbit, dog, cat, rat or mouse), either alone or as a pharmaceutical composition in an appropriate dosage form. The dosage differs depending on the administration target, type of disease, symptoms, administration route and the like, but for example in the case of an adult, the substance that has the effect of suppressing the expression of at least one selected from the group consisting of glycoalbumin, hemoglobin A1c and the glycoalbumin/hemoglobin A1c ratio may be administered in the amount of normally about 0.1 to 100 mg/kg body weight, or preferably about 0.5 to 50 mg/kg body weight as a single dose about 1 to 2 times per month, and preferably the same dose can be administered for 2 to 3 days consecutively at the beginning of treatment. A corresponding dose can be administered in the case of other non-oral or oral administration. The dose can also be increased according to the symptoms when the symptoms are particularly severe.

[0153] A composition for oral administration may be in a solid or liquid dosage form, and specific examples include tablets (including sugar-coated tablets and film-coated tablets), pills, granules, powders, capsules (including soft capsules), syrups, emulsions, suspensions and the like. Such a composition may be manufactured by ordinary methods, and may contain carriers, diluents or excipients commonly used in the pharmaceutical field. Examples of pharmaceutical carriers and excipients include lactose, starch, sucrose, magnesium stearate and the like.

[0154] Examples of compositions for non-oral administration include injections, suppositories and the like, and an injection may be in the form of an intravenous injection, subcutaneous injection, intradermal injection, intramuscular injection, drip injection or the like. Such an injection can be prepared by dissolving, suspending or emulsifying a substance that has the effect of suppressing the expression of at least one selected from the group consisting of glycoalbumin, hemoglobin A1c and the glycoalbumin/hemoglobin A1c ratio in a sterile aqueous or oily liquid commonly used for injections. Examples of aqueous liquids for injection include physiological saline and isotonic liquids containing glucose and other adjuvants, and a suitable solubilizing agent such as an alcohol (such as ethanol), polyalcohol (such as propylene glycol or polyethylene glycol), nonionic surfactant (such as Polysorbate 80 or HCO-50 (polyoxyethylene (50 mol) adduct of hydrogenated castor oil)) or the like may also be included. Examples of oily liquids include sesame oil and soy bean oil, and benzyl benzoate, benzyl alcohol or the like may also be included as a solubilizing agent. The prepared injection is filled in a suitable common ampoule. A suppository for rectal administration can be prepared by mixing the substance having the effect of suppressing Smad1 expression with a suitable suppository base.

[0155] This oral or non-oral drug composition can be prepared in a unit dosage form suited to the administered dose of the active ingredient. Examples of unit dosage forms include tablets, pills, capsules, injections (ampoules), suppositories and the like.

[0156] Moreover, this drug composition may also contain other active ingredients to the extent that these do not cause unwanted interactions when compounded with the substance that has the effect of suppressing the expression of at least one selected from the group consisting of glycoalbumin, hemoglobin A1c and the glycoalbumin/hemoglobin A1c ratio.

[0157] To screen candidate compounds from clinical studies, the glycoalbumin/hemoglobin A1c ratio is diagnosed in target cases of diabetes, impaired glucose tolerance (IGT) and impaired fasting glycemia (IFG), the administered drugs and environmental factors such as exercise and diet are tracked throughout the investigation period, and the glycoalbumin/hemoglobin A1c ratio is measured for a period of at least a year after the start of follow-up. Drugs that are shown by frequency method analysis to lower the glycoalbumin/hemoglobin A1c level in these target cases are subjected to drug repositioning and evaluated as compounds exhibiting effectiveness in treating or preventing dementia or Alzheimer's disease, or as seed compounds for potential use as active ingredients in drugs for treating or preventing dementia or Alzheimer's disease.

[0158] The present invention is explained in more detail below using examples, but the present invention is not limited to these examples.

EXAMPLES

Example 1

[0159] Hazard Ratios for Alzheimer's disease Onset According to Levels of Blood Glucose-Associated Indicators

[0160] Of the residents aged 65 and older who received a cardiovascular checkup in Hisayama-machi in 2007, 1,187 individuals without dementia were followed prospectivelyf for 5 years. The subjects were separated into 4 quartiles (Q1 to Q4) according to their hemoglobin A1c, glycoalbumin and 1,5-anhydroglucitol levels and glycoalbumin/hemoglobin A1c ratios. The end point was the onset of Alzheimer's disease or vascular dementia. The Cox proportional hazard model was used to calculate the hazard ratio (HR). Alzheimer's disease was diagnosed by the diagnostic criteria of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA) (Non-Patent Document 2: Guy McKhann et al., Clinical diagnosis of Alzheimer's disease, Neurology 1984, 34:939-944), and vascular dementia by the clinical standard of the National Institute of Neurological Disorders and Stroke and Association Internationale pour la Recherche et l'Enseignement en Neurosciences (NINDS-AIREN) (Roman G C, Tatemichi T K, Erkinjuntti T et al., Vascular dementia: diagnostic criteria for research studies: report of the NINDS-AIREN International Workshop, Neurology 1993, 43:250-260).

[0161] The following conditions 1) to 4) must be met in order for a subject to be diagnosed with Alzheimer's disease according to the NINCDS-ADRDA diagnostic criteria.

[0162] 1) Meets the diagnostic criteria for clinical examination (probable Alzheimer's disease)

[0163] 2) Progressive impairment or the like of specific cognitive function is confirmed, and findings support Alzheimer's disease

[0164] 3) There are other clinical features consistent with probable Alzheimer's disease once causes of dementia other than Alzheimer's disease have been excluded

[0165] 4) Cases in which probable Alzheimer's disease diagnosis is doubtful, or patients with features that do not match probable Alzheimer's disease are excluded from probable Alzheimer's disease diagnosis

[0166] In this study, cranial imaging (CT/MRI) was also performed to obtain findings (see 2 above) supporting the NINCDS-ADRDA diagnosis of probable Alzheimer's disease. Specifically, patients who exhibited no clear organic lesions typical of cerebral vascular disease and who exhibited cerebral atrophy (especially internally, in the temporal and parietal lobe regions) were diagnosed with probable Alzheimer's disease.

[0167] 91.4% of the Alzheimer's disease diagnoses in the current study were cases of probable Alzheimer's disease according to NINCDS-ADRDA, and included some subjects who died during the follow-up period and received a conclusive diagnosis of definite Alzheimer's disease based on an autopsy. That is, the 91.4% are subjects who:

[0168] 1) Fulfill the diagnostic criteria for clinical examination (probable Alzheimer's disease), that is: [0169] Dementia found in clinical testing and testing by the Mini-Mental State Examination (MMSE), Blessed Dementia Scale or similar, and confirmed by neuropsychological testing [0170] Deficits in two or more cognitive areas [0171] Progressive decline in memory and other cognitive functions [0172] Consciousness not impaired [0173] Onset at age 40 to 90, most common after age 65 [0174] Progressive memory and cognitive impairment not attributable to any other system or brain disease.

[0175] 2) Supporting findings are: [0176] Progressive impairment in specific cognitive functions: language impairment (aphasia), behavioral impairment (apraxia), impaired recognition (agnosia) and impairment of other activities of daily living (ADL), changes in behavioral patterns, family history of similar impairment, especially when there is neuropathological confirmation, and [0177] As a clinical test finding, brain atrophy or the like is confirmed by cranial imaging, or in other words by CT/MIR or the like.

[0178] 3) There are other clinical features consistent with the diagnosis of probable Alzheimer's disease once causes of dementia other than Alzheimer's disease have been excluded

[0179] 4) There are no features that call the diagnosis of probable Alzheimer's disease into question, or that are not characteristic of probable Alzheimer's disease.

[0180] Regarding the age range given in the 1) diagnostic criteria for clinical examination (probable Alzheimer's disease), this was re-interpreted as onset at age 40 or above for purposes of the diagnosis because subjects who may have developed Alzheimer's disease at age 90 or above could be included.

[0181] Finally, probable Alzheimer's disease was diagnosed after a detail examination of the subject's medical history and symptomology.

[0182] The remaining 8.6% of subjects diagnosed with Alzheimer's disease were a group of subjects who could not be examined by cranial imaging, but who were diagnosed with Alzheimer's disease based on medical history and symptomology (current symptoms), and who came very close to meeting the NINCDS-ADRDA criteria for probable Alzheimer's disease. More specifically, although clinical test findings have been established as supporting findings for a NINCDS-ADRDA diagnosis of probable Alzheimer's disease, the 8.6% were those who were diagnosed with Alzheimer's disease after a detailed investigation of medical history and symptomology because MRI imaging could not be performed even though they fulfilled all the other conditions. Medical history here means the history of previous diseases and the history of the current disease. Symptoms here mean pathological changes appearing in the body, and symptomatology in the case of Alzheimer's disease symptomology is a field of medicine and is a methodology for defining, classifying and interpreting the various complaints and medical examination findings (and symptoms) of a patient.

[0183] In other words, the most important factors in Alzheimer's disease diagnosis are medical history and symptomology (symptoms displayed), and the collecting of these is emphasized in the present study. Moreover, other clinical data was collected as much as possible and cranial imaging was performed to support the diagnosis of probable Alzheimer's disease, and when the subject died an autopsy was performed to obtain a final diagnosis of definite Alzheimer's disease, thereby completing a 3-stage Alzheimer's disease diagnosis.

[0184] During the follow-up period 116 subjects developed Alzheimer's disease, and 43 developed vascular dementia. The incidences (sex- and age-adjusted) of Alzheimer's disease and vascular dementia were investigated against the levels of each blood glucose-associated indicator. The results show that the Alzheimer's disease incidence rose significantly as the glycoalbumin/hemoglobin A1c levels rose, and although not significant, the same tendency was shown for glycoalbumin. However, no clear relationship was shown for hemoglobin A1c and 1,5-anhydroglucitol. No relationship was shown between any of these blood glucose-associated indicators and the incidence of vascular dementia. In a multivariate analysis, adjusting the variables of sex, age, high blood pressure, total cholesterol, body mass index (BMI, obesity) category (low body weight, normal body weight, obese), history of stroke, education, alcohol consumption, smoking and exercise habits had no effect on the significant positive relationship between glycoalbumin/hemoglobin A1c level and Alzheimer's disease (trend p=0.01), and the hazard ratio for Alzheimer's disease incidence was significantly higher in both the Q3 group (2.1) and Q4 group (2.0) in comparison with the Q1 group (FIG. 1, p<0.05).

[0185] These results show that the glycoalbumin/hemoglobin A1c ratio is a biomarker associated with the incidence of or the existence or non-existence of a development risk of Alzheimer's disease, and that the hazard of developing Alzheimer's disease is significantly higher when the glycoalbumin/hemoglobin A1c level is 2.83 or more, or in other words that the incidence of or the existence or non-existence of a development risk of Alzheimer's disease can be determined using the glycoalbumin/hemoglobin A1c level as an indicator.

[0186] This also shows that the risk of developing Alzheimer's disease can be determined based on measurement results obtained with a commercial glycoalbumin measurement kit and hemoglobin A1c measurement kit or device, and that a kit or device for determining the risk of Alzheimer's disease can be provided.

Example 2

Hazard Ratios of Blood Glucose-Associated Indicator Levels for Alzheimer's Disease According to Presence or Absence of Abnormal Glucose Metabolism

[0187] The hazard ratios (after multivariate adjustment) of each blood glucose-associated indicator level in Alzheimer's disease onset were calculated depending on the presence or absence of abnormal glucose metabolism (diabetes+pre-diabetes) under the same conditions as in Example 1. As a result, in the group without abnormal glucose metabolism the high glycoalbumin/hemoglobin A1c ratio groups (Q3 and Q4) had significantly higher hazard ratios (1.82, p=0.03) for Alzheimer's disease onset than the low-ratio groups (Q1 and Q2), and the same tendency was seen in the group with abnormal glucose metabolism (FIG. 2, hazard ratio 1.7, p=0.07). However, no significant relationship was found between Alzheimer's disease onset and hemoglobin A1c, glycoalbumin or 1,5-anhydroglucitol regardless of whether there was abnormal glucose metabolism (p>0.1 in all cases). There were no interactions between the blood glucose-associated indicator levels and abnormal glucose metabolism.

[0188] These results show that the glycoalbumin/hemoglobin A1c ratio can be used to determine the incidence of or the existence or non-existence of a development risk of Alzheimer's disease even in subjects with no abnormal glucose metabolism.

[0189] Because the incidence of or the existence or non-existence of a development risk of Alzheimer's disease can be determined based on measurement results obtained with a commercial glycoalbumin measurement kit or device and a commercial hemoglobin A1c measurement kit or device as in Example 1, the present invention can clearly provide a kit and device for determining the the incidence of or or the existence or non-existence of a development risk of Alzheimer's disease.

Example 3

Calculating Cut-Off Values for Glycoalbumin/Hemoglobin A1c Levels in Alzheimer's Disease Onset

[0190] Cut-off values and sensitivity/specificity of glycoalbumin and glycoalbumin/hemoglobin A1c levels were determined for Alzheimer's disease onset using ROC (receiver operating characteristics curve) analysis under the same conditions as in Example 1. The results are shown in Table 1.

TABLE-US-00001 TABLE 1 Cut-off values for blood glucose-associated indicators in Alzheimer's disease onset (1,187 Hisayama residents aged 65 and older, 2007-2012, ROC analysis) BG-associated indicator Cut-off value Sensitivity (%) Specificity (%) GA 15.8% 55.2% 59.9% GA/HbA1c 2.85 65.5% 53.3%

[0191] As shown in Table 1, the glycoalbumin/hemoglobin A1c ratio cut-off value for detecting Alzheimer's disease is 2.85, with a sensitivity of 65.5% and a specificity of 53.3%. These results show that the glycoalbumin/hemoglobin A1c ratio is a biomarker for Alzheimer's disease, and that the presence or absence of Alzheimer's disease can be predicted or the existence or non-existence of a development risk of Alzheimer's disease can be determined using a cut-off value for detecting the incidence of or the existence or non-existence of a development risk of Alzheimer's disease, or in other words that many tests can be performed at once easily and cheaply without the advanced and complex diagnosis and testing used in the past.

[0192] This also shows, similarly to Example 1, that the risk of developing Alzheimer's disease can be determined based on measurement results obtained with commercial glycoalbumin measurement kit or device and a commercial hemoglobin A1c measurement kit or device, and that a kit or device for determining the incidence of or the existence or non-existence of a development risk of Alzheimer's disease can be provided by the present invention.

[0193] In this Example 3, the presence or absence of Alzheimer's disease can be predicted using the glycoalbumin/hemoglobin A1c ratio as an indicator for example. Specifically, if the glycoalbumin/hemoglobin A1c ratio is at least 2.85, the likelihood that Alzheimer's disease is present can be judged to be high, while if it is less than 2.85 the likelihood can be judged to be low. If it is at least 2.85, the patient can then be diagnosed with Alzheimer's disease using brain imaging requiring expensive equipment, invasive cerebrospinal fluid testing, or the complex diagnostic criteria of NINCDS-ADRDA or the like, and treatment can be initiated as necessary. This means that with the present invention the likelihood of Alzheimer's disease can be predicted easily and cheaply without the need for brain imaging requiring expensive equipment, invasive cerebrospinal fluid testing, or elaborate and complex Alzheimer's disease diagnoses.

[0194] Moreover, even if a subject has been judged not to have Alzheimer's disease based on such a detailed Alzheimer's disease diagnosis, a subject with a glycoalbumin/hemoglobin A1c ratio of at least 2.85 has a high likelihood of developing Alzheimer's disease in the future. Consequently, with the present invention the likelihood of developing Alzheimer's disease in the future can be predicted and future onset can be prevented easily and cheaply without using brain imaging requiring expensive equipment, invasive cerebrospinal fluid testing, or the complex diagnostic criteria of NINCDS-ADRDA or the like, and patients can receive necessary health guidance and take measures to prevent onset.

INDUSTRIAL APPLICABILITY

[0195] In the present invention, the incidence of or the existence or non-existence of a development risk of Alzheimer's disease can be determined easily, rapidly and cheaply by measuring glycoalbumin and hemoglobin A1c and determining the glycoalbumin/hemoglobin A1c ratio. Moreover, in the present invention useful compounds for preventing or treating Alzheimer's disease can be selected easily, rapidly and cheaply. The method, device and kit of the present invention are useful for clinical testing and compound selection, and are useful for screening, diagnosis, risk assessment and prevention of Alzheimer's disease.

REFERENCE SIGNS LIST

[0196] 10 Blood sample device [0197] 11 Input device [0198] 12 Input part [0199] 13 Memory part [0200] 14 Calculation part [0201] 15 Evaluation part [0202] 16 Output part [0203] 17 Output device