Analysis of Cerebrospinal Fluid and [ 11 C]PIB PET Biomarkers for Alzheimer’s Disease with Updated Protocols

Abstract

Background: Recently, a Korean research group suggested a consensus protocol, based on the Alzheimer’s Disease Neuroimaging Initiative study protocol but with modifications for minimizing the confounding factors, for the evaluation of cerebrospinal fluid (CSF) biomarkers.

Objective: Here, we analyzed fluid and imaging biomarkers of Alzheimer’s disease (AD) in Korean population. We used the updated protocol to propose a more accurate CSF biomarker value for the diagnosis of AD.

Methods: Twenty-seven patients with AD and 30 cognitively normal controls (NC) were enrolled. CSF was collected from 55 subjects (patients with AD = 26, NC = 29) following the Korea consensus protocol. CSF biomarkers were measured using the INNO-BIA AlzBio3 immunoassay, and Pittsburgh compound B (PIB) positron emission tomography (PET) scans were also performed.

Results: The cutoff values of CSF amyloid beta 1-42 (Aβ₄₂), total tau (t-Tau), and phosphorylated tau (p-Tau) proteins were 357.1 pg/ml, 83.35 pg/ml, and 38.00 pg/ml, respectively. The cutoff values of CSF t-Tau/Aβ₄₂ and p-Tau/Aβ₄₂ ratio- were 0.210 (sensitivity 100%, specificity 86.21%) and 0.1350 (sensitivity 88.46%, specificity of 92.86%). The concordance rate with PIB-PET was higher using the CSF t-Tau/Aβ₄₂ ratio (κ= 0.849, CI 0.71–0.99) than CSF Aβ₄₂ alone (κ= 0.703, CI 0.51–0.89).

Conclusions: Here, we improved controversial factors associated with the previous CSF study protocol and suggested a new cutoff value for the diagnosis of AD. Our results showed good diagnostic performance for differentiation of AD. Thus, we expect our findings could be a cornerstone in the establishment and clinical application of biomarkers for AD diagnosis.

Keywords

Alzheimer’s disease biomarker cerebrospinal fluid diagnosis Pittsburgh compound B protocol

INTRODUCTION

The accuracy of an Alzheimer’s disease (AD) diagnosis obtained by using purely clinical information is generally low to predict a neuropathological AD diagnosis [1]. Reports from drug trial studies show that 10% to 30% of clinically diagnosed AD cases have no evidence of amyloid pathology on amyloid positron emission tomography (PET) scans [2]. Therefore, recently established AD diagnostic criteria provide the probability of AD pathology according to biomarker state [3 –5]. The major AD biomarkers can be divided into two classes based on the biology they measure. The first class is known as biomarkers of brain amyloid-β (Aβ) protein deposition, which are decreases in Aβ₄₂ protein in the cerebrospinal fluid (CSF) and positive amyloid PET imaging. The other category of biomarker relates to neuronal injury, which include measures of elevated total tau (t-Tau) and phosphorylated tau (p-Tau) protein in the CSF, decreased temporoparietal hypometabolism on 18-fluorodeoxyglucose (FDG) PET, and medial temporal atrophy on brain magnetic resonance imaging (MRI) [3]. A combination of these biomarkers could well discriminate patients with AD from cognitively normal controls.

According to the hypothetical model of dynamic biomarkers for AD pathology, the evidence of amyloid pathology or neurodegenerative markers precedes clinical symptoms by many years [6]. Verification of biomarker state is beneficial to detect a patient with prodromal AD. Therefore, interpretation of the results of each biomarker as “positive” or “negative” is crucial to AD diagnosis. However, due to differences in protocols or analytical methods, there is considerable variability in measured biomarker values across laboratories [7, 8]. Specifically, the source of variations stems from preanalytical, analytical, postanalytical, and assay manufacturing factors [9]. In 2015, a Korean research group suggested a protocol for measuring AD biomarkers in the CSF in order to standardize preanalytical factors. The proposed approach was based on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) study protocol, but with modifications for minimizing the confounding influence of various factors [10].

Here, we report findings from a study we designed, called “Alzheimer’s Disease All Markers (ADAM),” in which we collected data of well-known AD biomarkers using brain MRI, ¹¹Carbon- Pittsburgh compound B PET (PIB PET), and CSF assessment. We used the updated protocol proposed by the abovementioned Korean research group in an attempt to suggest a more accurate CSF cutoff value to discriminate between patients with AD and cognitively normal individuals.

METHODS

Subjects

Twenty-seven patients with AD and 30 cognitively normal controls (NC) were enrolled from the Neurocognitive Behavior Center of Seoul National University Bundang Hospital and Department of Neurology of Chung-Ang University Hospital, Republic of Korea from April 2012 to November 2014. All subjects underwent a comprehensive clinical assessment, including medical history, physical and neurologic examination, and routine blood tests including vitamin B12, folate, syphilis serology, and thyroid function test. The patients also underwent neuropsychological testing via the Mini–Mental State Examination (MMSE), and neuropsychiatric testing, which evaluated attention, memory, language, visuospatial, and frontal function. On the basis of clinical assessments and neuropsychological tests, we selected clinically well-characterized patients with AD for participation in this study. The inclusion criteria of the AD group were as follows: 1) age between 50 and 90 years, 2) probable AD by National Institute on Aging-Alzheimer’s Association (NIA-AA) criteria [3], 3) ≥6 years of education, 4) Clinical Dementia Rating (CDR): From 0.5 to 2 and CDR sum of box (SOB) ≥2.5, 5) Modified Hachinski Ischemia score≤4, and 6) a caregiver who knew the patient well. Cognitively normal controls met the following inclusion criteria: 1) age between 50 and 90 years, and satisfaction of normal aging criteria [11], 2) scores on the MMSE of at least 1.0 standard deviation below the mean for their age- and education-matched norm, 3) ≥6 years of education, 4) short form Geriatric Depression Score≤7. We excluded individuals if they had 1) significant or unstable medical problems, 2) psychiatric problems, 3) a cardiac pacemaker, and/or 4) a history of substance abuse or dependence within the past 10 years.

This study was approved by the Seoul National University Bundang Hospital Institutional Review Board. Written informed consent was obtained from all patients or the caregivers of the patients who participated in this study.

CSF study

CSF was collected from the 55 subjects (NC = 29, AD = 26) via the consensus protocol put forth by the Korea Research Group regarding the preanalytical processing of CSF AD biomarkers [10]. In brief, a lumbar puncture was conducted at the space between levels L3/L4 or L4/L5 in the morning (8–12 AM). An experienced interventional radiologist carried out the lumbar puncture using fluoroscopy with either cutting-edge Quincke needles (20 gauge) or “atraumatic” Whitacre needles (20 or 22 gauge). A total of 10–15 ml of CSF was collected in a 15-ml centrifuge tube by free flow of CSF. Centrifugation at 2000 × g for 10 min at room temperature was conducted within 4 hours after the lumbar puncture. Five hundred μL polypropylene tubes with screw caps were used to aliquot 400 μL of CSF. These aliquots were immediately frozen at –80°C. CSF biomarkers of AD (Aβ₄₂, p-Tau, and t-Tau) were analyzed using INNO-BIA AlzBio3 immunoassay kit-based reagents [12]. The INNO-BIA AlzBio3 is a fluorimetric immunoassay in which different parameters are selectively captured on beads of a specific region by a primary monoclonal antibody (AT270 for p-tau, AT120 for t-Tau, 4D7A3 for Aβ₄₂), coupled covalently. The INNO-BIA AlzBio3 is tested using xMap^® technology.

PET study

Static PET acquisition for 20 min was performed 40 min after intravenous bolus injection of ¹¹Carbon- Pittsburgh compound B (658.6 ± 122.1 MBq) in each subject. Each PIB PET image was coregistered to a T1 MRI of each subject and spatially normalized to the T1 brain magnetic resonance (MR) template. Then, MR-based segmentation of cerebral gray and white matter was performed, and the mean standardized uptake value (SUV) was calculated in each brain region using modified automated anatomical labeling. A mean cortical PIB standardized uptake value ratio (PIB SUVR) was introduced by calculating the mean uptake over voxels in the prefrontal, anterior cingulate, posterior cingulate, precuneus, lateral temporal, and parietal regions of interest (ROIs) for each subject and dividing this by the mean uptake over voxels in the cerebellar gray matter.

Statistical analysis

Baseline characteristics were compared between groups using an unpaired t-test and chi-square test in SPSS 20.0 software. The mean ± standard deviation values of CSF biomarkers were also analyzed by an unpaired t-test in SPSS 20.0 software. To identify the cutoff values of fluid biomarkers, we used receiver operating characteristics (ROC) analysis in GraphPad Prism, v5. The cutoff values of PIB PET SUVR based on estimated percentiles were also calculated. For example, the cutoff value corresponding to 90% sensitivity was the estimated 10th percentile of the AD group. The cutoff values corresponding to five levels of sensitivity (80%, 85%, 90%, 95%, and 99%) were similarly calculated. Comparisons between CSF Aβ₄₂ and PIB PET studies were analyzed using percent agreement as well as the κ statistic with 95% confidence intervals (CI) using SPSS 20.0 software.

RESULTS

Baseline characteristics

Table 1 shows the baseline characteristics of patients with AD (n = 27) and normal controls (n = 30). There were no differences in the age, sex, and years of education between the two groups. Apolipoprotein E ɛ4 carrier (homozygote or heterozygote) was greater in the AD group (46.2%, n = 12) than in the NC group (16.7%, n = 5). There were significant differences in MMSE scores between the two groups. The completion rate of PET and CSF assessments for each diagnostic group is also given in Table 1.

Analysis of CSF biomarkers

Mean ± standard deviation values for CSF biomarkers in AD and NC groups are provided in Table 2. There are significant differences between the two groups in all CSF biomarkers (p < 0.05). Figure 1 shows the distribution of each biomarker concentration and ratio. Similar to other studies about CSF biomarkers of AD [13 –22], we confirmed trends of decreases in Aβ₄₂ and increases in t-Tau, p-Tau, the t-Tau/Aβ₄₂, and p-Tau/Aβ₄₂ ratio in patients with AD when compared to normal controls. Although there were definite differences in mean levels of CSF biomarkers, the distribution considerably overlapped between the two groups. However, the t-Tau/Aβ₄₂ and p-Tau/Aβ₄₂ ratio presented relatively less overlapping areas than the single analyte.

We also calculated the cutoff values for discriminating between patients with AD and NCs using the ROC analysis (Table 3). For a single parameter, CSF Aβ₄₂ had the greatest ROC AUC, 0.9191. The cutoff value of CSF Aβ₄₂ was 357.1 pg/ml with a sensitivity of 82.76% and specificity of 88.46%. The greatest ROC AUC value was obtained for the CSF t-Tau/Aβ₄₂ ratio (0.9542), with sensitivity and specificity values of 100% and 86.21%, respectively. For the CSF p-Tau/Aβ₄₂ ratio, the AUC value is 0.9478, comparable to for CSF t-Tau/Aβ₄₂ ratio, with sensitivity of 88.46% and specificity of 92.86%. The cutoff values of CSF t-Tau/Aβ₄₂ and p-Tau/Aβ₄₂ ratio were 0.210 and 0.135.

Analysis of amyloid PET

To determine the cutoff values of PIB PET biomarkers, we used estimated percentile scores for different sensitivity levels, which had been reported in a previous study [23]. Figure 2 presents the distribution pattern of mean cortical PIB SUVR in patients with AD and NCs. The points corresponding to different diagnostic sensitivities (80%, 85%, 90%, 95%, and 99%) for AD are included in Fig. 2. The cutoff value with a 90% sensitivity was 1.259. The proportion of NCs that was classified as being in the abnormal range at each cutoff point is also provided in Fig. 2.

Concordance of CSF and PIB PET biomarkers

Figure 3 shows the concordance between the PIB PET and CSF biomarkers for AD and NC groups. A portion of the patients with AD showed inconsistencies between fluid and imaging amyloid biomarkers. Of the 27 patients with AD, three patients showed high CSF Aβ₄₂ with high PIB SUVR while two patients had low CSF Aβ₄₂ and low PIB SUVR. In the NC group, six patients showed evidence of amyloid deposit, low CSF Aβ_42, or high PIB SUVR. Among six patients, three patients presented amyloidosis in both CSF and PIB PET studies. Thus, there was 85.2% agreement (κ= 0.703, CI 0.51–0.89) between PIB PET and CSF Aβ₄₂ (Fig. 3A). However, six of eight subjects with disagreements between PIB and CSF Aβ₄₂ showed values that were very close to the cutoff points for differentiation. The concordance rate of PIB PET was higher using the CSF t-Tau/Aβ₄₂ ratio (κ= 0.849, CI 0.71–0.99) than CSF Aβ₄₂ alone or the CSF p-Tau/Aβ₄₂ ratio (Fig. 3B and 3C).

DISCUSSION

The main purpose of this study was to establish cutoff values of biomarkers for the diagnosis of AD in clinical practice. We collected AD biomarker data from CSF and PIB PET studies in well-characterized patients with AD and age-matched cognitive NCs. In order to reduce variability of the CSF biomarker and control for preanalytical factors, we followed an updated CSF study protocol that had been modified from the ADNI protocol.

In this study, the CSF biomarker profiles of AD were characterized by decreased Aβ₄₂ and increased p-Tau and t-Tau. These findings were consistent with previous studies [12 , 25]. The cutoff value of each analyte was <357.1 (pg/ml, sensitivity of 82.76% and specificity of 88.46%) for CSF Aβ₄₂, >0.210 for the CSF t-Tau/Aβ₄₂ ratio (sensitivity 100% and specificity 86.21%), and >0.135 for the CSF p-Tau/Aβ₄₂ ratio (sensitivity of 88.46% and specificity of 92.86%). The greatest ROC AUC was shown on the CSF t-Tau/Aβ₄₂ ratio and then on the p-Tau/Aβ₄₂ ratio. For the differentiation of AD from NCs, the CSF t-Tau/Aβ₄₂ or p-Tau/Aβ₄₂ ratios were superior to tau protein or Aβ₄₂ alone. Expert opinions have also suggested that combined Aβ₄₂, t-Tau, and p-Tau provide reliable biomarkers for AD [26].

Numerous studies have reported on CSF biomarker levels, however, measured values and diagnostic accuracy significantly vary (Table 4). In addition, inter- and intra-center variabilities are high [27]. These variations result from preanalytical, analytical factors, or assay manufacturing processes [9, 27]. Possible confounding preanalytical factors include patient selection, CSF collection procedure, and sample storage. Analytical factors arise from assay kit handling techniques, laboratory equipment, and analytical variations such as pipetting error or assay acceptance criteria. Finally, factors associated with assay manufacturing variability can be attributed to optimization methods, as well as within-assay and lot-to-lot inconsistencies in reagents. To overcome these limitations, the Alzheimer’s Association has been conducting an international quality control (QC) program since 2009 with an aim of identifying factors that influence intra- and inter-center variations in AD CSF biomarkers. In 2011, Mattsson et al. [27] reported the total coefficient variability (CV) between laboratories participating in the QC program were 13 to 36%, and attributed variability to analytical procedures and differences in analytical kits.

The two most commonly used commercial AD CSF assay platforms are the INNOTEST enzyme-linked immunosorbent assay (ELISA) and INNO-BIA AlzBio3 xMAP technology (Innogenetics NV, Ghent, Belgium). A previous study that compared these two analytical platforms reported that the assays exhibit different (approximately two- to six-fold) absolute values, but that they both perform equally well in identifying individuals with underlying amyloid plaque pathology [28]. However, the microsphere-based xMAP technology provides several advantages over INNOTEST ELISA, which include simultaneous quantification of several analytes in a small volume of sample and reductions in total assay time as well as in the number of handling steps. Furthermore, xMAP might also provide higher reproducibility and lower assay variability than INNOTEST ELISAs [12]. With these reasons in mind, we employed INNO-BIA AlzBio3 xMAP technology for CSF analysis in the present study.

In Korea, two studies on AD CSF biomarkers have been reported; however, preanalytical factors were not controlled for and the analytical kits employed were different from what was used in the current study [19, 25]. Using data from the ADNI study, Shaw et al. [21] suggested the following CSF biomarker cutoff values: 192 pg/ml for Aβ₄₂, >0.10 for the p-Tau/Aβ₄₂ ratio and >0.39 for the t-Tau/Aβ₄₂ ratio. Although the analytical kit used in our study was identical to that employed by Shaw et al., we observed differences in concentration values. As mentioned, data in the Shaw et al. study had been gathered using a protocol proposed by the ADNI study. In contrast, our study used a modified version of that protocol in order to minimize the confounding influence of various factors [10]. Thus, the discrepancy in concentration values was presumably caused by variability in preanalytical factors between the two protocols (see Table 5 for the full list of differences).

With regard to how specific protocol discrepancies can affect biomarker concentrations, it has been shown that the number of CSF transfers can influence the value of Aβ₄₂. More specifically, the ADNI protocol permits two CSF transfers; however, Toombs et al. [29] reported that each transfer could decrease Aβ₄₂ levels by 28.6%. Therefore, the current study only allowed for one CSF transfer. In addition, we employed centrifugation to account for the instability of proteomic analytes caused by the time delay before freezing [30]. Further, we only one freeze-thaw cycle was permitted in the current study. Although there are still controversies about the influence of freezing and thawing on biomarker concentrations, some studies have reported a reduction in Aβ₄₂ levels even after the first freeze-thaw cycle [31]. Finally, it has been shown that CSF collection volume can influence biomarker concentrations via rostrocaudal gradients, and previous protocols [32] have recommended a collection volume of at least 12 ml. Therefore, we collected the minimum required volume of CSF. Thus, our close attention to the controversial issues surrounding CSF AD biomarker collection could be a strength of this study.

In order to evaluate diagnostic performance using CSF biomarkers, we assessed the concordance between fluid and imaging amyloid biomarkers. Although there is currently no established consensus regarding “PIB-positive” categorization [23 , 34], the cutoff point of PIB PET was calculated using estimated percentiles scores of the ADNI AD distribution data. The cutoff values of PIB PET for the diagnosis of AD were mean cortical PIB SUVR 1.5 and 1.41, yielding a sensitivity of 90% and 95%, respectively [23]. To settle the cutoff point for PIB-positive classification with our data, we used the same method. The cutoff point with 90% and 95% sensitivity was PIB SUVR 1.259 and 1.140, respectively. Because the way we calculated PIB SUVR varied from that of previous studies [16 , 35], the absolute cutoff value reported here was different from the value associated with the ADNI data. The parameters we chose in the current study (ROI and reference regions) were selected for their high diagnostic performance and good correlation with cognitive functions [36]. As it showed a good correlation rate (κ> 0.7) between CSF and PIB PET biomarkers, this methodology could be reliably applied for the differentiation of patients with AD from healthy individuals.

In the present study, some subjects exhibited a mismatch between fluid and imaging amyloid biomarkers. Low CSF Aβ₄₂ and high PIB SUVR is commonly observed in patients with AD [6], however, five of the patients with AD (17.2%) showed inconsistencies in these amyloid biomarkers. Previous studies have reported similar levels of concordance [37, 38]. One explanation for this phenomenon could be the chronological order of amyloid measurement; CSF Aβ₄₂ levels could become abnormal prior to PIB PET [39]. However, three patients did not show low Aβ₄₂ levels on CSF analysis in spite of a high amyloid load on PIB PET. The reason for this is unclear, but it could be that measured amyloid protein differs between PET and CSF analysis. Amyloid PET is related to as fibrillar Aβ [40], whereas CSF Aβ₄₂ is a marker of soluble Aβ₄₂ [41]. In the NC group, six patients showed evidence of amyloid deposit. In the very early or prodromal stage, AD biomarkers are present even in the absence of AD signs/symptoms [23]. Thus, although all these individuals displayed normal neuropsychological test and brain MRI results, follow-up examinations for these subjects will be needed.

The current study had several limitations. First, the study sample was small and the study itself was conducted in a single country. However, we were trying to select well-characterized AD patient by experts. In addition, the performance rates of individual tests exceeded 95% in both imaging and fluid biomarker studies. Moreover, the AUC indicated high sensitivity and specificity in all biomarkers. Concordance between fluid and imaging measures was also good; thus, we expect our results using updated protocol could be applied to differentiate individuals with AD from those without AD. Second, we were unable to evaluate intra- and inter-center variation. To apply the cutoff values reported here in a clinical setting, it is necessary to validate these findings via a large multi-center study using this protocol. However, as we followed the updated standard protocol to measure CSF AD biomarkers, our results may contribute to future multi-center studies.

In summary, this is the first study to assess comprehensive biomarkers of AD in Korea. Although this study was conducted in one country, the cutoff values proposed for CSF biomarkers of AD were derived using an updated protocol that had been designed to control for previously identified controversial factors. We expect the present study will be a cornerstone in the establishment and clinical application of biomarkers for the diagnosis of AD.

Footnotes

ACKNOWLEDGMENTS

This work was supported by grants from the Korea Healthcare Technology R & D Project (HI14C3331 and HD15A0040) through the Korea Health Industry Development Institute (KHIDI), Korea Ministry of Health & Welfare, Republic of Korea.

Authors’ disclosures available online ().

References

Beach

, Monsell

, Phillips

, Kukull

(2012) Accuracy of the Clinical Diagnosis of Alzheimer Disease at National Institute on Aging Alzheimer Disease Centers, 2005–2010, J Neuropathol Exp Neurol 71, 266–273.

Salloway

, Sperling

, Fox

, Blennow

, Klunk

, Raskind

, Sabbagh

, Honig

, Porsteinsson

, Ferris

, Reichert

, Ketter

, Nejadnik

, Guenzler

, Miloslavsky

, Wang

, Lu

, Lull

, Tudor

, Liu

, Grundman

, Yuen

, Black

, Brashear

(2014) Two phase 3 trials of Bapineuzumab in mild-to-moderate Alzheimer’s disease, N Engl J Med 370, 322–333.

McKhann

, Knopman

, Chertkow

, Hyman

, Jack

Jr , Kawas

, Klunk

, Koroshetz

, Manly

, Mayeux

, Mohs

, Morris

, Rossor

, Scheltens

, Carrillo

, Thies

, Weintraub

, Phelps

(2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease, Alzheimers Dement 7, 263–269.

Dubois

, Feldman

, Jacova

, Hampel

, Molinuevo

, Blennow

, DeKosky

, Gauthier

, Selkoe

, Bateman

(2014) Advancing research diagnostic criteria for Alzheimer’s disease: The IWG-2 criteria, Lancet Neurol 13, 614–629.

Dubois

, Feldman

, Jacova

, Dekosky

, Barberger-Gateau

, Cummings

, Delacourte

, Galasko

, Gauthier

, Jicha

, Meguro

, O’Brien

, Pasquier

, Robert

, Rossor

, Salloway

, Stern

, Visser

, Scheltens

(2007) Research criteria for the diagnosis of Alzheimer’s disease: Revising the NINCDS-ADRDA criteria. Lancet Neurol 6 734–746.

Jack

, Knopman

, Jagust

, Petersen

, Weiner

, Aisen

, Shaw

, Vemuri

, Wiste

, Weigand

(2013) Tracking pathophysiological processes in Alzheimer’s disease: An updated hypothetical model of dynamic biomarkers. Lancet Neurol 12, 207–216.

Ritchie

, Smailagic

, Noel-Storr

, Takwoingi

, Flicker

, Mason

, McShane

(2014) Plasma and cerebrospinal fluid amyloid beta for the diagnosis of Alzheimer’s disease dementia and other dementias in people with mild cognitive impairment (MCI). Cochrane Database Syst Rev 6, CD008782.

Zhang

, Smailagic

, Hyde

, Noel-Storr

, Takwoingi

, McShane

, Feng

(2014) (11)C-PIB-PET for the early diagnosis of Alzheimer’s disease dementia and other dementias in people with mild cognitive impairment (MCI). Cochrane Database Syst Rev 7, CD010386.

Blennow

, Dubois

, Fagan

, Lewczuk

, de Leon

, Hampel

(2015) Clinical utility of cerebrospinal fluid biomarkers in the diagnosis of early Alzheimer’s disease. Alzheimers Dement 11, 58–69.

10.

Park

, Kang

, Ki

, Roh

, Youn

, Kim

(2015) A Consensus in Korea regarding a protocol to reduce preanalytical sources of variability in the measurement of the cerebrospinal fluid biomarkers of Alzheimer’s disease. J Clin Neurol 11, 132–141.

11.

Christensen

(2001) What cognitive changes can be expected with normal ageing?. Aust N Z J Psychiatry 35, 768–775.

12.

Olsson

, Vanderstichele

, Andreasen

, De Meyer

, Wallin

, Holmberg

, Rosengren

, Vanmechelen

, Blennow

(2005) Simultaneous measurement of β-amyloid (1-42), total tau, and phosphorylated tau (Thr181) in cerebrospinal fluid by the xMAP technology. Clin Chem 51, 336–345.

13.

Tapiola

, Alafuzoff

, Herukka

, Parkkinen

, Hartikainen

, Soininen

, Pirttila

(2009) Cerebrospinal fluid β-amyloid 42 and tau proteins as biomarkers of Alzheimer-type pathologic changes in the brain. Arch Neurol 66, 382–389.

14.

Galasko

, Chang

, Motter

, Clark

, Kaye

, Knopman

, Thomas

, Kholodenko

, Schenk

, Lieberburg

, Miller

, Green

, Basherad

, Kertiles

, Boss

, Seubert

(1998) High cerebrospinal fluid tau and low amyloid beta42 levels in the clinical diagnosis of Alzheimer disease and relation to apolipoprotein E genotype. Arch Neurol 55, 937–945.

15.

Hulstaert

, Blennow

, Ivanoiu

, Schoonderwaldt

, Riemenschneider

, De Deyn

, Bancher

, Cras

, Wiltfang

, Mehta

(1999) Improved discrimination of AD patients using β-amyloid (1-42) and tau levels in CSF. Neurology 52, 1555.

16.

Jagust

, Landau

, Shaw

, Trojanowski

, Koeppe

, Reiman

, Foster

, Petersen

, Weiner

, Price

, Mathis

(2009) Relationships between biomarkers in aging and dementia. Neurology 73, 1193–1199.

17.

Kapaki

, Paraskevas

, Zalonis

, Zournas

(2003) CSF tau protein and β-amyloid (1-42) in Alzheimer’s disease diagnosis: Discrimination from normal ageing and other dementias in the Greek population. Eur J Neurol 10, 119–128.

18.

Mulder

, Verwey

, van der Flier

, Bouwman

, Kok

, van Elk

, Scheltens

, Blankenstein

(2010) Amyloid-β (1-42), total tau, and phosphorylated tau as cerebrospinal fluid biomarkers for the diagnosis of Alzheimer disease. Clin Chem 56, 248–253.

19.

Park

, Kim

Y-J

, Lee

, Park

, Chae

, Yim

, Seo

(2013) Preliminary study for a multicenter study of Alzheimer’s disease cerebrospinal fluid biomarkers. Dement Neurocognitive Disord 12, 1–8.

20.

Schoonenboom

, Reesink

, Verwey

, Kester

, Teunissen

, Van De Ven

, Pijnenburg

, Blankenstein

, Rozemuller

, Scheltens

(2012) Cerebrospinal fluid markers for differential dementia diagnosis in a large memory clinic cohort. Neurology 78, 47–54.

21.

Shaw

, Vanderstichele

, Knapik-Czajka

, Clark

, Aisen

, Petersen

, Blennow

, Soares

, Simon

, Lewczuk

(2009) Cerebrospinal fluid biomarker signature in Alzheimer’s disease neuroimaging initiative subjects. Ann Neurol 65, 403–413.

22.

Toledo

, Brettschneider

, Grossman

, Arnold

, Hu

, Xie

, Lee

VM-Y

, Shaw

, Trojanowski

(2012) CSF biomarkers cutoffs: The importance of coincident neuropathological diseases. Acta Neuropathol 124, 23–35.

23.

Jack

, Knopman

, Weigand

, Wiste

, Vemuri

, Lowe

, Kantarci

, Gunter

, Senjem

, Ivnik

(2012) An operational approach to National Institute on Aging–Alzheimer’s Association criteria for preclinical Alzheimer disease. Ann Neurol 71, 765–775.

24.

Galasko

, Clark

, Chang

, Miller

, Green

, Motter

, Seubert

(1997) Assessment of CSF levels of tau protein in mildly demented patients with Alzheimer’s disease. Neurology 48, 632–635.

25.

Lee

, Youn

, Park

K-Y

, Min

J-H

, Kwon

O-S

, Lee

H-O

, Hong

H-J

(2008) The Effect of acetylcholine esterase inhibitor on cerebrospinal fluid β-amyloid 1-42 and phosphorylated tau protein in Korean Alzheimer’s disease patients: Preliminary study. J Korean Neurol Assoc 26, 220–230.

26.

Blennow

, Hampel

, Weiner

, Zetterberg

(2010) Cerebrospinal fluid and plasma biomarkers in Alzheimer disease. Nat Rev Neurol 6, 131–144.

27.

Mattsson

, Andreasson

, Persson

, Arai

, Batish

, Bernardini

, Bocchio-Chiavetto

, Blankenstein

, Carrillo

, Chalbot

(2011) The Alzheimer’s Association external quality control program for cerebrospinal fluid biomarkers. Alzheimers Dement 7, 386–395. e386

28.

Fagan

, Shaw

, Xiong

, Vanderstichele

, Mintun

, Trojanowski

, Coart

, Morris

, Holtzman

(2011) Comon of analytical platforms for cerebrospinal fluid measures of β-amyloid 1-42, total tau, and p-tau181 for identifying alzheimer disease amyloid plaque pathology. Arch Neurol 68, 1137–1144.

29.

Toombs

, Paterson

, Schott

, Zetterberg

(2014) Amyloid-beta 42 adsorption following serial tube transfer. Alzheimers Res Ther 6, 5.

30.

Rosenling

, Slim

, Christin

, Coulier

, Shi

, Stoop

, Bosman

, Suits

, Horvatovich

, Stockhofe-Zurwieden

(2009) The effect of preanalytical factors on stability of the proteome and selected metabolites in cerebrospinal fluid (CSF). J Proteome Res 8, 5511–5522.

31.

Schoonenboom

, Mulder

, Vanderstichele

, Van Elk

E-J

, Kok

, Van Kamp

, Scheltens

, Blankenstein

(2005) Effects of processing and storage conditions on amyloid β (1-42) and tau concentrations in cerebrospinal fluid: Implications for use in clinical practice. Clin Chem 51, 189–195.

32.

Teunissen

, Petzold

, Bennett

, Berven

, Brundin

, Comabella

, Franciotta

, Frederiksen

, Fleming

, Furlan

(2009) A consensus protocol for the standardization of cerebrospinal fluid collection and biobanking. Neurology 73, 1914–1922.

33.

Jack

, Lowe

, Senjem

, Weigand

, Kemp

, Shiung

, Knopman

, Boeve

, Klunk

, Mathis

(2008) 11C PiB and structural MRI provide complementary information in imaging of Alzheimer’s disease and amnestic mild cognitive impairment. Brain 131, 665–680.

34.

Zhang

, Han

, Tan

, Feng

, Guo

, Ding

(2012) Diagnostic accuracy of 18F-FDG and 11C-PIB-PET for prediction of short-term conversion to Alzheimer’s disease in subjects with mild cognitive impairment. Int J Clin Pract 66, 185–198.

35.

Mormino

, Kluth

, Madison

, Rabinovici

, Baker

, Miller

, Koeppe

, Mathis

, Weiner

, Jagust

(2009) Episodic memory loss is related to hippocampal-mediated β-amyloid deposition in elderly subjects. Brain 132, 1310–1323.

36.

Chun

, Jang

J-W

, Park

, Bang

, Park

, Kim

(2014) Improved quantification methods for brain β-amyloid burden on 11C-PIB PET in patients with Alzheimer’s disease. Alzheimers Dement 10, P538.

37.

Leuzy

, Carter

, Chiotis

, Almkvist

, Wall

, Nordberg

(2015) Concordance and diagnostic accuracy of [11 C] PIB PET and cerebrospinal fluid biomarkers in a sample of patients with mild cognitive impairment and Alzheimer’s disease. J Alzheimers Dis 45, 1077–1088.

38.

Zwan

, van Harten

, Ossenkoppele

, Bouwman

, Teunissen

, Adriaanse

, Lammertsma

, Scheltens

, van Berckel

, van der Flier

(2014) Concordance between cerebrospinal fluid biomarkers and [11C]PIB PET in a memory clinic cohort. J Alzheimers Dis 41, 801–807.

39.

Bateman

, Xiong

, Benzinger

, Fagan

, Goate

, Fox

, Marcus

, Cairns

, Xie

, Blazey

(2012) Clinical and biomarker changes in dominantly inherited Alzheimer’s disease. N Engl J Med 367, 795–804.

40.

Ikonomovic

, Klunk

, Abrahamson

, Mathis

, Price

, Tsopelas

, Lopresti

, Ziolko

, Bi

, Paljug

(2008) Post-mortem correlates of in vivo PiB-PET amyloid imaging in a typical case of Alzheimer’s disease. Brain 131, 1630–1645.

41.

Schöll

, Wall

, Thordardottir

, Ferreira

, Bogdanovic

, Långström

, Almkvist

, Graff

, Nordberg

(2012) Low PiB PET retention in presence of pathologic CSF biomarkers in Arctic APP mutation carriers. Neurology 79, 229–236.