Alzheimer's disease (AD) will become a prominent public health issue in the future given its cognitively debilitating nature. As the advent of global ageing society is expected, AD may bring tremendous socioeconomical costs if current diagnosis methods stay put. In this article, we performed a systematic review of a recent (less than 10 years) ultrasensitive technology, the immunomagnetic reduction (IMR), which shows promising potential of early diagnosis of AD.
We searched the Pubmed and Embase databases for studies that included keywords “early-stage Alzheimer's disease” and “immunomagnetic signal reduction.”
After full-text review, a total of 7 studies were included for final analysis. Most included studies have reported on Aβ40, Aβ42, t-tau, and levels of these biomarkers in the plasma of early AD patients comparing those in the healthy population. The ranges of the mean Aβ40 levels are as follows: 59.2 to 60.9 for control groups and 36.9 to 39.5 pg/mL for AD. Aβ42 and t-tau concentrations are both markedly lower than Aβ40, Aβ42 at 15.5 to 16.1 for control groups and 17.9 to 19 pg/mL for AD; t-tau levels were 13.5 to 14.3 for control groups and 39.4 to 46.7 pg/mL for AD. There is a significant increasing level of plasma Aβ42 by IMR assays in early AD patients across nearly all the included studies. There is a possible relationship between plasma levels of IMR AD biomarkers and (1) degree of hippocampal atrophy using magnetic resonance imaging, and (2) amount of brain amyloid accumulation using positron emission tomography.
IMR assay is an ultrasensitivity technique that is useful for detection of early AD, which can provide benefits on understanding the disease progression of AD and encourage early medical invention for AD patients.
Alzheimer's disease (AD) is a chronic and lethal neurological disorder caused by progressive severe cognitive and physical function impairment. It is a disease continuum that may ultimately lead to symptoms of Alzheimer's dementia, including short-term memory loss, personality and behavioral change, limited verbal communication, and movement dysfunction, reflecting the degree of neuron damage in different parts of brain.[1,2] Under updated National Institute on Aging and the Alzheimer Association Guideline in 2011, Alzheimer's dementia is redefined as the stage describing the dementia stage of AD, which also accounts for 60% to 80% of dementia cases. The number of dementia cases currently increases at a rate of 10 million per year, and there is an estimate of 50 million people with dementia in the world.[4,5]
Long-term care and hospice service for dementia patients aged older than 65 have been a huge burden for medical systems across the globe. How to confirm and intervene as early as possible in a new-onset AD patient, therefore, has become a key issue. Currently, with the aid of AD biomarkers, molecular changes in brain and circulation system have the potential to early diagnose AD conveniently before clinical symptoms arises, endowing great benefit in personal health and substantial reduction in medical costs.
Grossly, any apparent change of brain structure in patients with AD cannot be diagnostic. However, significant atrophy of the hippocampus and dilatation of the adjacent temporal horn of the lateral ventricle, may be a reliable indicator suggesting further microscopic evidence from a patient with AD. At the microscopic level, extracellular deposition of beta amyloid-Aβ and intracellular accumulation of tau protein are the 2 principal mechanisms leading to AD.[7–9]
Aβ is part of amyloid precursor protein, which will be degraded into Aβ amyloid residues. Aβ monomers and oligomers are further broken down by other enzymes. Abnormal clearance of Aβ from abnormal cleavage of amyloid precursor protein will lead to its accumulation. An example of its larger accumulation fragments is Aβ42.[10–12]
Aβ is toxic to neurons, especially for the hippocampus and entorhinal cortex, which are the most affected areas in AD patients. These accumulated Aβ42 fragments form the insoluble structures that account for the formation of senile plaques and oxyradicals, resulting in loss of long-term potentiation, synapses damaged, and neurons killed.[13,14]
Another key player in AD is tau protein, which provides microtubule stability. Hyperphosphorylation of tau protein results in neurofibrillary tangles. By distorting the spacing of microtubules, neurofibrillary tangles impair the axonal transport and its nutrition, as well as impairment of mitochondrial oxidative metabolism. Some studies[16–18] demonstrate strong evidence of the relationship between Aβ accumulation and tau protein aggregation, proposing a Aβ plaques–mediated inflammatory process, which ultimately leads to tau protein exacerbation, finishing the final step of the pathogenesis of AD.
However, autopsy studies reveal that people aged over 65 and without clinical dementia, may contain a few senile plaques and neurofibrillary tangles in the hippocampus and entorhinal cortex of their brain, suggesting these also belong to natural aging process.[19–22] On the other hand, the abundancy and wide-spreading of the senile plaque and neurofibrillary tangles can tell an AD patient from a healthy elderly person, and quantifying the senile plaque and neurofibrillary tangles may be able to demonstrate the severity of the dementia.
A definitive diagnosis of AD can only be made by pathological examination of brain tissue. Biomarkers of AD, such as reduced Aβ42, elevation of tau protein in cerebrospinal fluid (CSF), and Aβ plaques deposited in neuronal tissue identified by positron emission tomography (PET), are believed to be useful biomarkers for early detection of AD.[24–27] Fluorodeoxyglucose PET scans of the same individuals also demonstrate reduced glucose metabolism in the posterior temporoparietal regions bilaterally in AD. However, the costs and limited access to CSF and PET analysis restrict the use in these cases. There is thus a great need for another accessible, reliable, and promising technique to detect biomarkers of AD at an early stage.
Recently, there are increasing easier and simpler tests by blood-based biomarkers detection developed, including immunomagnetic reduction (IMR), single molecule assay, immunoinfrared sensor technology, multimer detection system, and immunoprecipitation/mass spectrometry.[29–35] All these techniques are similar to enzyme-linked immunosorbent assay (ELISA), which uses specific antibodies to target antigens, but add new technologies to enhance sensitivity and accuracy. We will discuss the newest technology, IMR, and its results, comparing with other detections in early diagnosis of AD.
Immunomagnetic Reduction (IMR) Assay
IMR is a novel technique to quantize target molecules by measuring the decrease in a mixture of frequency magnetic propensity of magnetic reagent because of binding of the targeting proteins to magnetic nanoparticles. In IMR, the reagent is a solution which has distributed magnetic nanoparticles equally, and magnetic nanoparticles are thus covered with hydrophilic surface-active agents and bioprobes (e.g., antibodies). With external multiple AC magnetic fields, magnetic nanoparticles vibrate with the fields by means of magnetic interaction. Therefore, the reagent demonstrates a magnetic property under external multiple AC magnetic fields, which is called mixed-frequency ac magnetic susceptibility χac. Via the bioprobes on the outermost shell, magnetic nanoparticles are associated and magnetically labeled biomolecules, such as antigens, are therefore identified. Magnetic nanoparticles become bigger from the association as a result, as shown in the scheme (Fig. 1). The inertia of these larger magnetic nanoparticles in response to the external magnetic fields in comparison to that of originally individual magnetic nanoparticles leads to a reduced χac of the reagent. The method is therefore called “immunomagnetic reduction.” Theoretically, when more amounts of to-be-detected biomolecules are mixed with a reagent, more magnetic nanoparticles with larger sizes are produced. A larger reduction in χac could be expected for reagents.
IMR exhibits several unique merits. First, owing to its wash-free processes and bioprobes used, the unbound to-be-identified biomolecules and magnetic nanoparticles are not essentially to be washed out. Second, magnetic nanoparticles have a property of high ratio of surface to volume. There is abundant area for binding between antibodies and targeted biomolecules, leading to a high sensitivity. Third, the cross reactions are inhibited by IMR. As magnetic nanoparticles vibrate with the external AC magnetic fields, the χac of the magnetic reagent is detected. At high oscillating frequencies, the centrifugal force is augmented. When the centrifugal force is higher than the binding force between the bioprobes and nontarget biomolecules, the weak binding will break down, which suppresses the nonspecific binding between the bioprobes and nontarget biomolecules.[36–45]
We searched the Pubmed and Embase databases for studies published since inception until March 1, 2020 that included keywords “early-stage Alzheimer's disease” and “immunomagnetic signal reduction.” We did not set limitations on the languages used in full-text articles.
Two reviewers (PUT and ICW) independently screened the titles and abstracts to identify potentially relevant studies. Full texts of relevant publications were screened independently by two reviewers (PUT and ICW). A third reviewer (CJH) confirmed the inclusion or exclusion of the studies.
Initial electronic database searches yielded 93 results. After title and abstract screening, 66 studies were eligible for full-text review. After full-text review, a total of seven studies were included for the final analysis (Fig. 2). Reasons for exclusion include (1) no mentioning of either Aβ42 and t-tau protein, (2) AD not to be the main theme discussed in the paper, (3) no emphasis put on early-stage AD, (4) p-tau instead of t-tau to be discussed in the paper, and (5) review, conference abstract, or editorial.
Characteristics of Included Studies
Most included studies have reported on Aβ40, Aβ42, t-tau, and levels of these biomarkers in the plasma of early AD patients comparing those in the healthy population. Significant discrepancy was found in most studies, and thus a suitable cutoff level of these biomarkers may be determined for facilitating diagnosis of early AD patients. Some papers evaluated the relationship between levels of IMR plasma markers and image findings on PET and MRI imaging. Other than plasma markers, APOE genotyping was proposed by Szu-Ying Lin et al in 2019 to be added in an algorithm to further increase the accuracy for detecting early AD. Compiled in Table 1 are summarized findings from the included studies using IMR assays of plasma AD pathological markers. The table is organized by publication date, from the earliest to the latest. In addition to the major findings, the number of patients in the cognitively healthy control group versus AD group, possible imaging modalities, and the types of IMR plasma biomarkers are also provided.
Alzheimer's Disease Marker Levels Measured by IMR Assays
Included in Table 2 are the means and ranges of AD marker levels measured by IMR assays. The ranges of the mean Aβ40 levels were 59.2 to 60.9 for control groups and 36.9 to39.5 pg/mL for AD. Aβ42 and t-tau concentrations are both significantly lower than Aβ40, with Aβ42 fluctuating between 15.5 to 16.1 for control groups and 17.9 to 19 pg/mL for AD; t-tau levels standing between 13.5 to 14.3 for control groups and 39.4 to 46.7 pg/mL for AD. There is a significant increasing level of plasma Aβ42 by IMR assays in early AD patients across nearly all the included studies. Plasma Aβ42 levels in AD patients was on average 10% to 20% greater than the levels in plasma from control groups. The same phenomenon happened more prominently with t-tau plasma levels. When compared with control groups, t-tau plasma levels in early AD patients were 2.8- to 3.5-fold of the levels in control groups. IMR Aβ40 plasma levels, however, when discussed, invites more controversies. Its IMR plasma levels in AD patients were lower, but not higher as in the cases of Aβ42 and t-tau, by 40%, than those in control groups.
Kai-Yuen Tzen et al reported a satisfactory result in diagnosing early AD with the usage of the ratio of Aβ42/Aβ40 instead of the absolute values of each biomarker. When compared with values from the control group, there were significantly higher Aβ42/Aβ40 values in AD patients.
Relationship Between IMR AD Markers and Imaging
Several imaging modalities have been used for assessing possible relationship between plasma levels of IMR AD biomarkers and (1) degree of hippocampal atrophy using MRI, and (2) the amount of brain amyloid accumulation using PET.
A regression analysis by Ming-Jang Chiu et al demonstrates that total hippocampal volume as measured by MRI accounted for 39.4% of the variance in the plasma tau level, while the gray matter density of the superior frontal accounted for 5.4%.
Ling-Yun Fan et al used 11C-labeled Pittsburgh compound-B positron emission tomography to show brain amyloid accumulation, as well as mean standardized uptake value ratio (SUVR) of all cortical regions, increased significantly in patients with AD. A cut-off level was also suggested for differentiating between the control group and AD patients (Pib−: tau 37.54 and Aβ42 21.92 pg/mL; Pib+: tau 25.57 and Aβ42 16.81 pg/mL). Sensitivity and specificity are 100% for both tau and Aβ42 plasma levels of Pib− group, while standing at 93.3%/100% for tau and 86.7%/81.8% for Aβ42 for Pib+ group. They also tried to depict a linear relation between plasma biomarkers, amyloid deposition, and cortical atrophy, as indicated by mean cortical thickness. The results showed that mean cortical thickness was influenced the most by mean 11C-labeled Pittsburgh compound-B positron emission tomography SUVR (which could account for ∼20%), followed by plasma Aβ40 level, which explained an additional 6.5%.
Szu-Ying Lin et al proposed a cut-off level of plasma Aβ42 at 15.58 pg/mL for discriminating PET+ from PET− patients, with sensitivity and specificity being 59.1% and 60.0%, respectively.
This is the first systematic review to use all literature available regarding application of IMR-mediated plasma biomarker measurement, namely Aβ40, Aβ42 and t-tau, in early AD diagnosis.
Ming-Jang Chiu et al in 2013 is the first study to propose the application of IMR-assayed combined plasma biomarkers in early AD diagnosis. In this paper, not only healthy controls could be distinguished from AD patients by IMR, but further they suggested the difference in IMR AD marker plasma levels could allow identification of different stages of mild cognitive impairment (MCI), namely the prodromal and the dementia phases of AD. Another paper by his team in 2014 showed plasma AD biomarker levels are linked to image findings. The authors employed MRI imaging to show that plasma tau levels are negatively correlated to the performance of logical memory, visual reproduction, and verbal fluency, and also are negatively correlated to the volume of hippocampus, amygdala, total gray matter, and the densities of gray matter in different regions.
Kai-Yuen Tzen et al in 2014 found that the level of plasma AD biomarkers could indirectly predict early or preclinical pathological conditions of AD. All recruited members underwent 11C-labeled Pittsburgh compound-B PET scans and detected the plasma concentration of several AD biomarkers. The team then reached a conclusion that the severity of Aβ deposition and the volume of the hippocampus could be predicted only by the ratio of Aβ42/Aβ40 level and the level of plasma tau protein, respectively.
Teunissen et al in 2018 reported that the negative correlation of plasma Aβ42 and CSF Aβ42 could be caused by the different assaying methodologies. The team assayed IMR for plasma Aβ42 and ELISA for CSF Aβ42, resulting in a negative correlation. By reviewing previous studies, they found an opposite result compared with their findings, indicating a positive correlation between the levels of plasma Aβ42 and CSF Aβ42 that both were measured with ELISA. The team came up with a hypothesis that while plasma Aβ42 often binds to carrier proteins in blood, ELISA, a method using 2 antibodies could induce a higher potential of encountering obstacles with binding to target plasma biomarkers. On the contrary, IMR is a method of single-antibody immunoassay, which potentially can be a better way of detecting plasma biomarkers.
In 2019, Lin et al demonstrated the idea of combining APOE genotypes with plasma Aβ1–42 levels as a predicting factor of positive amyloid PET findings in early AD patients. This study also proposed the cut-off values of plasma Aβ1–42 quantified by IMR technique in differing PET+ from PET− patients within either group of APOE ε4 gene carriers/noncarriers. This surprising finding may further participate in future pharmaceutical researches and the development of early-stage AD patients.
Chen et al in 2019 researched whether plasma t-tau and Aβ42 levels can be used to calculate the degree of cognitive decline in participants with amnestic MCI. T-tau and plasma Aβ42 were measured in 22 candidates with amnestic MCI using IMR. This research found out that higher levels of Aβ42 and t-tau offered were negatively correlated with follow through period cognitive decline and episodic verbal memory performance at baseline. This result implies that higher levels of plasma t-tau and Aβ42 in amnestic MCI are associated with subsequent cognitive decline.
The World Health Organization predicts that by the mid-21st century, AD will increase 300% comparing to the status quo and will dominate the cause of death in most countries. According to the World Health Organization report in 2017, the number of dementia cases currently increases at a rate of 10 million per year, and there is an estimate of 50 million people with dementia in the world. These figures are guessed to escalate to more than 150 million in the mid-21st century because of prolonged life expectancies in developing countries. The global social expense is projected to reach over US $2 trillion by 2030, a 3-fold increase from US $818 billion in 2015. The current social cost related to AD has already reached 1.1% of the global gross domestic product, and especially higher, on average 1.4% of the local gross domestic product, in developed countries.
Early diagnosis of AD is crucial to successful treatment as currently available treatment could only provide small benefits for late stage AD patients. There are several preclinical or clinical trials therapeutics that are promising in neuroprotection before widespread brain damage and dementia happen. The ultrasensitivity technique is useful for detection of early AD, providing benefits on understanding the disease progression of AD and encourage early medical invention.
There are several limitations in this study that could be addressed in future researches. First, the studies reviewed are based on cross-sectional study method. Further analysis under well-designed longitudinal study is needed to determine whether IMR-assayed plasma markers could predict different stages of AD. Second, there is a large variation between the stages of AD that were diagnosed by clinical presentations and those that were determined by pathological findings. For example, education and intelligence quotient can improve or worsen the clinical staging with the same pathological severity. However, the plasma and tau-protein levels can still be a useful tool to distinguish healthy controls from the patients satisfactorily with both high sensitivity and specificity. Third, previous studies yielded conflicting results regarding the levels of Aβ42 proteins in plasma. Several studies assayed using the IMR method yielded an increase of plasma Aβ42 in AD patient, while other studies conducting research using the ELISA method reported opposite or no difference. The possible reason might be related to the different designs of the platform employed. The ELISA method is based on the sandwich assay, which requires 2 antibodies to detect the Aβ42 protein. Plasma Aβ42 is often bound to carrier proteins in the blood, such as albumin or lipoproteins, leading to a possible obstacle for 2 antibodies to bind normally. However, the IMR method involves the use of single-antibody immunoassay, which can be a more feasible way of avoiding interference and binding target proteins in the blood. Last, the papers reviewed measured the total tau of the plasma but not the phosphorylated form of tau (p-tau). Production of total tau protein also happens in such situations as acute stroke, Creutzfeldt disease, and traumatic brain injury. T-tau is not as specific as p-tau in detecting neurodegenerative diseases. On the other hand, p-tau has less concentration in the plasma, making it challenging to be used as a biomarker. Therefore, the detection of p-tau could be further studied in the future.
In summary, we reviewed that utilization of IMR technique may be considered a highly sensitive and specific test for early detecting AD. However, because researches conducted by now are heterogeneous and had no consensus, future studies relating to using IMR technique combining plasma marker cut-off point and its prediction value should be well and thoroughly conducted. Study design of longitudinal researches in a larger population is encouraged, so as to have a more specific and accurate association, and to make IMR technique more applicable in future clinical situations.
Source of Support: None. Conflict of Interest: None.
These authors contributed equally to this work.