Immunomagnetic Reduction Detects Plasma Aβ_1–42 Levels as a Potential Dominant Indicator Predicting Cognitive Decline

This article is published with digital features to facilitate understanding of the article. To view digital features for this article go to https://​doi.​org/​10.​6084/​m9.​figshare.​12982070.

Pathological evidence of Alzheimer’s disease (AD) includes the presence of amyloid plaques and tau protein tangles in the brain [1‐4], which are typically detected with positron emission tomography (PET) [5‐8]. Although amyloid or tau PET has been approved for clinical use, its high cost and low availability seriously limit its utility in clinical practice. Instead, the detection of amyloid β (Aβ) and tau protein in cerebrospinal fluid (CSF) has been suggested as a method to probe for pathological evidence of AD [9‐12]. However, it has not been easy to implement CSF biomarker detection more widely in clinical practice because of the need to perform a lumbar puncture. The detection of Aβ and tau protein in other body fluids such as blood, urine, or saliva has been proposed as an alternative method. Compared to those in the CSF, the levels of Aβ and tau protein in such body fluids are extremely low, on the order of picograms per milliliter [13‐15]. Thus, ultrasensitive assays are needed to precisely quantify the levels of Aβ and tau proteins in blood, urine, or saliva. At present, several ultrasensitive assaying technologies have been developed [16‐20], but there is no consistency in the measured concentrations of plasma Aβ and tau protein among the different assaying technologies. For example, the levels of plasma amyloid β 1–42 (Aβ_1–42) were found to be reduced in AD patients as compared to normal controls using single-molecule array (SIMOA), whereas they were found to be higher using immunomagnetic reduction assay (IMR). The discrepancies in measured results among assaying technologies could be attributable to several factors, including antibodies, assay methodology, and sample preparation. Hence, it would be better to concentrate on the results obtained by assay with one of these technologies. In this review, the assay results obtained with superconducting quantum interference device-based immunomagnetic reduction (SQUID IMR) are the subject of interest [21, 22].

Most studies utilizing SQUID IMR have focused on exploring the associations between Aβ or tau protein in plasma and in CSF [15], the results of amyloid or tau PET [23‐25] or magnetic resonance imaging [24, 26], or clinical diagnosis [27‐29]. These studies have yielded promising results regarding the feasibility of detecting AD in patients by assaying blood Aβ and tau protein. We will not summarize such works in this review, because previous review articles have already investigated the working mechanism and clinical impact of SQUID IMR [30, 31]. Instead, according to published papers [32‐35], we will aim to discuss the possible dominance of plasma Aβ and tau protein in the progression to dementia in at-risk patients, such as those with Down syndrome, stroke, or amnestic mild cognitive impairment (aMCI). Such research is so novel that few papers have been published. In this review, five studies are referenced to show the roles of plasma Aβ and tau in the progression to dementia in these at-risk patients. This is a review article that is based on reported studies and does not contain any new studies with human participants or animals performed.

Down syndrome (DS) is a common chromosomal abnormality [36]. Due to the presence of a triplicated chromosome 21, DS patients usually suffer from amyloid deposition in the brain, which increases the risk of developing dementia. The onset age of dementia in DS patients is earlier than that in normal controls by 30 years. Thus, a high prevalence of dementia has been observed in DS patients younger than 35 years of age [37]. However, in clinical practice it is very difficult to diagnose dementia in a DS patient, regardless of the results of magnetic resonance imaging, amyloid PET, neuropsychological tests, or CSF biomarker detection. Therefore, a delayed diagnosis of dementia for DS patients often occurs in the clinic. On the other hand, drawing blood is relatively easy for DS patients. This has motivated studies exploring the feasibility of discriminating dementia in DS patients via plasma biomarkers.

Fang et al. investigated the difference in levels of plasma biomarkers between DS individuals without and with dementia [32]. They enrolled 35 DS patients without dementia and 16 DS patients with dementia, using the Adaptive Behavior Dementia Questionnaire (ABDQ) score to discriminate between the two groups. SQUID IMR was used to assay plasma Aβ_1–42, amyloid β 1–40 (Aβ_1–40), and tau protein (tau) in the enrolled subjects. All three biomarkers were reported to be moderately associated with the ABDQ score, as illustrated in Table 1. Notably, the plasma Aβ_1–42 level showed the most significant correlation with the ABDQ score in DS patients. The results of the receiver operating characteristic (ROC) curve analysis for discriminating dementia in DS using the plasma concentrations of Aβ_1–42, Aβ_1–40, and tau are listed in Table 2. It was observed that the plasma Aβ_1–42 level was the predominant indicator for the early detection of neurodegeneration in DS. These results reveal that the plasma Aβ_1–42 level might be the most powerful parameter for predicting the progression of dementia in DS.

Stroke causes serious brain damage or death in afflicted patients. It can also lead to cerebral ischemia, which triggers the deposition of Aβ, the hyperphosphorylation of tau, and neuroinflammation in the brain [38‐40]. Hence, post-stroke patients are at risk for the progression of dementia. It has been reported that post-stroke dementia occurs in 7% of patients 1 year after a stroke [41], which increases to 52% at 6 months if mild post-stroke cognitive impairment is included in the diagnosis [42]. Thus, management for both stroke and dementia is needed in patients post-stroke. However, such patients usually experience impairment in activities of daily living, language, etc. [43, 44]. It is difficult to evaluate neurodegeneration in post-stroke patients using neuroimaging or neuropsychological tests. Blood tests, on the other hand, would be a more accessible method.

Tang et al. utilized an IMR assay for plasma Aβ_1–40, Aβ_1–42 and tau in post-stroke patients with (n = 34, age = 76.9 ± 6.2 years) or without (n = 27, age = 71.0 ± 6.6 years) dementia [33]. Neuropsychological tests, including the clinical dementia ranking (CDR) scale, the Mini-Mental State Examination (MMSE), and the Montreal Cognitive Assessment (MoCA), were performed for every participant. The clinical diagnosis of vascular dementia was made in accordance with the criteria for dementia from the National Institute of Neurological Disorders and Stroke and the Association Internationale pour la Recherche et l’Enseignement en Neurosciences (NINDS-AIREN). The results of the study showed values of CDR ≥ 1, MMSE = 13.1 ± 4.2, and MoCA = 7.3 ± 4.7 for post-stroke patients with dementia, and CDR ≤ 0.5, MMSE = 28.7 ± 0.9, and MoCA = 23.6 ± 3.4 for those without dementia. Among the measured plasma biomarkers, only the Aβ_1–42 level was significantly different between post-stroke patients with (17.6 ± 2.6 pg/ml) and without dementia (15.4 ± 1.8 pg/ml, p < 0.01). Receiver operating characteristic curve analysis determined a cutoff value of 15.59 pg/ml for discriminating non-dementia from dementia in post-stroke patients, resulting in clinical sensitivity, specificity, and area under the curve of 0.765, 0.741, and 0.760, respectively, as shown in Fig. 1. Furthermore, plasma Aβ_1–42 showed a moderately negative correlation with the MMSE (r = −0.46, p < 0.01) and MoCA scores (r = −0.44, p < 0.01) in post-stroke patients, as shown in Table 3. The levels of plasma Aβ_1–40 and tau were not associated with the MMSE or MoCA scores. Hence, the plasma Aβ_1–42 level is a promising index for determining neurodegeneration in post-stroke patients.

Chi et al. investigated the prediction of cognitive impairment in post-stroke patients 1 year after stroke using levels of plasma biomarkers 3 months after stroke [34]. Forty-two post-stroke patients with and 13 post-stroke patients without cognitive impairment after 1 year were enrolled. The plasma Aβ_1–42, Aβ_1–40, and tau levels 3 months after stroke were assayed. The level of plasma Aβ_1–42 at 3 months was the most significant biomarker in predicting post-stroke cognitive impairment 1 year after stroke. Therefore, plasma Aβ_1–42 is a promising index for both predicting and determining cognitive impairment in post-stroke patients.

Amnestic mild cognitive impairment (aMCI) is defined as a transition from normal aging to Alzheimer’s disease (AD). Approximately 30% of aMCI patients progress to AD in 1–1.5 years, and more than 50% progress to AD in 3 years [45, 46]. One of the symptoms during this progression is cognitive decline. In the clinic, the objective is to predict cognitive decline in aMCI patients. Tsai et al. investigated the predictive power for cognitive decline of the baseline levels of plasma biomarkers at a 1.2-year follow-up in aMCI patients [34]. Twenty-four aMCI patients were enrolled. The value of Pearson’s correlation coefficient r between the baseline levels of the plasma biomarkers and the annual changes in the MMSE score are listed in Table 4. Among the three different plasma biomarkers, Aβ_1–42 showed a significant and moderate negative correlation with the annual change in the MMSE score.

Similar results were reported independently by Chen et al. and are listed in Table 5. Twenty-two aMCI patients were enrolled [35], and changes in the MMSE scores were measured at the 1–1.5-year follow-up. The value of Pearson’s r between the baseline levels of the plasma biomarkers and the annual change in the MMSE score was analyzed. Compared to the levels of Aβ_1–40 and tau, the level of plasma Aβ_1–42 at baseline showed the strongest association with the annual change in the MMSE score. The two independent studies by Tsai et al. and Chen et al. illustrate that plasma Aβ_1–42 at baseline is a useful parameter for predicting cognitive decline in aMCI.

In populations at risk for developing dementia, such as patients with Down syndrome, stroke, and amnestic mild cognitive impairment, it is important to be able to predict the progression to dementia or cognitive decline. In clinical practice, a convenient examination that can evaluate progression to dementia or cognitive decline is urgently needed. According to this review, plasma Aβ_1–42 may play a more key role in this prediction than Aβ_1–40 or tau. This implies that abnormal levels of Aβ_1–42 might be a critical trigger of the onset of dementia or cognitive decline. Therefore, there may be hope for stopping the development of dementia or delaying cognitive decline by eliminating the abnormal changes in the level of plasma Aβ_1–42. Nevertheless, there are limitations in this work that should be mentioned. Notably, because of the novelty of such research, only five studies were available for this review. Additional independent studies should be performed to further explore the promising results described in this review.