Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Arch Neurol. Author manuscript; available in PMC 2013 August 1.
Published in final edited form as:
PMCID: PMC3528180

Comparison of Cerebrospinal Fluid Levels of Tau and Aß1-42 in Alzheimer’s disease and Frontotemporal Degeneration Using Two Analytical Platforms



To utilize values of cerebrospinal fluid (CSF) tau and ß-amyloid obtained from two different analytical immunoassays to differentiate Alzheimer’s disease (AD) from frontotemporal lobar degeneration (FTLD).


CSF values of total tau (t-tau) and ß-amyloid (Aß1-42) obtained using the INNOTEST® ELISA were transformed using a linear regression model to equivalent values obtained using the INNO-BIA AlzBio3™ (xMAP Luminex) assay. Cutoff values obtained from the xMAP assay were developed in a series of autopsy-confirmed cases and cross-validated in another series of autopsy-confirmed samples using transformed ELISA values to assess sensitivity and specificity for differentiating AD from FTLD.


Tertiary memory disorders clinics and neuropathological and biomarker core centers.


75 samples from patients with CSF data obtained from both assays were used for transformation of ELISA values. 40 autopsy-confirmed cases (30 AD, 10 FTLD) were used to establish diagnostic cutoff values, and then cross-validated in a second sample set of 21 autopsy-confirmed cases (11 AD, 10 FTLD) with transformed ELISA values.

Main outcome measure

Diagnostic accuracy using transformed biomarker values.


Data obtained from both assays were highly correlated. The t-tau:Aß1-42 ratio had the highest correlation between measures (r=0.928, p<0.001) and high reliability of transformation (ICC=0.89). A cutoff of 0.34 for the t-tau:Aß1-42 ratio had 90% and 100% sensitivity and 96.7% and 91% specificity to differentiate FTLD cases in the validation and cross-validation samples, respectively.


Values from two analytical platforms can be transformed into equivalent units, which can distinguish AD from FTLD more accurately than the clinical diagnosis.


Prediction of underlying neuropathology of neurodegenerative disease patients is difficult in clinical practice due to the vast heterogeneity and overlapping clinical presentations of these disorders. This is exemplified by atypical presentations of Alzheimer’s disease (AD) mimicking the behavioral-variant of frontotemporal degeneration (bvFTD)1, 2, corticobasal syndrome (CBS)3, primary progressive aphasia,4, 5 and other FTLD spectrum disorders. Indeed, approximately 20% of clinically-diagnosed FTLD patients have a diagnosis of AD at autopsy.6 Conversely, FTLD spectrum pathology can present with an amnestic syndrome clinically resembling AD.7 With the emergence of disease-modifying treatments for neurodegenerative diseases, it will be of upmost importance to accurately identify the underlying neuropathology in these patients. Biomarkers of disease are crucial for this purpose, and new diagnostic criteria for AD8, 9 and FTLD10, 11 incorporate biofluid and neuroimaging biomarkers for research purposes.

Cerebrospinal fluid (CSF) values of the major constituents of AD pathology, tau and ß-amyloid, (Aß) have been widely studied in AD and mild cognitive impairment (MCI) patients during the ongoing Alzheimer’s disease neuroimaging initiative (ADNI) study, with higher levels of total tau (t-tau) and lower Aβ1-42 values observed compared with controls.12-16 Using these measurements, our group has recently reported high sensitivity and specificity in differentiating AD from non-demented controls13 and predicting MCI conversion to AD.13, 14 These biomarkers are less established in FTLD patients, with some studies showing higher levels of CSF t-tau in FTLD compared with controls17-23 while others find no difference,24 or decreased levels in some FTLD subtypes.19 The observed CSF t-tau elevation in these reports is intermediate to the higher values observed in AD cases. Aβ1-42 has also been reported at levels intermediate to controls and the lower levels seen in AD,21, 22 or similar to control patient values.18, 24 Using autopsy confirmed cases, our group previously showed lower levels of t-tau and t-tau:Aß1-42 ratio in FTLD CSF compared with AD.25, 26 Comparative studies are crucial to demonstrate that findings do not merely reflect the non-specific presence of any central nervous system change. Nevertheless, reasons for these discrepancies most importantly include lack of autopsy-confirmed cases in a disease with considerable clinical heterogeneity. Other contributing factors include small patient numbers, and variability in test center processing of samples.26, 27

Two commercially available immunoassays are available to measure these CSF analytes. Concentrations of tau and Aß1-42 obtained using the INNOTEST® (ELISA) compared with the INNO-BIA AlzBio3™ luminex (xMAP) platform differ substantially; however, values from these two immunoassays are highly correlated,28, 29 suggesting values from one platform can be transformed into equivalent units of the other. Combining these data is advantageous, as it allows for increased sample sizes to fully utilize valuable research samples.

In this work, we utilize a linear regression model to transform values obtained from the ELISA method to equivalent units of tau and Aß detected by the xMAP luminex platform. Using these transformed data, we show that autopsy-confirmed AD and FTLD patients can be differentiated with high sensitivity and specificity.



Data from patients followed at the Alzheimer’s disease center (ADC) or Frontotemporal degeneration center (FTDC) at the University of Pennsylvania (PENN) were included for analysis. ELISA and xMAP CSF values of total tau (t-tau), phosphorylated tau (p-tau181), Aß1-42, t-tau:Aß1-42 ratio, p-tau181:Aß1-42 ratio, and the neuropathological and genetic diagnosis were obtained from the integrated neurodegenerative disease database at PENN.30 10 autopsy-confirmed cases (FTDC) were previously reported using ELISA analysis only,25, 26 and 36 autopsy cases (ADC) had previous xMAP values reported in an exploratory study of novel AD CSF biomarkers.31

Transformation of ELISA values was performed using data from 75 patients with available CSF biomarker data obtained from both methods. Different aliquots from the same initial CSF collection were utilized for these cases, with limitation to one freeze-thaw cycle in most instances. Five cases used in the transformation data set were also used in the autopsy-confirmed samples.

Evaluation and establishment of diagnostic cut-off values for CSF analytes using the xMAP system was performed in a sample of 40 autopsy-confirmed cases (Sample 1) with a neuropathological diagnosis of AD or FTLD spectrum disorders from the ADC. Cross-validation of the diagnostic cut-off value was performed in a second sample set of 21 autopsy-confirmed cases from the FTDC (Sample 2) using transformed ELISA values. To balance these groups, five FTLD cases (four with known pathogenic mutations in the MAPT or PGRN gene, as the underlying neuropathology is universally FTLD-tau and FTLD-TDP, respectively)32 from the FTDC were included in Sample 1 and one ADC AD case added to Sample 2. Autopsy-confirmed cases of FTLD included the following neuropathological diagnoses: FTLD with TDP-43 inclusions (n=5), amyotrophic lateral sclerosis with FTLD (ALS-FTLD, n=1)-collectively referred to as FTLD-TDP (n=9 including the non-deceased PGRN mutation cases, n=3), and corticobasal degeneration (CBD, n=5), progressive supranuclear palsy (PSP, n=2), tangle-predominant senile dementia (TPSD, n=2)-collectively referred to as FTLD-tau (n=10, including the non-deceased MAPT mutation case, n=1). One FTLD case did not contain significant TDP-43, tau, α-synuclein or FUS inclusions, and was classified as dementia lacking distinctive histopathology (DLDH).32 Thus, the autopsy-confirmed data set had roughly equal numbers of FTLD-TDP and FTLD-tau. All AD cases carried a primary neuropathological diagnosis of “high probability AD.”33 Demographic data were compared between groups using chi-square tests for categorical variables and independent t-tests or Mann-Whitney U tests for continuous variables, as appropriate (Table 1).34 Missing data included three cases in the transformation sample (age of onset) and one case in the transformation and sample 2 (age at CSF collection).

Table 1
Demographics of study patients

All procedures, including CSF fluid collection and autopsy, required informed consent and were performed in accordance with the rules of the local institutional review board at PENN.

Neuropathologic diagnosis

Autopsy was performed as previously described.6 Briefly, fresh brain and spinal cord tissue obtained at autopsy was fixed in neutral buffered formalin or 70% ethanol and 150-mmol sodium chloride, embedded in paraffin blocks, and cut into 6 μm sections for microscopic analysis. Routine staining was performed on each case, including hematoxylin & eosin and the amyloid-binding dye Thioflavin-S, and immunohistochemistry using well-characterized monoclonal antibodies specific for α-synuclein, tau, and TDP-43, which are found in characteristic inclusions seen in most neurodegenerative diseases. Microscopic diagnosis was made by an experienced neuropathologist (JQT) using current neuropathological diagnostic criteria for neurodegenerative diseases.32, 33, 35

Biofluid Collection/Analysis

CSF samples were obtained during routine diagnostic lumbar puncture as previously described.25 In brief, lumbar puncture was performed at the L3/L4 lumbar space using a 20-gauge needle to collect about 20 ml of CSF in polypropylene tubes (Corning Life Sciences, Lowell, MA). Samples were centrifuged at 3,000 rpm for 15 minutes at 4°C, aliquotted, and immediately stored at −80 °C until analysis.

Samples were analyzed using the ELISA assay (INNOTEST®, Innogenetics, Ghent, Belgium) or the Luminex xMAP platform (INNO-BIA AlzBio3™ for research only reagents, Innogenetics-Fujirebio, Ghent, Belgium) at the biomarker core at PENN according to previous reports.13, 25, 28 Monoclonal (MAb) capture and reporting antibodies used in the ELISA method for detection of t-tau and p-tau181 in CSF were AT120/HT7 and BT2, HT7/AT270, respectively. The ELISA values for Aß1-42 were measured using an “in house” ELISA method36 with the MAb BAN-50 as the capture, and BC-05 as the reporting MAb. The xMAP platform utilized the capture MAbs 4D7A3 (Aß1-42), AT120 (t-tau), and AT270 (p-tau181) bound to color-specific beads. The biomarker analytes were detected using the reporting MAbs 3D6 (Aß1-42) and HT7 (t-tau, p-tau181).

Statistical Analysis

Intra-assay coefficients of variation (%CVs) were calculated for both immunoassays using measurements from duplicate analysis from single runs (data missing for one case) and reported as mean and standard deviation.

The transformation, validation, and cross validation steps are summarized in Figure 1. To transform the ELISA values to xMAP, a linear regression model was applied on the raw and natural-log transformed values of the training dataset (n=52). Then, the obtained formula was applied on ELISA values in the test dataset (n=23) and the intra-class correlation coefficient (ICC) was measured. We selected the best transformation results (based on raw or natural log transformation) to select the transformation formula.

Figure 1
Flow chart of the (A) transformation and (B) validation and cross-validation steps.

The diagnostic utility of CSF biomarker levels in differentiating AD from FTLD cases was established in a separate sample set of autopsy-confirmed cases with available xMAP values (n=40). A receiver operating characteristic (ROC) curve analysis was performed for all analytes and assessed for optimal sensitivity and specificity for best test accuracy. The t-tau: Aß1-42 ratio had the highest area under the curve (AUC) compared to exploratory analyses assessing t-tau, p-tau181, Aß1-42, and p-tau181:Aß1-42 ratio, and thus was used in subsequent analysis. The diagnostic cutoff value of the t-tau:Aß1-42 ratio obtained in the xMAP sample was applied to the transformed ELISA data in a separate cross-validation sample set (n=21). Analyses were performed using SPSS 19.0 (SPSS, Chicago, Ill) and R version 2.13.37

Sensitivity and specificity of the ante mortem clinical diagnosis (FTLD spectrum or AD) was calculated for comparison. A clinical diagnosis of logopenic progressive aphasia (LPA, n=3) was considered an accurate identification of AD pathology, as the majority of these cases are atypical presentations of AD neuropathology5.


Transformation of ELISA values

Percent CVs for ELISA and xMAP were: tau (5.3+7.6%, 4.9+8.2%), ptau (3.4+7.6%, 3.9+4.3%) and Aβ1-42 (8.6+6.7%, 3.8+5.2%), respectively. Seventy-five subjects with natural log-transformed CSF values from both ELISA and xMAP immunoassays were used for transformation of values (Table 1). This sample was divided randomly into training (n=52) and test (n=23) samples. Natural log transformed data had the best correlation between the two immunoassays for most analytes, with CSF values of Aß1-42 (r=0.819, p<0.001), t-tau (r=0.890, p<0.001), p-tau181 (r=0.779, p<0.001), t-tau:Aß1-42 ratio (r=0.928, p<0.001) and p-tau181: Aß1-42 ratio (r=0.834, p<0.001) (Figure 2, Panels A-E). When the regression model was used to transform data in the test sample, the ICCs showed modest to high reliability, ranging from 0.63 to 0.89 (Figure 2, Panels A-E). The linear regression model for the t-tau:Aß1-42 ratio yielded the formula: ((ln(value)-1.513562)/1.040762) to convert ELISA values, which was used in subsequent analyses.

Figure 2
Transformation of CSF analytes into equivalent values between platforms. Shown are plots of raw and natural-log transformed values of (A) Aß1-42, (B) t-tau, (C) p-tau181, (D) t-tau:Aß1-42 ratio, and (E) p-tau181:Aß1-42 ratio obtained ...

Diagnostic Accuracy of transformed values

ROC curve analysis using xMAP values from a cohort of autopsy-confirmed cases (n=30, 20 AD, 10 FTLD) showed the highest diagnostic accuracy using the t-tau:Aß1-42 ratio (AUC=0.989, sensitivity= 90% and specificity=96.7% for best test accuracy) (Figure 3). Using the cutoff of 0.34 (ln value= −1.078) we correctly identified 29 of 30 AD patients and 9 of 10 FTLD cases (90% sensitivity, 96.7% specificity) and outperformed the clinical diagnosis (86.7% sensitivity, 66.7% specificity) (Figure 4). This t-tau:Aß1-42 ratio value was then used for cross-validation in the transformed ELISA data set due to its high diagnostic accuracy and correlation between assays.

Figure 3
Receiver operating characteristic curve analysis of xMAP analyte values in an autopsy-confirmed sample (neuropathological sample 1). The t-tau:Aß1-42 ratio had the highest area under the curve at the optimal diagnostic cut-point of 0.34.
Figure 4
Diagnostic accuracy of t-tau:Aß1-42 ratio cutoff in the validation and cross-validation data sets. Shown is a (A) box-plot of AD and FTLD values from both samples. (B) The t-tau:Aß1-42 ratio was highly sensitive and specific for identifying ...

The cutoff value of 0.34 for the t-tau:Aß1-42 ratio obtained using xMAP data correctly identified 10 of 11 AD cases and 10 of 10 FTLD cases in the transformed value cross-validation sample (100% sensitivity and 91% specificity) compared with the clinical diagnosis (36.5% sensitivity, 100% specificity). Thus, the t-tau:Aß1-42 ratio effectively distinguished FTLD from AD autopsy cases in both the xMAP and cross-validation transformed value data sets with superior accuracy than the ante mortem clinical diagnosis (Figure 4, panel B). Individual analysis of the cases misclassified by our system reveal one genetic (PGRN) FTLD case (t-tau:Aß1-42 ratio = 0.40) and one high probability AD case in both the xMAP sample (t-tau:Aß1-42 ratio = 0.29), and the transformed ELISA set (t-tau:Aß1-42 ratio=0.27).


We have confirmed our previous data showing a lower CSF t-tau:Aß1-42 ratio in FTLD compared with AD in a much larger autopsy-confirmed sample.25, 26 In addition, we demonstrate that CSF biomarker analysis can be compared directly between the ELISA and xMAP analytical platforms. The transformed data were highly sensitive and specific in correctly differentiating autopsy-confirmed cases of AD from FTLD in a clinically demented sample, with added sensitivity and specificity to the clinical diagnosis.

These findings compliment previous work showing that AD biomarkers obtained from these two immunoassays are highly correlated28, 29, 38 and can be transformed by a conversion factor.28 Others have suggested that values obtained from these platforms cannot be converted due to a high coefficient of variation for the xMAP:ELISA ratio of raw biomarker values.39 Recent work from our group has shown effective transformation of ELISA biomarker data into equivalent xMAP values in differentiating AD from normal controls.40 The linear regression model used in that study was similar to our formula here, extending the generalizability of such a transformation method. Moreover, the present report extends this approach to a comparative study, and provides autopsy-confirmed validation. Further validation of this method is exemplified by previous work showing an equivalent ability of the t-tau:Aβ1-42 ratio values independently obtained from both platforms to distinguish patients with evidence of in vivo amyloidosis.29 Thus, the t-tau:Aβ1-42 ratio values obtained from these two assays have comparable diagnostic accuracy for AD neuropathology, despite differing absolute values.

The individual cases misclassified by our system reveal one non-deceased genetic (PGRN) FTLD case and two AD cases, both of whom had no co-morbid neuropathologic findings, and interestingly had atypical clinical presentations of LPA and bvFTD. Since the PGRN case carries a known pathogenic mutation (c.102delC), it certainly will contain TDP-43 pathology at autopsy; however, co-morbid AD pathology cannot be ruled out. The age of this patient at time of CSF collection was 68 years old, indicating the possibility of age-associated Aß amyloidosis which could influence the t-tau:Aß1-42 ratio. Indeed, another FTLD case that was very close to the diagnostic threshold but correctly identified in the transformed data set (t-tau:Aß1-42 ratio=0.290) had a neuropathological diagnosis of CBD pathology with co-morbid Aβ amyloidosis (The Consortium to Establish a Registry for Alzheimer’s Disease41 CERAD plaque score C). The close-to-diagnostic-threshold elevated ratio in this case is most likely due to the relative lower value of Aß1-42 (ELISA value of 321.94 pg/ml), suggesting that FTLD cases with significant co-morbid AD pathology may have values of tau and Aß1-42 that are more typical of AD, which can complicate clinical interpretation of CSF biomarker analysis in living patients. Since most FTLD cases are relatively young, this reduces the likelihood of age-associated amyloidosis. Utilizing in vivo amyloid imaging or other modalities may help improve diagnostic accuracy of mixed-pathology cases.

Limitations to this study include lack of autopsy-confirmed non-demented controls and other neurodegenerative dementias, as study of mixed dementia groups may be more applicable to clinical practice;42 however, this represents a diagnostic challenge beyond the scope of this work. We have shown previously that CSF levels of these biomarkers cannot accurately differentiate FTLD cases from non-demented control patients,26 although the recent availability of clinical criteria for bvFTD10 and PPA11 reduces the likelihood that individuals with an FTLD spectrum clinical disorder will be confused with healthy adults. Additionally, non-progressive, non-neurodegenerative patients with cognitive/behavioral symptoms resembling FTLD (phenocopy syndrome) can be accurately distinguished from patients with underlying FTLD-spectrum neurpathology by serial clinical evaluations.43

A major strength of this study is the use of autopsy-confirmed cases in the validation and cross-validation steps.44 Indeed, the importance of autopsy confirmed samples in FTLD biomarker research is highlighted here, as the diagnostic accuracy outperformed the clinical diagnosis in both centers. Since sample 2 was derived mainly from the FTDC, the majority these AD cases had atypical clinical syndromes (i.e. CBS, bvFTD, and semantic dementia), with resultant lower clinical diagnostic sensitivity for AD pathology. This discrepancy in clinical presentations of AD pathology between samples should not influence our findings here, as these cases do not have a CSF biomarker signature that would alter the t-tau:Aß1-42 ratio4, 45, 46; however, it does exemplify the vast heterogeneity and diagnostic challenges of this clinical spectrum of disease, and underlines the usefulness of CSF biomarkers to distinguish FTLD from atypical presentations of AD.

The transformed ELISA sample had an earlier age of onset (p=0.008), CSF collection (p=0.003) and death (p=0.001) compared with the xMAP sample, and a shorter interval between CSF collection and autopsy (p=0.001) (Table 1). This is most likely due to the majority of typical amnestic AD cases in the xMAP sample which would be expected to have a longer duration of illness compared with FTLD-spectrum diseases.47 The annual variation in AD CSF biomarkers is small for AD patients after the onset of dementia,48, 49 while the longitudinal profile of these biomarkers in FTLD is less clear; there was no significant difference between groups in the interval from reported onset of dementia to CSF collection (p=0.955), thus, these differences in demographics between groups should have minimal influence on CSF analyte levels.

There is significant utility in combining values obtained from these analytical platforms, as obtaining CSF samples from patients is invasive and may be limited in size for multiple analyses. In addition, samples from longitudinally followed autopsy-confirmed cases are extremely valuable research tools. Combining data sets from these two methods helps conserve these precious biofluid samples and expands available sample sizes for future studies. Previous studies have shown that developing a universal AD CSF biomarker diagnostic cutoff value for use between centers is very difficult, due to multiple sources of variability within- and between-laboratories that need to be harmonized,15, 50 limiting the immediate clinical application of CSF analysis in dementia diagnosis; however, our data here support the combined use of these immunoassay platforms in a research setting. Of note, the data was obtained from two different laboratories within one institution with acceptable intra-assay variability.

That said, this study emphasizes the continuing need to standardize all aspects of biomarker methods and research protocols so that data from different centers can be compared worldwide. This will greatly facilitate understanding the pathobiology of biomarker changes and define best practices for applying biomarker technologies, especially in the context of AD clinical trials that increasingly are carried out on a global scale.

With these caveats in mind, our work provides a method for maximizing use of valuable research samples and reinforces the utility of AD biomarker profiles, most specifically the t-tau:Aβ1-42 ratio, in an autopsy-confirmed sample differentiating FTLD from AD. These findings further highlight the need for FTLD-specific biomarkers51-53 and the potential value of a multi-modal approach combining clinical, neuroimaging and biofluid biomarkers to increase ante-mortem diagnostic accuracy for neurodegenerative diseases27 in clinical practice.


This study was supported by the NIH P30AG010124-20, P01 AG017586, R01 NS44266, R01 AG15116, P01 AG32953, P01 NS53488 and the Wyncote Foundation. DJI is supported by the NIH T32-AG000255 training grant and JBT is supported by a grant of the Alfonso Martín Escudero Foundation.


Author Contributions

Study concept and design Grossman, McMillan, Irwin, and Toledo. Acquisition of data: Grossman, Arnold, Trojanowski, and Lee. Analysis and interpretation of data: McMillan, Toledo, Irwin, Grossman, Trojanowski, Lee, Shaw, and Wang. Drafting of the manuscript: Irwin, Toledo, McMillian, and Grossman Critical revision of the manuscript for important intellectual content Irwin, Toledo, McMillan, Grossman, Arnold, Trojanowski, Lee, Shaw, and Wang. Statistical Analysis Toledo, McMillan, and Irwin. Obtained funding Grossman, Arnold, Trojanowski, and Lee. Study Supervision Grossman.


1. Johnson JK, Head E, Kim R, Starr A, Cotman CW. Clinical and pathological evidence for a frontal variant of Alzheimer disease. Arch Neurol. 1999 Oct;56(10):1233–1239. [PubMed]
2. Grossman M, Libon DJ, Forman MS, et al. Distinct antemortem profiles in patients with pathologically defined frontotemporal dementia. Arch Neurol. 2007 Nov;64(11):1601–1609. [PubMed]
3. Boeve BF, Maraganore DM, Parisi JE, et al. Pathologic heterogeneity in clinically diagnosed corticobasal degeneration. Neurology. 1999 Sep 11;53(4):795–800. [PubMed]
4. Hu WT, McMillan C, Libon D, et al. Multimodal predictors for Alzheimer disease in nonfluent primary progressive aphasia. Neurology. 2010 Aug 17;75(7):595–602. [PMC free article] [PubMed]
5. Grossman M. Primary progressive aphasia: clinicopathological correlations. Nat Rev Neurol. 2010 Feb;6(2):88–97. [PMC free article] [PubMed]
6. Forman MS, Farmer J, Johnson JK, et al. Frontotemporal dementia: clinicopathological correlations. Ann Neurol. 2006 Jun;59(6):952–962. [PMC free article] [PubMed]
7. Graham A, Davies R, Xuereb J, et al. Pathologically proven frontotemporal dementia presenting with severe amnesia. Brain. 2005 Mar;128(Pt 3):597–605. [PubMed]
8. Jack CR, Jr., Albert MS, Knopman DS, et al. Introduction to the recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011 May;7(3):257–262. [PMC free article] [PubMed]
9. McKhann GM, Knopman DS, Chertkow H, et al. 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. 2011 May;7(3):263–269. [PMC free article] [PubMed]
10. Rascovsky K, Hodges JR, Knopman D, et al. Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia. Brain. 2011 Aug 2; [PMC free article] [PubMed]
11. Gorno-Tempini ML, Hillis AE, Weintraub S, et al. Classification of primary progressive aphasia and its variants. Neurology. 2011 Mar 15;76(11):1006–1014. [PMC free article] [PubMed]
12. Weiner MW, Aisen PS, Jack CR, Jr., et al. The Alzheimer’s disease neuroimaging initiative: progress report and future plans. Alzheimers Dement. 2010 May;6(3):202–211. e207. [PMC free article] [PubMed]
13. Shaw LM, Vanderstichele H, Knapik-Czajka M, et al. Cerebrospinal fluid biomarker signature in Alzheimer’s disease neuroimaging initiative subjects. Ann Neurol. 2009 Apr;65(4):403–413. [PMC free article] [PubMed]
14. De Meyer G, Shapiro F, Vanderstichele H, et al. Diagnosis-independent Alzheimer disease biomarker signature in cognitively normal elderly people. Arch Neurol. 2010 Aug;67(8):949–956. [PMC free article] [PubMed]
15. Shaw LM, Vanderstichele H, Knapik-Czajka M, et al. Qualification of the analytical and clinical performance of CSF biomarker analyses in ADNI. Acta Neuropathol. 2011 May;121(5):597–609. [PMC free article] [PubMed]
16. Trojanowski JQ, Vandeerstichele H, Korecka M, et al. Update on the biomarker core of the Alzheimer’s Disease Neuroimaging Initiative subjects. Alzheimers Dement. 2010 May;6(3):230–238. [PMC free article] [PubMed]
17. van Harten AC, Kester MI, Visser PJ, et al. Tau and p-tau as CSF biomarkers in dementia: a meta-analysis. Clin Chem Lab Med. 2011 Mar;49(3):353–366. [PubMed]
18. Pijnenburg YA, Schoonenboom NS, Rosso SM, et al. CSF tau and Abeta42 are not useful in the diagnosis of frontotemporal lobar degeneration. Neurology. 2004 May 11;62(9):1649. [PubMed]
19. Arai H, Morikawa Y, Higuchi M, et al. Cerebrospinal fluid tau levels in neurodegenerative diseases with distinct tau-related pathology. Biochem Biophys Res Commun. 1997 Jul 18;236(2):262–264. [PubMed]
20. Green AJ, Harvey RJ, Thompson EJ, Rossor MN. Increased tau in the cerebrospinal fluid of patients with frontotemporal dementia and Alzheimer’s disease. Neurosci Lett. 1999 Jan 8;259(2):133–135. [PubMed]
21. Riemenschneider M, Wagenpfeil S, Diehl J, et al. Tau and Abeta42 protein in CSF of patients with frontotemporal degeneration. Neurology. 2002 Jun 11;58(11):1622–1628. [PubMed]
22. Kapaki E, Paraskevas GP, Papageorgiou SG, et al. Diagnostic value of CSF biomarker profile in frontotemporal lobar degeneration. Alzheimer Dis Assoc Disord. 2008 Jan-Mar;22(1):47–53. [PubMed]
23. de Souza LC, Lamari F, Belliard S, et al. Cerebrospinal fluid biomarkers in the differential diagnosis of Alzheimer’s disease from other cortical dementias. J Neurol Neurosurg Psychiatry. 2011 Mar;82(3):240–246. [PubMed]
24. Sjogren M, Minthon L, Davidsson P, et al. CSF levels of tau, beta-amyloid(1-42) and GAP-43 in frontotemporal dementia, other types of dementia and normal aging. J Neural Transm. 2000;107(5):563–579. [PubMed]
25. Grossman M, Farmer J, Leight S, et al. Cerebrospinal fluid profile in frontotemporal dementia and Alzheimer’s disease. Ann Neurol. 2005 May;57(5):721–729. [PubMed]
26. Bian H, Van Swieten JC, Leight S, et al. CSF biomarkers in frontotemporal lobar degeneration with known pathology. Neurology. 2008 May 6;70(19 Pt 2):1827–1835. [PMC free article] [PubMed]
27. Bian H, Grossman M. Frontotemporal lobar degeneration: recent progress in antemortem diagnosis. Acta Neuropathol. 2007 Jul;114(1):23–29. [PubMed]
28. Olsson A, Vanderstichele H, Andreasen N, et al. Simultaneous measurement of beta-amyloid(1-42), total tau, and phosphorylated tau (Thr181) in cerebrospinal fluid by the xMAP technology. Clin Chem. 2005 Feb;51(2):336–345. [PubMed]
29. Fagan AM, Shaw LM, Xiong C, et al. Comparison of Analytical Platforms for Cerebrospinal Fluid Measures of {beta}-Amyloid 1-42, Total tau, and P-tau181 for Identifying Alzheimer Disease Amyloid Plaque Pathology. Arch Neurol. 2011 May 9; [PMC free article] [PubMed]
30. Xie SX, Baek Y, Grossman M, et al. Building an integrated neurodegenerative disease database at an academic health center. Alzheimers Dement. 2011 Jul;7(4):e84–e93. [PMC free article] [PubMed]
31. Hu WT, Chen-Plotkin A, Arnold SE, et al. Novel CSF biomarkers for Alzheimer’s disease and mild cognitive impairment. Acta Neuropathol. 2010 Jun;119(6):669–678. [PMC free article] [PubMed]
32. Mackenzie IR, Neumann M, Bigio EH, et al. Nomenclature and nosology for neuropathologic subtypes of frontotemporal lobar degeneration: an update. Acta Neuropathol. 2010 Jan;119(1):1–4. [PMC free article] [PubMed]
33. Hyman BT, Trojanowski JQ. Consensus recommendations for the postmortem diagnosis of Alzheimer disease from the National Institute on Aging and the Reagan Institute Working Group on diagnostic criteria for the neuropathological assessment of Alzheimer disease. J Neuropathol Exp Neurol. 1997 Oct;56(10):1095–1097. [PubMed]
34. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology. 1984 Jul;34(7):939–944. [PubMed]
35. McKeith IG, Dickson DW, Lowe J, et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology. 2005 Dec 27;65(12):1863–1872. [PubMed]
36. Winton MJ, Lee EB, Sun E, et al. Intraneuronal APP, not free Abeta peptides in 3xTg-AD mice: implications for tau versus Abeta-mediated Alzheimer neurodegeneration. J Neurosci. 2011 May 25;31(21):7691–7699. [PMC free article] [PubMed]
37. RDC T. A language and Environment for Statistical Computing R. Foundation for Statistical Computing. 2011 Published Last Modified Date|. Accessed Dated Accessed|.
38. Lewczuk P, Zimmermann R, Wiltfang J, Kornhuber J. Neurochemical dementia diagnostics: a simple algorithm for interpretation of the CSF biomarkers. J Neural Transm. 2009 Sep;116(9):1163–1167. [PubMed]
39. Reijn TS, Rikkert MO, van Geel WJ, de Jong D, Verbeek MM. Diagnostic accuracy of ELISA and xMAP technology for analysis of amyloid beta(42) and tau proteins. Clin Chem. 2007 May;53(5):859–865. [PubMed]
40. Wang L LY, Chang S, Leight S, Malgorzata KC, Shaw L, Lee VMY, Trojanowski JQ, Clark C. Comparison of xMAP and ELISA assays for detecting CSF biomarkers of Alzheimer’s Disease. Journal of Alzheimer’s disease. 2011 under review. [PMC free article] [PubMed]
41. Mirra SS, Heyman A, McKeel D, et al. The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology. 1991 Apr;41(4):479–486. [PubMed]
42. Clark CM, Xie S, Chittams J, et al. Cerebrospinal fluid tau and beta-amyloid: how well do these biomarkers reflect autopsy-confirmed dementia diagnoses? Arch Neurol. 2003 Dec;60(12):1696–1702. [PubMed]
43. Kipps CM, Hodges JR, Hornberger M. Nonprogressive behavioural frontotemporal dementia: recent developments and clinical implications of the ‘bvFTD phenocopy syndrome’ Curr Opin Neurol. 2010 Dec;23(6):628–632. [PubMed]
44. Toledo JB BJ, Grossman M, Arnold SE, Hu WT, Xie S, Lee VMY, Shaw LM, Trojanowski JQ. Improving diagnostic accuracy for dementia: CSF biomarker cutoffs based on clinical and neuropathological criteria. Annals of Neurology. 2011 under review.
45. Seguin J, Formaglio M, Perret-Liaudet A, et al. CSF biomarkers in posterior cortical atrophy. Neurology. 2011 May 24;76(21):1782–1788. [PubMed]
46. Koric L, Felician O, Ceccaldi M. [Use of CSF biomarkers in the diagnosis of Alzheimer’s disease in clinical practice] Rev Neurol (Paris) 2010 Jun-Jul;167(6-7):474–484. [PubMed]
47. Roberson ED, Hesse JH, Rose KD, et al. Frontotemporal dementia progresses to death faster than Alzheimer disease. Neurology. 2005 Sep 13;65(5):719–725. [PubMed]
48. Jack CR, Jr., Knopman DS, Jagust WJ, et al. Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol. 2010 Jan;9(1):119–128. [PMC free article] [PubMed]
49. Vemuri P, Wiste HJ, Weigand SD, et al. Serial MRI and CSF biomarkers in normal aging, MCI, and AD. Neurology. 2010 Jul 13;75(2):143–151. [PMC free article] [PubMed]
50. Mattsson N, Andreasson U, Persson S, et al. The Alzheimer’s Association external quality control program for cerebrospinal fluid biomarkers. Alzheimers Dement. 2011 Jul;7(4):386–395. e386. [PMC free article] [PubMed]
51. Hu WT, Chen-Plotkin A, Grossman M, et al. Novel CSF biomarkers for frontotemporal lobar degenerations. Neurology. 2010 Dec 7;75(23):2079–2086. [PMC free article] [PubMed]
52. Hu WT, Chen-Plotkin A, Arnold SE, et al. Biomarker discovery for Alzheimer’s disease, frontotemporal lobar degeneration, and Parkinson’s disease. Acta Neuropathol. 2010 Sep;120(3):385–399. [PMC free article] [PubMed]
53. Hu WT, Trojanowski JQ, Shaw LM. Biomarkers in frontotemporal lobar degenerations-Progress and challenges. Prog Neurobiol. 2011 Dec;95(4):636–648. [PMC free article] [PubMed]