Blood amyloid β oligomerization as a biomarker of Alzheimer’s disease: a blinded validation study

INTRODUCTION: Oligomeric amyloid ß (Aß) is one of the major contributors to the pathomechanism of AD; Aß oligomerization in plasma can be measured using a Multimer Detection System-Oligomeric Aß (MDS-OAß) after incubation with spiked synthetic Aß. METHODS: We evaluated the clinical sensitivity and specificity of the MDS-OAß values by inBlood TM OAß test using heparin-treated plasma samples from 52 AD patients in comparison with 52 community-based subjects with normal cognition (NC). The inclusion criterion was proposed by the NINCDS-ADRDA and additionally required for the least 6 months of follow-up from the initial clinical diagnosis in the course of AD. RESULTS: The MDS-OAβ values were 1.43 ± 0.30 ng/ml in AD and 0.45 ± 0.19 ( p <0.001) in NC, respectively. Using a cut-off value of 0.78 ng/ml, the results revealed that 100% sensitivity 92.31% specificity. DISCUSSION: MDS-OAß to measure plasma Aβ oligomerization is a valuable blood-based biomarker for clinical diagnosis of AD, with high sensitivity and specificity.


Introduction
Studies investigating the development of a biomarker for the early diagnosis of Alzheimer's disease (AD) have increased throughout medical communities worldwide.
Many studies have suggested that amyloid β (Aβ) oligomers could present as potent toxic species compared to insoluble fibrils (1)(2)(3)(4). Aβ oligomers, playing a key-role in AD pathogenesis, have become a promising candidate for AD diagnosis. Efforts to measure Aβ oligomers in humoral fluids are currently underway (5-7). However, cost-effective and non-invasive diagnostic methods to detect Aβ oligomers are not currently available in clinical practice. Furthermore, a clinically applicable method to measure Aβ oligomers in the blood has not yet been reported. Due to the invasiveness of lumbar puncture, the accessibility of patients for cerebrospinal fluid (CSF) analysis is low, limiting its general usage. Therefore, blood analysis would be a beneficial complement to this shortcoming.
However, there are several limitations of using plasma Aβ as blood-based biomarker of AD. Firstly, Aβ is highly diluted in blood (8,9). Furthermore, Aβ can bind to and interfere with other protein and peptides (10, 11), and undergo degradation (12) and selfaggregation (13) in the blood.
An atypical approach, called the Multimer Detection System-Oligomeric Aβ (MDS-OAβ), was introduced to measure the oligomerization dynamics in plasma samples after spiking synthetic Aβ (14). This essentially utilizes the MDS technique, which can selectively detect oligomers in a given sample (15,16). One study found that the level of Aβ oligomers increased after spiking Aβ and incubation, in plasma samples of AD patients but not in healthy normal subjects (14). The elevated levels of Aβ oligomers closely correlated with conventional AD biomarkers, such as CSF Aβ42 and Pittsburgh compound B (PIB) positron emission tomography (PET) standard uptake ratio, CSF phosphorylated tau, and CSF total tau (17). In this study, we aimed to validate the accuracy of MDS-OAβ for measuring Aβ oligomerization dynamics in heparin-treated plasma samples from patients with AD and healthy controls. The objective of this study was to assess the sensitivity and specificity of the MDS-OAβ test in differentiating plasma from AD patient and subjects with normal cognition (NC). The positive predictive value (PPV) and negative predictive value (NPV) were also evaluated. In addition, the MDS-OAβ levels in AD patients was compared to the Clinical Dementia Rating (CDR) scores, a numeric scale used to quantify the severity of dementia symptoms.

Methods
The study was supported by a contract research organization (CRO). This clinical study was approved by the Korea Ministry of Food and Drug Safety (MFDS) and performed in accordance with its approved protocol (No.753). Approval by the institutional review board  Table 1.

Results
None of the 104 samples were excluded during the test procedure. The average MDS-OAβ values of the AD and NC samples were 1.43 ± 0.30 ng/ml and 0.45 ± 0.19 ng/ml, respectively, and there was a significant difference between the two groups (p < 0.001) (Fig. 2). The ROC analysis indicated an optimal cut-off value (0.78 ng/ml), which allowed the best differential discrimination between patients with AD and NC subjects. In order to closely verify the optimal cut-off value, we induced the reference interval by biding by the  Abbreviations: NC, community-based normal control; ADD, Alzheimer's disease dementia * Ratio difference as to whether the sensitivity is greater than 70.9%; Z-test Ratio difference as to whether the specificity is greater than 70.8%; Z-test The MDS-OAβ levels were significantly higher in patients with a CDR score of 0.5 (1.46 ± 0.33 ng/ml, n = 25), 1 (1.53 ± 0.21, n = 17), and 2 (1.26 ± 0.21, n = 9), compared with CDR 0 (0.45 ± 0.19, n = 25) (p < 0.001) (Fig. 3). Interestingly, the average of MDS-OAβ levels decreased as the CDR score in AD patients increased; however, there was no significance (p > 0.05).

Discussion
Many researchers have chosen Aβ oligomers as a biomarker for the diagnosis of AD as it satisfies the criteria of an ideal biomarker, which was proposed by the Ronald and Nancy Reagan Research Institute-NIA in 1998 (25). Several studies have substantiated the positive correlation between the levels of Aβ oligomers in plasma and the likelihood of AD (26), and found the sensitivity and specificity to be less than 85% (6). The MDS-OAβ measures oligomerization dynamics of Aβ in the blood without the need for specialized equipment, unlike conventional techniques which directly measure Aβ molecules using larger machines due to the low concentration of target molecules in the blood (14). In this study, the oligomerization tendencies of Aβ in AD and normal plasma were measured using the MDS-OAβ, and we demonstrated that the sensitivity and specificity were 100% and 92.3%, respectively; therefore MDS-OAβ has very high sensitivity and specificity in distinguishing AD from NC. We used a stringent patient recruitment criterion for this study. For example, AD patients were followed-up for at least 6 months by experienced neurologists to rule out the possibility of cognitive impairment caused by any other disease, and community-based NC subjects without cognitive decline, were enrolled.
Although the data was not shown in this study, while assessing 29 cases of the AD, all patients showed high MDS-OAβ levels, and a PIB or Florbetaben PET was conducted. The standardized uptake value radio (SUVR) of 25 cases were positive, and 3 cases had a positive visual rating but negative SUVR. One case had a negative amyloid PET but had a typical CSF profile of AD. This patient may have a soluble form Aβ, which failed to produce Aβ plaques in the brain; therefore, only showing changes in the CSF biomarker (27)(28)(29)(30).
While the MDS-OAβ measures dynamics of Aβ oligomerization (14), amyloid PET only reveals the fibrillar form of Aβ in the brain (31), which may have caused the discrepancy.
The other possible reason is that changes in CSF biomarkers may have occurred before the amyloid PET change. Aβ oligomerization tendency in plasma may reflect as early as changes in the CSF biomarker, and signify AD regardless of the solubility of Aβ.
Many studies have argued that biomarkers, such as brain volume and CSF Aβ42, p-tau, and t-tau which indicates the downstream effects of AD, show an increase in severity as the disease progress and formed a graph of the sigmoid curve (32,33). However, the average of MDS-OAβ level in this study was the highest at CDR score 0.5 and the lower as AD progress (Fig. 3). MDS-OAβ measures the oligomerization dynamics of Aβ, which corresponds to the derivative of the sigmoid function of Aβ accumulation. It is possible that this biomarker changes during the early phase of AD, as shown with other biomarkers associated with processes upstream of the AD pathomechanism, and decreases in expression as the disease progresses (34)(35)(36). Another possible explanation could be that the concentrations of neuronal injury/death biomarkers decrease after symptom onset, which suggests slowing of the acute neurodegenerative processes with symptomatic disease progression (37).
One limitation of the present study was the age difference between the AD and NC groups.
However, we found that there was no correlation between MDS-OAβ levels and age in the 52 NC subjects (range 51-77, mean 60.5 ± 7.4), and speculated that age difference was not a significant variable in influencing MDS-OAβ levels. Second, the MMSE score of AD increased in patients with CDR 2 (Table 1), but this is thought to be due to the patients being of a younger age than those with CDR 1.

Conclusions
Plasma samples of AD and NC subjects were differentiated using MDS-OAβ, which measured the Aß oligomerization tendency of plasma. Furthermore, MDS-OAβ was found to have high sensitivity and specificity. Based on the current findings, measuring the Aβ oligomerization tendency in plasma could be a simple and reliable blood-based biomarker for the diagnosis of AD.  OAß mean with an increase of CDR score was not significant (p>0.05); however, the MDS-OAβ levels were significantly higher in patients with a CDR score of 0.5, 1, and 2 versus those with a CDR score 0 was significant (p<0.001).