KR102634018B1 - Amyloid PET positivity prediction model in patients with cognitive impairment and prediction method using the model - Google Patents

Abstract

The present invention relates to three brain amyloid PET positive prediction models according to cognitive function level that integrate demographic variables, blood-based markers, and neuropsychological measurement data that can generally be collected in clinical practice settings. The model of the present invention has greatly improved prediction accuracy compared to existing models, and has the advantage of being widely used as a screening tool because the model was created with data generally obtained in a clinical environment. Additionally, early prediction of brain amyloid PET positivity may be helpful in selecting cognitively impaired patients with a high positive rate of brain amyloid accumulation.

Description

Amyloid PET positivity prediction model in patients with cognitive impairment and prediction method using the same {Amyloid PET positivity prediction model in patients with cognitive impairment and prediction method using the model}

The present invention relates to a prediction model for amyloid PET positivity in patients with cognitive impairment and a prediction method using the same. More specifically, the present invention relates to age, gender, education, APOE4 positivity, MMSE score, and history of hypertension data collected from a group of cognitive impairment patients. This relates to a method and system for calculating the amyloid PET positivity rate (P) of a subject using an equation derived by modeling through multivariate logistic regression analysis, and predicting amyloid PET positivity using P.

Alzheimer's disease (AD) is the most common degenerative brain disease that causes dementia, and there are currently about 35 million people with Alzheimer's dementia worldwide. Symptoms of Alzheimer's dementia include not only cognitive decline but also mental behaviors such as personality changes, nervousness, depression, delusions, hallucinations, aggression, and sleep disorders. In the final stages of Alzheimer's dementia, neurological disorders such as stiffness and gait abnormalities appear, as well as physical disabilities such as incontinence of urine and feces, infections, and bedsores. Depending on the age of onset, Alzheimer's dementia is classified into early-onset (senior-onset) Alzheimer's dementia in those under 65 years of age and late-onset (senile-onset) Alzheimer's dementia in those over 65 years of age. Cases of Alzheimer's dementia have been reported even in relatively young people, and it is known to be the most common form of dementia in the elderly population, with the incidence doubling for each 5-year increase in age within the range of 65 to 85 years. .

When examining the brain tissue of a patient with Alzheimer's disease, amyloid plaques, neurofibrillary tangles, and neuronal cell death can be observed. The exact pathogenesis and cause of Alzheimer's dementia are not known, but currently, the core mechanism of its development is an imbalance of neurotransmitters and excessive production of amyloid β protein, which is deposited in the brain and causes toxicity. It is known as In addition, hyperphosphorylation of tau protein, neuroinflammation, and oxidative substances that cause brain cell damage are known to be the causes of Alzheimer's dementia. In patients with Alzheimer's disease, beta-amyloid protein and tau protein accumulate about 15 years before symptoms such as cognitive decline begin.

Because there is still no way to completely treat Alzheimer's dementia, it is very important to assess the risk of developing Alzheimer's dementia and diagnose it early. Alzheimer's dementia is diagnosed according to clinical diagnostic criteria in most clinical practice settings. However, the diagnostic method for Alzheimer's dementia based on clinical diagnostic criteria has the problem of low sensitivity (71-81%) and specificity (about 70%) because it does not reflect histopathologic findings. Cerebrospinal fluid (CSF) amyloid β (Aβ) testing is a test using a useful and inexpensive fluid biomarker for Aβ pathology. However, this test is not only invasive but also cannot be applied to patients with elevated intracranial pressure, coagulopathy, or patients using anticoagulants, so its actual clinical use is limited.

Meanwhile, the amyloid PET (amyloid positron emission tomography) test is a test using an imaging marker that can non-invasively check whether amyloid β (Aβ) is deposited in the patient's brain. The amyloid PET test is a method of confirming the presence of amyloid protein deposited in the brain by combining a substance that binds well with amyloid protein with a radioisotope and injecting it into the patient. Amyloid PET testing is useful for predicting the risk of patients with mild cognitive impairment (MCI) developing dementia in the future. Patients who are positive for amyloid PET have a risk of converting to dementia within 3 years about 3 to 4 times higher than patients who are negative. Therefore, amyloid PET testing is important in clinical practice to enable medical staff to predict and manage the prognosis of patients with mild cognitive impairment. However, because it is difficult to recommend to patients who do not have dementia due to its high cost, a model that can predict the amyloid PET positivity rate in advance is needed.

In 2013, the Alzheimer's Association and the Society for Nuclear Medicine and Molecular Imaging convened the Amyloid Imaging Taskforce to collect empirical evidence supporting the clinical utility of amyloid imaging. Patients with persistent unexplained mild cognitive impairment (MCI), patients with an atypical presentation of Alzheimer's dementia (AD), and patients with symptoms of progressive neurocognitive disorder (NCD) at an atypically early age. Patients with this condition reported that performing amyloid imaging tests was clinically helpful. However, in order to determine whether to perform amyloid PET testing in actual clinical settings, specific guidelines including demographic and genetic factors are needed.

Several previous studies have created prediction models for brain amyloid PET positivity, but the accuracy is low, making it difficult to apply them as guidelines, variables that cannot be used clinically are applied to the model, or variables that cannot be used clinically are applied to the model, or there is a difficulty in applying them as a guideline, such as MCI or AD. There were limitations in presenting a model for disability. For example, in the case of the positive prediction model for brain amyloid accumulation presented in Patent Publication No. 10-2019-0063275, word list recall can be obtained from a detailed dementia diagnosis assessment such as CERAD-K, which takes about 40-50 minutes. ) score and constructional recall score are included as variables, and it is limited in that only models for MCI patients and Alzheimer's dementia patients are presented. In addition, the present inventor's previous model had the problem of low accuracy because it did not limit the type of dementia (Korean J Biol Psychiattry 2020;27(2):94-100).

Accordingly, the present invention attempted to develop a prediction model for amyloid PET positivity in various cognitive impairment patients using clinical data of cognitive impairment patients that can be obtained with a diagnosis time of about 5 minutes, and as a result, it was found that patients with three cognitive function states We developed a model to predict brain amyloid PET positivity in subjects, that is, all cognitive impairment patients suffering from cognitive decline, patients with mild cognitive impairment (MIC), and patients with Alzheimer's dementia (AD), and found that all three models above had high prediction accuracy. By confirming this, the present invention was completed.

The purpose of the present invention is to provide a prediction model for amyloid PET positivity in patients with cognitive impairment and a prediction method using the same.

In order to achieve the above object, the present invention provides a method of providing information for predicting whether amyloid PET is positive in a patient with cognitive impairment using the following equation.

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit])

[Equation 2]

Logit = 0.158 + 0.029*(age) + 0.247*(gender) + 0.034*(education) + 1.937*(APOE4 positive) + -0.155*(MMSE score)

In the present invention, P refers to the amyloid PET positivity rate.

In the present invention, when the P value is 0.556 or more, the patient may be determined to be positive for amyloid PET.

The present invention also provides a method of providing information for predicting whether amyloid PET is positive in a patient with mild cognitive impairment using the following equation.

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit]),

[Equation 3]

Logit = -3.668 + 0.037*(age) + -0.016*(gender) + 0.004*(education) + 1.915*(APOE4 positive),

In the present invention, P refers to the amyloid PET positivity rate.

In the present invention, when the P value is 0.504 or more, the patient may be determined to be positive for amyloid PET.

The present invention also provides a method of providing information for predicting whether a patient with Alzheimer's dementia is positive for amyloid PET using the following equation.

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit])

[Equation 4]

Logit = -0.012 + 0.039*(age) + 1.043*(gender) + 0.075*(education) + 2.631*(APOE4 positive) + -0.198*(MMSE score) + -1.615*(history of hypertension)

In the present invention, P refers to the amyloid PET positivity rate.

In the present invention, when the P value is 0.691 or more, the patient may be determined to be positive for amyloid PET.

The present invention also includes a calculation unit that calculates the amyloid PET positivity rate (P) value from the age, gender, education, APOE4 positivity, and MMSE score data of the subject to be analyzed using the following [Equation 1] and [Equation 2]; Provided is a system for predicting amyloid PET positivity in patients with cognitive impairment, including:

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit])

[Equation 2]

Logit = 0.158 + 0.029*(age) + 0.247*(gender) + 0.034*(education) + 1.937*(APOE4 positive) + -0.155*(MMSE score)

In the present invention, [Equation 1] and [Equation 2] are derived by modeling age, gender, education, APOE4 positivity, and MMSE score data collected from a group of cognitively impaired patients through multivariate logistic regression analysis. You can do this.

In the present invention, the system may further include a prediction unit that predicts whether amyloid PET is positive from the analyzed P value.

In the present invention, when the P value is 0.556 or more, the patient may be determined to be positive for amyloid PET.

The present invention also includes a calculation unit that calculates the amyloid PET positivity rate (P) value from the age, gender, education, and APOE4 positivity data of the subject to be analyzed using the following [Equation 1] and [Equation 3], We provide a prediction system for amyloid PET positivity in patients with mild cognitive impairment.

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit])

[Equation 3]

Logit = -3.668 + 0.037*(age) + -0.016*(gender) + 0.004*(education) + 1.915*(APOE4 positive)

In the present invention, the above [Equation 1] and [Equation 3] are characterized in that they are derived by modeling the age, gender, education level, and APOE4 positivity data collected from a group of patients with mild cognitive impairment through multivariate logistic regression analysis. You can.

In the present invention, the system may further include a prediction unit that predicts whether amyloid PET is positive from the analyzed P value.

In the present invention, when the P value is 0.504 or more, the patient may be determined to be positive for amyloid PET.

The present invention also uses the following [Equation 1] and [Equation 4] to calculate the amyloid PET positivity rate (P) value from the age, gender, education, APOE4 positivity, MMSE score, and hypertension history data of the subject to be analyzed. Provides a prediction system for amyloid PET positivity in patients with Alzheimer's dementia, including:

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit])

[Equation 4]

Logit = -0.012 + 0.039*(age) + 1.043*(gender) + 0.075*(education) + 2.631*(APOE4 positive) + -0.198*(MMSE score) + -1.615*(history of hypertension)

In the present invention, [Equation 1] and [Equation 4] are derived by modeling the age, gender, education, APOE4 positivity, MMSE score, and history of hypertension data collected from a group of Alzheimer's dementia patients through multivariate logistic regression analysis. It can be characterized as being.

In the present invention, the system may further include a prediction unit that predicts whether amyloid PET is positive from the analyzed P value.

In the present invention, if the P value is 0.691 or more, the patient may be determined to be positive for amyloid PET.

The amyloid PET positive prediction model for cognitive impairment patients of the present invention and the prediction method using the same are generally obtained in a clinical environment for three models for all cognitive impairment patients suffering from cognitive decline, patients with mild cognitive impairment, and patients with Alzheimer's dementia. It has the advantage of being able to easily and accurately predict whether brain amyloid PET is positive through a simple and rapid diagnosis in about 5 minutes using data.

Figure 1 presents receiver operating characteristic (ROC) curves for the three final prediction models for brain amyloid PET positivity. Curves are shown for (A) all cognitive impairment patients, (B) mild cognitive impairment (MIC) patients, and (C) Alzheimer's (AD) dementia patients.

Hereinafter, the present invention will be described in detail.

Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as commonly understood by a person skilled in the art to which the present invention pertains. In general, the nomenclature used herein is well known and commonly used in the art.

Several studies prior to the present invention proposed prediction models for brain amyloid PET positivity.

Bahar-Fuchs et al. reported that decline in episodic memory was related to brain amyloid PET positivity in 45 patients with mild cognitive impairment (MCI), focusing on neuropsychological test results (J Alzheimers Dis2013;33 :451-462). Additionally, several prediction models focusing on blood-based biomarkers have been reported. Some studies have used blood amyloid β to predict amyloid PET positivity in the patient's cerebrum. However, the blood-based biomarkers presented in the above studies are not commonly used in actual clinical settings, so their use has limitations.

Haghighi et al. developed an amyloid PET positivity prediction model by integrating neuropsychological tests and blood-based biomarkers in 168 patients with mild cognitive impairment (MCI) and 50 normal cognitive controls (J Alzheimers Dis2015;43:1261-1270 ). However, this study also included blood-based biomarkers that are not commonly available in clinical practice.

Lee et al.23 compared the age, gender, education, APOE4, and hypertension of 107 patients with mild cognitive impairment and 69 patients with early Alzheimer's dementia in the KBASE cohort and 332 patients with mild cognitive impairment and 144 patients with Alzheimer's dementia in the ADNI-2 cohort. A brain amyloid PET positivity prediction model was proposed using data commonly available in clinical practice, such as medical history and word list recall test scores (Front Aging Neurosci 2018;10:309). However, in the case of the brain amyloid PET positive prediction model, the word list recall score and constructional recall that can be obtained from a detailed cognitive function test for dementia diagnosis such as CERAD-K, which takes about 40-50 minutes There is a limitation in that the score is included as a variable and only models for patients with mild cognitive impairment and Alzheimer's dementia are presented.

Therefore, in the present invention, blood tests and mini mental state examination (MMSE) commonly used in clinical dementia evaluation and the sum of the box score for Clinical Dementia Rating (CDR-SB) are used. ) The present invention was completed by developing a model that can predict whether brain amyloid PET is positive for all cognitive impairment patients, mild cognitive impairment patients, and Alzheimer's dementia patients suffering from cognitive decline through measurement. Therefore, the present invention has the advantage of being able to easily and accurately predict whether brain amyloid PET is positive through a simple and rapid diagnosis in about 5 minutes using data generally obtained in actual clinical practice settings.

In the present invention, all equations used to predict whether amyloid PET is positive were derived through multivariate logistic regression analysis. Age, gender, and education level were inserted as fixed variables, and other variables well known as risk factors (e.g., body mass index, family history of dementia, history of hypertension, history of diabetes, MMSE, CDR-SB, APOE4 gene possession, total blood cholesterol) By sequentially inserting the blood thyroid-stimulating hormone level, blood vitamin B12 level, systolic blood pressure, and diastolic blood pressure), an optimized equation was completed to predict whether amyloid PET was positive.

In the present invention, the term “amyloid PET positivity” generally refers to a state in which a radioactive tracer that binds to amyloid beta protein in the patient's brain binds to gray matter.

In the present invention, the term “educational background” means years of education.

In the present invention, the term “cognitive disorder” refers to a condition in which memory, attention, language ability, visuospatial ability, judgment, etc. are reduced. The degree of cognitive impairment varies from very mild to severe. When cognitive dysfunction is so severe that it interferes with daily life or social life, it is called “dementia.” Compared to people of the same age, cognitive function, especially memory, is reduced, but the ability to perform daily life remains, so it is not yet severe enough to be called dementia. It is called “mild cognitive impairment.” In addition to Alzheimer's dementia, dementia can be caused by a variety of causes, including cerebrovascular disease, Lewy body dementia, frontotemporal lobe degeneration, and normal pressure hydrocephalus.

Accordingly, the present invention generally relates to a method of providing information for predicting whether a patient is positive for amyloid PET using the following equation.

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit])

[Equation 2]

Logit = 0.158 + 0.029*(age) + 0.247*(gender) + 0.034*(education) + 1.937*(APOE4 positive) + -0.155*(MMSE score)

In the present invention, P refers to the amyloid PET positivity rate.

In the present invention, cognitive impairment patients refer to all patients suffering from cognitive decline and include patients with mild cognitive impairment and patients with Alzheimer's dementia.

In the present invention, the method of providing information to predict whether the patient with cognitive impairment is positive for amyloid PET cannot confirm the cognitive function status because comprehensive cognitive function tests such as SNSB and CERAD-K tests cannot be performed, but the MMSE score is based on age, It can be used when there is a decline in academic ability or when tests such as MRI have been performed but mild cognitive impairment and Alzheimer's dementia are unclear.

In the present invention, when the P value is 0.556 or more, the patient may be determined to be positive for amyloid PET.

From another perspective, the present invention relates to a method of providing information for predicting whether amyloid PET is positive in a patient with mild cognitive impairment using the following equation.

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit])

[Equation 3]

Logit = -3.668 + 0.037*(age) + -0.016*(gender) + 0.004*(education) + 1.915*(APOE4 positive)

In the present invention, P refers to the amyloid PET positivity rate.

In the present invention, the information provision method for predicting whether the patient with mild cognitive impairment is positive for amyloid PET can be used when the patient is confirmed to be a patient with mild disability through MMSE, CDR, MRI, comprehensive neuropsychological test, blood test, etc. there is.

In the present invention, when the P value is 0.504 or more, the patient may be determined to be positive for amyloid PET.

From another perspective, the present invention relates to a method of providing information for predicting whether a patient with Alzheimer's dementia is positive for amyloid PET using the following equation.

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit])

[Equation 4]

Logit = -0.012 + 0.039*(age) + 1.043*(gender) + 0.075*(education) + 2.631*(APOE4 positive) + -0.198*(MMSE score) + -1.615*(history of hypertension)

In the present invention, P refers to the amyloid PET positivity rate.

In the present invention, the information provision method for predicting whether the Alzheimer's dementia patient is positive for amyloid PET can be used when the Alzheimer's dementia patient is confirmed to be a patient through MMSE, CDR, MRI, comprehensive neuropsychological test, blood test, etc. .

In the present invention, when the P value is 0.691 or more, the patient may be determined to be positive for amyloid PET.

From another perspective, the present invention is a method of calculating the amyloid PET positivity rate (P) value from the age, gender, education, APOE4 positivity, and MMSE score data of the subject to be analyzed using [Equation 1] and [Equation 2] below. It relates to a system for predicting amyloid PET positivity in patients with cognitive impairment, including:

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit])

[Equation 2]

Logit = 0.158 + 0.029*(age) + 0.247*(gender) + 0.034*(education) + 1.937*(APOE4 positive) + -0.155*(MMSE score)

In the present invention, [Equation 1] and [Equation 2] are derived by modeling age, gender, education, APOE4 positivity, and MMSE score data collected from a group of cognitively impaired patients through multivariate logistic regression analysis. You can do this.

In the present invention, the system may further include a prediction unit that predicts whether amyloid PET is positive from the analyzed P value.

In the present invention, when the P value is 0.556 or more, the patient may be determined to be positive for amyloid PET.

From another perspective, the present invention includes a calculation unit that calculates the amyloid PET positivity rate (P) value from the age, gender, education, and APOE4 positivity data of the subject to be analyzed using the following [Equation 1] and [Equation 3]; It relates to a system for predicting amyloid PET positivity in patients with mild cognitive impairment, including:

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit])

[Equation 3]

Logit = -3.668 + 0.037*(age) + -0.016*(gender) + 0.004*(education) + 1.915*(APOE4 positive)

In the present invention, the above [Equation 1] and [Equation 3] are characterized in that they are derived by modeling the age, gender, education, and APOE4 positivity data collected from a group of patients with mild cognitive impairment through multivariate logistic regression analysis. You can.

In the present invention, the system may further include a prediction unit that predicts whether amyloid PET is positive from the analyzed P value.

In the present invention, when the P value is 0.504 or more, the patient may be determined to be positive for amyloid PET.

From another perspective, the present invention calculates the amyloid PET positivity rate (P) value from the age, gender, education, APOE4 positivity, MMSE score, and history of hypertension data of the subject to be analyzed using [Equation 1] and [Equation 4] below. It relates to a system for predicting amyloid PET positivity in patients with Alzheimer's dementia, including a calculation unit that calculates.

[Equation 1]

P = (exp[Logit])/(1 + exp[Logit])

[Equation 4]

Logit = -0.012 + 0.039*(age) + 1.043*(gender) + 0.075*(education) + 2.631*(APOE4 positive) + -0.198*(MMSE score) + -1.615*(history of hypertension)

In the present invention, [Equation 1] and [Equation 4] are derived by modeling the age, gender, education, APOE4 positivity, MMSE score, and history of hypertension data collected from a group of Alzheimer's dementia patients through multivariate logistic regression analysis. It can be characterized as being.

In the present invention, the system may further include a prediction unit that predicts whether amyloid PET is positive from the analyzed P value.

In the present invention, if the P value is 0.691 or more, the patient may be determined to be positive for amyloid PET.

In the present invention, the final model for predicting brain amyloid PET positivity in all cognitive impairment patients, mild cognitive impairment patients, and Alzheimer's dementia patients has the ability to predict brain amyloid PET positivity compared to a simple model in which prediction is made with each variable one by one. It was confirmed that this was excellent.

In the present invention, each model can be applied depending on the cognitive level. The AUC value in each model was 0.775 in the total cognitive impairment patient group, 0.735 in the mild cognitive impairment patient group, and 0.845 in the Alzheimer's dementia patient group. This means that positive brain amyloid PET can be predicted in 77.5%, 73.5%, and 84.5% of subjects in each patient group.

In the present invention, the first model was applied to patients whose objective cognitive function level was not confirmed. The positivity rate of brain amyloid PET was 77.5% in the group of elderly female patients with low MMSE score and APOE4 gene.

In the present invention, the second model was applied to patients diagnosed with mild cognitive impairment. The positivity rate of brain amyloid PET was 73.5% in the elderly patient group with the APOE4 gene. In the present invention, patients predicted to have positive brain amyloid PET may need to undergo amyloid PET or take measures for early prevention.

In the present invention, the third model was applied to dementia patients with possible Alzheimer's dementia according to the NIAAA diagnostic guidelines. The brain amyloid PET positivity rate was 84.5% in the group of elderly patients with low MMSE scores, APOE4 gene, and no history of hypertension. In the present invention, the positive brain amyloid PET prediction result can be helpful in developing further diagnosis or treatment plans.

In the present invention, APOE4 was repeatedly confirmed to be associated with cerebral amyloid PET positivity in three models for all cognitive impairment patients, mild cognitive impairment patients, and Alzheimer's dementia patients. Inheritance of the ε4 allele of the APOE gene is the strongest genetic risk factor for Alzheimer's dementia. The interaction between APOE and cerebral amyloid β (A β) plays an important role in the development of Alzheimer's dementia. The present invention confirmed that among all variables, the APOE4 gene has the strongest correlation with positive brain amyloid accumulation. Therefore, patients with the APOE4 gene may accumulate cerebral Aβ and therefore require early amyloid PET or therapeutic intervention regardless of the level of cognitive decline.

In the present invention, the MMSE score is included in the final model for all cognitive impairment patients and Alzheimer's dementia patients. The MMSE test is one of the simplest and most widely used screening tests for patients complaining of cognitive decline. The present invention confirmed that a lower MMSE score was a predictor for brain amyloid PET positivity in all cognitive impairment patients and Alzheimer's dementia patients.

In the present invention, a history of hypertension is negatively correlated with brain amyloid PET positivity in patients with Alzheimer's dementia. Dementia patients with high blood pressure have a lower brain amyloid PET positivity rate than patients without high blood pressure. Because chronic hypertension can reduce cognitive reserve (brain reverse) through global or partial brain atrophy, patients with hypertension are more likely to have cognitive impairment due to relatively low amyloid accumulation.

In the present invention, APOE4, MMSE score, and history of hypertension were each significantly associated with brain amyloid PET positivity. In the Alzheimer's dementia patient group, the final model including all three variables showed superior screening ability compared to the simple model with only one variable.

Hereinafter, the present invention will be described in more detail through examples.

These examples are only for illustrating the present invention, and it will be obvious to those skilled in the art that the scope of the present invention is not to be construed as limited by these examples.

[Example]

Experimental method

(Example 1-1) Patient information of experimental subjects

A total of 410 people with various diagnoses such as mild cognitive impairment or Alzheimer's dementia were enrolled as experimental subjects in the cognitive impairment cohort at Seoul St. Mary's Hospital, Catholic Medical Center. We included participants who underwent both amyloid PET and apolipoprotein ε (APOE) genotyping between January 2016 and January 2019.

(Example 1-2) Screening and evaluation of brain amyloid PET positive predictor variables

For all subjects, a detailed history taking, a comprehensive physical examination including weight, height, and blood pressure (BP), blood tests, and brain imaging tests such as magnetic resonance imaging (MRI) or computed tomography (CT) was carried out. Regarding medical history, past medical history including age, gender, years of education, family history of dementia, diabetes mellitus (DM), and hypertension (HTN) were collected. Body mass index (BMI) is weight (kg) divided by the square of height (m). To exclude other causes of cognitive impairment, thyroid function tests (thyroid-stimulating hormone), lipid profile (total cholesterol), and evaluation of vitamin B12 were performed using blood from all subjects. APOE genotype was determined by PCR using blood samples in EDTA tubes. Patients with at least one ε4 allele were classified as APOE4 positive. A general cognitive evaluation was performed on all experimental patients using the mini mental state examination (MMSE) and the sum of the box score for Clinical Dementia Rating (CDR-SB). In addition, neuropsychological batteries such as the Seoul Neuropsychological Screening Battery (SNSB) or the Korean version of the Consortium to Establish a Registry for Alzheimer's Disease (CERAD-K) are used to assess multiple cognitive functions. It was used in . Clinical diagnoses include subjective memory complaint (SMC), mild cognitive impairment, Alzheimer's dementia, frontotemporal dementia (FTD), vascular dementia (VaD), and dementia with Lewy body (DLB). ) was performed according to the clinical diagnostic criteria for. The experimental subjects included MCI and AD patients.

(Example 1-3) Brain amyloid PET positive measurement

Brain amyloid PET/CT scans were performed with Discovery PET/CT 710 (GE Medical Systems, Milwaukee, WI, USA) or Biograph TruePoint 40 PET/CT (Siemens Medical Solutions USA, Knoxville, TN, USA). At this time, 18F-florbetaben and 18F-flutemetamol were used as 18F-labeled Aβ targeting tracers. The test subject received an intravenous injection of 8.0 mCi of 18F-florbetaben or 5.0 mCi of 18F-flutemetamol and rested in the waiting room. Image acquisition began after approximately 90 minutes, and the scan period was 20 minutes.

An experienced nuclear medicine physician visually assessed the patients as amyloid positive or amyloid negative using 18F-florbetaben PET and flutemetamol PET scans. 18F-Florbetaben PET images were interpreted by visually comparing the activity of cortical gray matter with the activity of adjacent cortical white matter. The brain regions of the lateral temporal cortex, frontal cortex, posterior cingulate cortex, precuneus cortex, and parietal cortex are the regions of interest, and the radioactive tracer is If it binds to the area of interest in the brain, it is read as an amyloid positive scan. On the other hand, negative scans have no tracer absorption in the target area. In the 18F-flutemetamol PET scan, the areas of interest included the frontal lobe, posterior cingulate cortex, precuneus, lateral temporal lobe, parietal lobe, and striatum. . In the case of a negative scan, more radioactive tracer binds to the white matter than the gray matter, confirming a clear gray matter contrast, whereas in the case of a positive scan, the radioactive tracer binds to the gray matter of the region of interest, resulting in a reduction or loss of gray matter contrast in at least one cortex. confirmed in the area.

(Example 1-4) Statistical analysis

Continuous variables, including age, education, BMI, MMSE, CDR-SB, systolic and diastolic BP, TSH, total cholesterol, and vitamin B12, were expressed with mean and standard error (SE). An independent t-test was used to compare the subject group with amyloid deposition and the subject group without amyloid deposition. Categorical variables, including gender, family history of dementia, DM, HTN, APOE genotype, and clinical diagnosis, were expressed as numbers and percentages. The group of subjects with amyloid accumulation and the group of subjects without amyloid accumulation were compared using the Chi-square test. p < 0.05 means statistically significant value.

The final model for predicting brain amyloid positivity was developed through multivariate logistic regression analysis using the forward conditional method. Age, gender, and education level (years of education) were entered as fixed variables. Gender, a categorical variable, was coded as 1 for male and 2 for female. Different variables were sequentially entered into the model using the forward likelihood ratio (forward LR) method. The association between clinical variables and amyloid positivity was confirmed through odds ratios (OR) and 95% confidence intervals (CIs). The present invention developed three brain amyloid PET positive prediction models for patients with cognitive impairment, patients with mild cognitive impairment, and patients with Alzheimer's dementia. The discriminatory power of the positive prediction model for brain amyloid accumulation was evaluated using receiver operating characteristic (ROC) curve analysis and area under the curve (AUC) values. To compare the final model with each simple model, a simple model for predicting brain amyloid accumulation positivity was developed using logistic regression including age, gender, education, and one of the other variables in the final model. We also explore how the Youden index, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) increase by 0.1 units from 0.1 to 0.9, finally increasing the probability of being positive for brain amyloid accumulation. The optimal cutoff point ^* was suggested for (Table 1-3). All statistical analyzes were performed using SAS version 9.2 (SAS Institute Inc., Cary, NC, USA).

Youden index, sensitivity, specificity, PPV and NPV in cognitively impaired patients Cutoff Youden index Sensitivity Specificity PPV NPV 0.1 1.005 One 0.005 0.543 One 0.2 1.094 0.982 0.112 0.566 0.840 0.3 1.273 0.932 0.340 0.625 0.810 0.4 1.333 0.806 0.527 0.668 0.697 0.5 1.385 0.694 0.691 0.726 0.657 0.556 ^* 1.452 0.676 0.777 0.781 0.670 0.6 1.423 0.631 0.793 0.782 0.645 0.7 1.394 0.500 0.894 0.847 0.602 0.8 1.267 0.320 0.947 0.877 0.541 0.9 1.125 0.131 0.995 0.967 0.492

Youden index, sensitivity, specificity, PPV and NPV in patients with mild cognitive impairment Cutoff Youden index Sensitivity Specificity PPV NPV 0.1 One One 0 0.541 0.2 1.006 0.996 0.011 0.543 0.667 0.3 1.031 0.919 0.112 0.550 0.538 0.4 1.227 0.833 0.394 0.619 0.667 0.5 1.397 0.599 0.798 0.778 0.628 0.504 ^* 1.402 0.599 0.803 0.782 0.629 0.6 1.366 0.536 0.830 0.788 0.602 0.7 1.360 0.509 0.851 0.801 0.595 0.8 1.240 0.293 0.947 0.867 0.531

Youden index, sensitivity, specificity, PPV and NPV in patients with Alzheimer's dementia Cutoff Youden index Sensitivity Specificity PPV NPV 0.1 1.036 One 0.036 0.759 One 0.2 1.095 0.988 0.107 0.771 0.750 0.3 1.274 0.988 0.286 0.808 0.889 0.4 1.334 0.976 0.357 0.822 0.833 0.5 1.405 0.941 0.464 0.842 0.722 0.6 1.513 0.871 0.643 0.881 0.621 0.691 ^* 1.574 0.788 0.786 0.918 0.550 0.7 1.550 0.765 0.786 0.915 0.524 0.8 1.456 0.671 0.786 0.905 0.440 0.9 1.447 0.482 0.964 0.976 0.380

(Example 1-5) Ethics approval

The Catholic Medical Center Institutional Ethics Review Committee approved this experimental protocol (KC20RESASI0639).

Assessment of predictors of brain amyloid PET positivity result

Table 4 shows the characteristics of 410 experimental patients according to positive brain amyloid accumulation. Among the 410 experimental patients, the proportion of amyloid PET-negative patients was 45.9% (n=188) and the proportion of amyloid PET-positive patients was 54.1% (n=222). Of the total experimental subjects, 297 were diagnosed with mild cognitive impairment and 113 with Alzheimer's dementia. Amyloid PET-positive patients were older than amyloid PET-negative patients. 70.3% of amyloid PET positive patients were female, and 61.2% of amyloid PET negative patients were female. BMI was lower in amyloid PET positive patients than in amyloid PET negative patients. 53.6% of amyloid PET positive patients were APOE4 positive, and 17.0% of amyloid PET negative patients were APOE4 positive. Amyloid PET-positive patients had lower cognitive function than amyloid PET-negative patients, which was confirmed by amyloid PET-positive patients having lower MMSE scores and higher CDR-SB scores. Serum total cholesterol was higher in amyloid PET positive patients than in amyloid PET negative patients.

Characteristics of experimental patients according to positive brain amyloid accumulation Amyloid PET-CT results p-value Negative Positive Number 188 222 Age, years 74.48±8.54 76.01±8.64 0.073 Female, N (%) 115 (61.17) 156 (70.27) 0.052 BMI, kg/ ^m2 23.73±3.31 23.16±3.14 0.079 Education, years 11.70±4.81 10.91±5.10 0.105 APOE4-positive, N (%) 32 (17.02) 119 (53.60) <0.001 Family history of dementia, N (%) 46 (25.41) 64 (30.77) 0.242 Hypertension, N (%) 92 (48.94) 102 (45.95) 0.546 Diabetes mellitus, N (%) 59 (31.38) 53 (23.87) 0.089 Diagnosis, N (%) <0.001 MCI 160 (85.11) 137 (61.71) AD dementia 28 (14.89) 85 (38.29) MMSE score 24.69±3.63 21.87±5.54 <0.001 CDR-SB score 2.5±2.32 4.09±3.64 <0.001 Systolic blood pressure, mm Hg 126.35±17.34 128.77±16.75 0.191 Diastolic blood pressure, mm Hg 72.54±10.56 73.40±11.25 0.469 TSH, mIU/L 2.75±7.60 2.64±2.89 0.859 Total cholesterol, mg/dL 178.88±41.05 186.04±40.54 0.111 Vitamin B12, pg/mL 823.89±768.07 751.12±552.97 0.344

In Table 4, PET is positron emission tomography, BMI is body mass index, APOE is apolipoprotein ε, MCI is mild cognitive impairment, and Ad is Alzheimer's disease. (Alzheimer's disease), MMSE is mini-mental state examination, CDR-SB, sum of the box score for Clinical Dementia Rating, and TSH is thyroid stimulating hormone. represents.

Development of brain amyloid PET positivity prediction model and prediction index

The following variables were used to develop a prediction model for brain amyloid PET positivity: age, gender, education, body mass index (BMI), family history of dementia, history of hypertension or diabetes mellitus (DM), and brief psychiatry. Mini-Mental State Examination (MMSE), Clinical Dementia Rating Scale (CDR) and Sum of Box (CDR-SB), APOE4 genotype, total cholesterol, Thyroid Stimulating Hormone , TSH), vitamin B12, blood pressure and clinical diagnosis. These variables were included because they are well known as risk factors for dementia.

In the present invention, three final brain amyloid PET positive prediction models were developed for all cognitive impairment patients, mild cognitive impairment patients, and Alzheimer's dementia patients through multivariate logistic regression analysis of the above data. Using the beta coefficient value of the final model, a logit equation, which is a prediction index of brain amyloid PET positivity, was derived for each model, and this logit value was used to calculate the brain amyloid PET positivity rate (P) in each model. The formula was derived.

(i) P _case = (exp [Logit _case ])/(1+exp[Logit _case ]),

(ii) Logit _all = β0 + β1 (age) + β2 (gender) + β3 (education) + β4 (APOE4 positive) + β5 (MMSE score),

However, β0 = 0.158, β1 = 0.029, β2 = 0.247, β3 = 0.034, β4 = 1.937, β5 = -0.155.

(iii) Logit _MCI = β0 + β1 (age) + β2 (gender) + β3 (education) + β4 (APOE4 positive);

However, β0=-3.668, β1=0.037, β2=-0.016, β3=0.004, β4=1.915.

(iv) Logit _AD = β0 + β1 (age) + β2 (gender) + β3 (education) + β4 (APOE4 positive) + β5 (MMSE score) + β6 (history of hypertension)

However, β0=-0.012, β1=0.039, β2=1.043, β3=0.075, β4=2.631, β5=-0.198, β6=-1.615.

The first model is a model that predicts brain amyloid PET positivity for all cognitive impairment patients with subjective cognitive decline in the memory clinic. Age, gender, education, BMI, APOE4, and MMSE score were selected as the final model for all cognitive impairment patients (Table 5). The AUC value of the ROC curve for the developed model was 0.775 (95% confidence interval [CI] = 0.731 to 0.820) (Figure 1A).

Model to predict brain amyloid PET positivity in patients with cognitive impairment Variable Beta Wald OR (95% CI) p-value Final model Age 0.029 4.374 1.030 (1.002, 1.058) 0.037 Sex 0.247 0.953 1.280 (0.780, 2.100) 0.329 Education 0.034 1.653 1.034 (0.983, 1.088) 0.199 APOE4 1.937 55.879 6.939 (4.176, 11.532) <0.0001 MMSE -0.155 27.068 0.857 (0.808, 0.908) <0.0001 Intercept 0.158 R-square ^* 0.302 APOE4-only model Age 0.038 8.213 1.038 (1.012, 1.066) 0.004 Sex 0.203 0.694 1.225 (0.760, 1.974) 0.405 Education -0.010 0.176 0.990 (0.946, 1.037) 0.675 APOE4 1.826 54.903 6.208 (3.830, 10.062) <0.0001 Intercept -3.517 R-square ^* 0.215 MMSE-only model Age 0.009 0.564 1.009 (0.985, 1.034) 0.453 Sex 0.354 2.344 1.425 (0.905, 2.243) 0.126 Education 0.021 0.776 1.021 (0.975, 1.070) 0.378 MMSE -0.138 25.617 0.871 (0.826, 0.919) <0.0001 Intercept 1.883 2.089 0.148 R-square ^* 0.122

In Table 5, R-square ^* represents Nagelkerke R-square, OR represents odds ratios, CI represents confidence interval, APOE represents apolipoprotein ε, and MMSE represents Mini-Mental State Examination.

The second model is a model that predicts brain amyloid PET positivity for patients with mild cognitive impairment. Age, gender, education, BMI, APOE4, and MMSE score were selected as the final model for patients with mild cognitive impairment (Table 6). The AUC value of the ROC curve for the developed model was 0.735 (95% CI=0.676~0.793) (Figure 1B).

Model to predict brain amyloid PET positivity in patients with mild cognitive impairment Final model Beta Wald OR (95% CI) p-value Age 0.037 5.679 1.038 (1.007, 1.071) 0.017 Sex -0.016 0.003 0.985 (0.565, 1.714) 0.956 Education 0.004 0.018 1.004 (0.949, 1.062) 0.893 APOE4 1.915 46.405 6.784 (3.911, 11.768) <0.0001 Intercept -3.668 R-square ^* 0.227

In Table 6, R-square ^* represents Nagelkerke R-square, OR represents odds ratios, CI represents confidence interval, and APOE represents apolipoprotein ε.

The third model is a model that predicts brain amyloid PET positivity for patients with Alzheimer's dementia. Age, gender, education, BMI, APOE4, MMSE score, and history of hypertension were selected as the final model for Alzheimer's dementia patients (Table 7). The AUC value of the ROC curve for the developed model was 0.845 (95% CI=0.765~0.925) (Figure 1C).

Model to predict brain amyloid PET positivity in Alzheimer's dementia patients Variable Beta Wald OR (95% CI) p-value Final model Age 0.039 1.016 1.040 (0.964, 1.122) 0.314 Sex 1.043 2.680 2.838 (0.814, 9.896) 0.102 Education 0.075 1.724 1.077 (0.964, 1.204) 0.189 APOE4 2.631 9.339 13.890 (2.569, 75.089) 0.002 MMSE -0.198 8.795 0.820 (0.720, 0.935) 0.003 Hypertension -1.615 6.443 0.199 (0.057, 0.692) 0.011 Intercept -0.012 R-square* 0.429 APOE4-only model Age -0.002 0.003 0.998 (0.939, 1.061) 0.958 Sex 1.213 4.271 3.365 (1.065, 10.634) 0.039 Education 0.050 0.943 1.051 (0.950, 1.164) 0.332 APOE4 2.453 9.888 11.619 (2.519, 53.589) 0.002 Intercept -1.831 R-square ^* 0.259 MMSE-only model Age -0.008 0.065 0.992 (0.932, 1.056) 0.799 Sex 1.061 3.851 2.890 (1.001, 8.342) 0.050 Education 0.048 0.865 1.049 (0.948, 1.162) 0.352 MMSE -0.125 6.066 0.882 (0.798, 0.975) 0.014 Intercept 1.955 R-square ^* 0.160 Hypertension-only model Age 0.006 0.036 1.006 (0.942, 1.075) 0.849 Sex 0.987 3.105 2.684 (0.895, 8.049) 0.078 Education 0.017 0.115 1.017 (0.924, 1.119) 0.735 Hypertension -1.579 7.940 0.206 (0.069, 0.618) 0.005 Intercept -0.143 R-square* 0.185

In Table 7, R-square ^* is Nagelkerke R-square, OR is odds ratios, CI is confidence interval, APOE is apolipoprotein ε, and MMSE is mini-mental state test. examination).

To compare the final model and each of the simple models, age, gender, education, and BMI were included as fixed variables and one of the other variables in the final model (APOE4, MMSE in the cognitive impairment patient model) / Simple models predicting brain amyloid PET positivity including APOE4, MMSE, and history of hypertension in the Alzheimer's dementia patient model were developed (Tables 5 and 7). The AUC values of the cognitive impairment patient model were 0.730 for the APOE4-only model and 0.663 for the MMSE-only model. The AUC values of the Alzheimer's dementia patient model were 0.747 for the APOE4-only model, 0.710 for the MMSE-only model, and 0.724 for the hypertension-only model. Through statistical comparison of the AUC values between the final model and each simple model, it was confirmed that the final model showed a much larger AUC value than the simple model (Table 8). As a result, it was confirmed that the final model can predict brain amyloid PET positivity more accurately than each simple model.

AUC values of the ROC curve for the final model and each simple model predicting brain amyloid PET positivity Model AUC Lower Upper All subjects Final model 0.775 0.731 0.820 APOE4-only model 0.730 0.681 0.779 MMSE-only model 0.663 0.611 0.715 MCI subjects Final model 0.735 0.676 0.793 AD subjects Final model 0.845 0.765 0.925 APOE4-only model 0.747 0.652 0.842 MMSE-only model 0.710 0.603 0.816 Hypertension-only model 0.724 0.615 0.832

In Table 8, AUC is the area under the curve, ROC is the receiver operating characteristic, APOE is apolipoprotein ε, MMSE is the mini-mental state examination, and MCI is mild cognitive impairment. Mild cognitive impairment, AD stands for Alzheimer's disease.

Claims (18)

As an information provision method to predict whether amyloid PET is positive,
(a) deriving the patient's P value based on Equation 1 and Equation 2 below; and
(b) A method of providing information for predicting whether a patient with cognitive impairment is positive for amyloid PET, including the step of predicting positive amyloid PET if the patient's value derived in step (a) is above the reference value:
[Equation 1]
P = (exp[Logit])/(1 + exp[Logit]),
[Equation 2]
Logit = 0.158 + 0.029*(age) + 0.247*(gender) + 0.034*(education) + 1.937*(APOE4 positive) + -0.155*(MMSE score);
However, P refers to the amyloid PET positivity rate.
According to paragraph 1,
When the P value is 0.556 or more, the patient is determined to be amyloid PET positive.
How to provide information to predict whether a patient with mild cognitive impairment will have a positive amyloid PET using the following equation:
[Equation 1]
P = (exp[Logit])/(1 + exp[Logit]),
[Equation 3]
Logit = -3.668 + 0.037*(age) + -0.016*(gender) + 0.004*(education) + 1.915*(APOE4 positive),
However, P refers to the amyloid PET positivity rate.
According to paragraph 3,
When the P value is 0.504 or more, the patient is determined to be amyloid PET positive.
How to provide information to predict whether a patient with Alzheimer's dementia will be positive for amyloid PET using the following equation:
[Equation 1]
P = (exp[Logit])/(1 + exp[Logit]),
[Equation 4]
Logit = -0.012 + 0.039*(age) + 1.043*(gender) + 0.075*(education) + 2.631*(APOE4 positive) + -0.198*(MMSE score) + -1.615*(history of hypertension),
However, P refers to the amyloid PET positivity rate.
According to clause 5,
When the P value is 0.691 or more, the patient is determined to be amyloid PET positive.
A calculation unit that calculates the amyloid PET positivity rate (P) value from the age, gender, education, APOE4 positivity, and MMSE score data of the subject to be analyzed using [Equation 1] and [Equation 2] below; Cognitive impairment, including; Patient's amyloid PET positivity prediction system:
[Equation 1]
P = (exp[Logit])/(1 + exp[Logit]),
[Equation 2]
Logit = 0.158 + 0.029*(age) + 0.247*(gender) + 0.034*(education) + 1.937*(APOE4 positive) + -0.155*(MMSE score).
In clause 7,
The system is characterized in that the above [Equation 1] and [Equation 2] are derived by modeling the age, gender, education level, APOE4 positivity, and MMSE score data collected from a group of cognitively impaired patients through multivariate logistic regression analysis.
In clause 7,
The system further includes a prediction unit that predicts whether amyloid PET is positive from the analyzed P value.
In clause 7,
If the P value is 0.556 or higher, the system determines that the patient is amyloid PET positive.
delete delete delete delete A calculation unit that calculates the amyloid PET positivity rate (P) value from the age, gender, education, APOE4 positivity, MMSE score, and history of hypertension data of the subject to be analyzed using [Equation 1] and [Equation 4] below; , Amyloid PET positivity prediction system in Alzheimer's dementia patients:
[Equation 1]
P = (exp[Logit])/(1 + exp[Logit]),
[Equation 4]
Logit = -0.012 + 0.039*(age) + 1.043*(gender) + 0.075*(education) + 2.631*(APOE4 positive) + -0.198*(MMSE score) + -1.615*(history of hypertension).
According to clause 15,
The above [Equation 1] and [Equation 4] are characterized by being derived by modeling the age, gender, education level, APOE4 positivity, MMSE score, and history of hypertension data collected from a group of Alzheimer's dementia patients through multivariate logistic regression analysis. , system.
According to clause 15,
The system further includes a prediction unit that predicts whether amyloid PET is positive from the analyzed P value.
According to clause 15,
The system is characterized in that, if the P value is greater than or equal to 0.691, the patient is determined to be positive for amyloid PET.