Impact of cerebro-spinal fluid biomarkers of Alzheimer s disease in clinical practice: a multicentric study

Transcription

1 DOI /s ORIGINAL COMMUNICATION Impact of cerebro-spinal fluid biomarkers of Alzheimer s disease in clinical practice: a multicentric study François Mouton-Liger David Wallon Anne-Cécile Troussière Rachida Yatimi Julien Dumurgier Eloi Magnin Vincent de la Sayette Emannuelle Duron Nathalie Philippi Emilie Beaufils Audrey Gabelle Bernard Croisile Philippe Robert Florence Pasquier Didier Hannequin Jacques Hugon Claire Paquet Received: 27 September 2013 / Accepted: 12 October 2013 Ó Springer-Verlag Berlin Heidelberg 2013 Abstract CSF biomarkers of Alzheimer s disease are well validated in clinical research; however, their pragmatic utility in daily practice is still unappreciated. These biomarkers are used in routine practice according to Health Authority Recommendations. In 604 consecutive patients explored for cognitive disorders, questionnaires were prospectively proposed and filled. Before and after CSF biomarker results, clinicians provided a diagnosis and an estimate of their diagnostic confidence. Analysis has compared the frequency of diagnosis before and after CSF biomarker results using the net reclassification improvement (NRI) method. We have evaluated external validity comparing with data of French Bank National of AD (BNA). A total of 561 patients [Alzheimer s disease (AD), F. Mouton-Liger, D. Wallon, A.-C. Troussière have contributed equally to this work. n = 253; non-ad, n = 308] were included (mean age, 68.6 years; women, 52 %). Clinically suspected diagnosis and CSF results were concordant in 65.2 % of cases. When clinical hypothesis and biological results were discordant, a reclassification occurred in favour of CSF biomarkers results in 76.9 %. The NRI was 39.5 %. In addition, the results show a statistically significant improvement in clinician confidence for their diagnosis. In comparison with BNA data, patients were younger and more frequently diagnosed with AD. Clinicians tend to heavily rely on the CSF AD biomarkers results and are more confident in their diagnoses using CSF AD biomarkers. Thus, these biomarkers appear as a key tool in clinical practice. Keywords Cerebrospinal fluid Biomarkers Alzheimer s disease Clinical practice For the ISAC network. F. Mouton-Liger J. Dumurgier J. Hugon C. Paquet (&) Centre Mémoire de Ressources et de Recherche (CMRR) Paris Nord Ile de France, Groupe hospitalier Lariboisière FW Saint Louis, APHP, Université Paris Diderot, 200, rue du Faubourg Saint Denis, Paris, France claire.paquet@inserm.fr F. Mouton-Liger C. Paquet Inserm U942, Paris, France F. Mouton-Liger C. Paquet Unité d Histologie et de Biologie du Vieillissement, Groupe Hospitalier Lariboisiere FW Saint Louis, APHP, Université Paris Diderot, Paris, France D. Wallon D. Hannequin Centre Mémoire (CMRR) and INSERM UMR1079, Hôpital Universitaire de Rouen, Rouen, France A.-C. Troussière F. Pasquier Centre Mémoire (CMRR) de Lille, Université Lille Nord de France, UDSL, Lille, France R. Yatimi P. Robert CMRR/EA COBTEK, CHU, Université de Nice Sophia Antipolis, Nice, France E. Magnin Centre Mémoire (CMRR) de Besançon, Hôpital Universitaire de Besançon, Besançon, France V. de la Sayette Centre Mémoire (CMRR) de Caen, Hôpital Universitaire de Caen, Caen, France E. Duron Centre Mémoire (CMRR) Paris Sud, Hôpital Broca, APHP, Université Paris V, Paris, France

2 Abbreviations Ab Amyloid b AD Alzheimer disease CT Computed tomography CBD Corticobasal degeneration CSF Cerebrospinal fluid FTD Frontotemporal dementia LBD Lewy body dementia LP Lumbar puncture MCI Mild cognitive impairment MRI Magnetic resonance imaging PET Positron emission tomography P-tau Tau phosphorylated at threonine-181 SPECT Single photon emission computed tomography T-tau Total tau Introduction Memory loss and cognitive alteration leading to dementia are devastating and affect the daily life, the future and the familial environment. In most cases, dementia is due to Alzheimer s disease (AD) and affected patients and their relatives can benefit from specific interventions, treatment or clinical trials in order to live as well as possible [7, 8]. Currently, in clinical practice, the diagnosis is based on the McKahnn s criteria [19] also used in recent cohort [23] and in international clinical trials concerning disease modifying therapies [25]. Although the sensitivity of these standardized criteria is rather high (80 %), their specificity is less than 70 % for probable AD [14]. Furthermore, neuropathological studies have shown that % of patients who fulfilled these criteria did not have AD neuropathological lesions [16, 27]. In living patients, comparable rates were found in clinical trials using new imaging biomarkers such as Pittsburg compound B (PIB) [3, 25]. In these cases, a discordance has emerged between the clinical presentation and the final neuropathological diagnosis, raising the N. Philippi Centre Mémoire (CMRR) de Strasbourg, Hôpital Universitaire Strasbourg, Strasbourg, France E. Beaufils Centre Mémoire (CMRR) de Tours, Hôpital Universitaire Tours, Tours, France A. Gabelle Centre Mémoire (CMRR) de Montpellier, Hôpital Universitaire Montpellier, Montpellier, France B. Croisile Centre Mémoire (CMRR) de Lyon, Hôpital Universitaire Lyon, Lyon, France question of the specificity of AD clinical criteria [24]. However, neuropathological studies, imaging and cerebrospinal fluid (CSF) biomarker studies can reveal atypical AD forms that are more and more frequently described [33] particularly in younger patients [28]. Schoonenbomm et al. [27]. and Andreasson et al. [2] have shown that the use of CSF biomarkers improve the validity of clinical criteria. In light of these works, AD diagnosis appears to be a challenge when only based on presenting clinical features without specific biomarkers, and could lead to a marked unwanted uncertainty [20]. Actually, positive biomarkers mean that patients displayed AD brain lesions that could not be necessarily associated with clinically detectable AD features. Inversely, AD diagnosis cannot only rely on positive biomarkers because the efficacy is not yet validated at the earliest stage of the disease. Taking into account these remarks, new criteria were proposed for AD [20], mild cognitive impairment (MCI) [1] and before clinical stage [29] including new specific markers such as CSF and amyloid imaging biomarkers. Through these works and the setting of an international task force, it seems necessary to determine very early whether the symptoms of patients are due to AD or not. While CSF biomarkers are now proposed and performed, little is known about how they impact the daily clinical practice. CSF is in direct contact with the brain and reflects its biochemical activity. CSF biomarkers are the only markers that give an image of AD classical neuropathological lesions: amyloid with CSF Ab1-42 and tau pathologies with CSF total (T-tau) and phosphorylated-tau (P-tau). Several studies have reported strong correlations between CSF biomarkers and neuropathology load [6, 27, 30, 31]. Then, CSF biomarkers are well validated in clinical research [4, 5, 26]; however, the real utility in the daily practice is still unappreciated. One single-site study explored the impact of CSF biomarker on final diagnosis and showed that the knowledge of these markers changes the diagnosis in 10 % of cases [13]. Our aim was to determine, in a large multicentric prospective observational cohort using the net reclassification improvement (NRI) method, the influence of CSF AD biomarker results on national daily practice. Methods Study subjects This study was designed to investigate how CSF biomarker results might impact the diagnosis of experienced clinicians in their routine daily approach. All recruiting centers are either secondary or tertiary memory centers, as detailed in results. For diagnosis, these centers use the same clinical

3 and biochemical procedures and international validated criteria for AD [19] and for all other dementia. All patients had a thorough examination, including clinical and neuropsychological evaluations and brain imaging. AsrecommendedbyHauteAutoritéde Santé (French Health Authority) in France, CSF biomarkers can be used in clinical practice in case of atypical clinical presentation and/or rapidly progressive cognitive decline and/or in case of diagnostic uncertainty, particularly in young patients. Consequently, in France, CSF AD biomarkers are used in clinical practice mainly in tertiary memory centers. For this work, the participation of centers was on a voluntary basis. When a clinician considers a patient eligible for CSF biomarkers, he fills out a questionnaire about the patient in which all information is anonymous. This questionnaire was divided in two parts First part Before the LP, clinicians were asked to provide the patient s sex and age. Clinicians were required to tick boxes indicating (1) the time when CSF biomarkers were obtained (2) the other explorations that were performed before LP and, if applicable, (3) which exams were done to aid their diagnostic. Diagnosis hypotheses of dementia were proposed as followed: AD, fronto-temporal lobar dementia, vascular dementia, semantic dementia, primary progressive aphasia, Parkinson s disease, dementia due to toxins, prion disease, cortico-basal dementia, supranuclear palsy, Lewy body disease, psychiatric disorders or others. Clinicians could choose several hypotheses, but finally they were required to write their most suspected diagnosis (initial diagnosis). Clinicians were asked to rate the level of confidence of their initial diagnosis with a numerical visual scale ranging from 0 to 10. Diagnosis hypothesis were establish by the clinician in charge of the patient. The clinician was blind to CSF results during this first part. Second part After LP and CSF biomarkers results, clinicians were asked to fill AD CSF biomarkers results as a biological AD profile or as not indicative of AD biological profile or non-contributory (meaning CSF with technical problem, intermediate CSF or CSF without interpretation). Clinicians had to write their final diagnosis, and to rate the level of confidence with a similar numerical visual scale. CSF biomarker tests were performed for all subjects in local laboratories. All biochemical teams are involved in two national cohorts and the share same standards for CSF analysis and interpretation [10, 11, 18, 22, 32]. In addition, the laboratories performed two external quality controls: one run by a working group of the Société Française de Biologie Clinique (French Society of Clinical Biology), and the other by the Alzheimer s Association [17]. As our aim was to gather the most reliable and realistic clinicians opinions in routine clinical practice, without interfering in their diagnosis processes; we did not ask how the CSF AD biomarkers were used for interpretation. Statistical analysis Baseline characteristics of the population study are presented overall, and by diagnosis groups (AD versus non- AD, before and after the LP). As CSF AD biomarkers reflected neuropathological AD lesions, we classified patients into the AD category whenever the Alzheimer box was checked and into the non-ad category when the AD box was not checked. These groups were compared using chi-square statistic for categorical variables and analysis of variance for continuous variables. To evaluate the impact of the initial diagnosis and the results of CSF biomarkers on the final diagnosis, we first computerized C-statistics using logistic regression models. In a first model, the final diagnosis was explained by the initial diagnosis as a dependent variable. In a second model, the final diagnosis was explained by CSF results of the biomarkers. Both models were compared using nonparametric methods. Secondly, we used the net reclassification improvement (NRI) method to compare the two ways of classifications (initial diagnostic, results of CSF biomarkers) on the final diagnosis [21]. Briefly, NRI is a statistical tool proposed to assess improvement in model performance offered by a new marker. The NRI is based on reclassification tables constructed separately for participants with and without the interest event (i.e., final diagnosis of AD or non-ad), and quantifies the correct movement in categories, upwards for events and downwards for non-events, according to CSF biomarkers results versus the initial clinical diagnostic. Using a t test for age and z test for sex ratio, we evaluate the external validity with a comparison to the BNA data. We evaluate the external validity by comparing with data of the National Bank of Alzheimer DGOS CHU Nice (BNA DGOS CHU Nice) that record each consultation performed in a memory consultation in France with or without biomarkers. P-values were two-tailed and values B0.05 considered as statistically significant. Analyses were performed using SAS 9.2 (SAS Institute, Cary, NC, USA). Results The participating clinicians from 29 memory clinics (12 secondary centers and 17 tertiary centers) were divided upon 43.3 % (n = 61) neurologists, 46.1 % (n = 65)

4 Table 1 Characteristics of the study population Table 2 Characteristics according to final diagnosis Overall Clinical diagnosis before LP Non-AD AD p value Final diagnosis after LP Non-AD AD p value N (62.0) 213 (38.0) Age, years, mean 68.6 (9.7) 68.4 (9.8) 68.9 (9.5) 0.53 (SD) Women, n (%) 289 (52.4) 183 (53.4) 106 (50.7) 0.55 CSF biomarkers 262 (46.7) 122 (35.1) 140 (65.7) \0.001 positivity, n (%) Final diagnosis of AD, n (%) 253 (45.1) 100 (28.7) 153 (71.8) \0.001 geriatricians and 1.4 % (n = 2) psychiatrists. A total of 604 questionnaires about patients who underwent LP were consecutively collected between March 2012 and December A total of 43 questionnaires (7.1 %) were excluded because of unknown final diagnosis or non-contributive CSF results and a total of 561 patients were finally included in the study. Characteristics of the study population before CSF biomarker results are presented in Table 1. The population was comparable to the national population recorded in BNA in terms of sex (p = 0.87) but not for age (p = 0.03) patients were younger in our study. Subjects with an AD diagnosis as initial diagnosis before CSF results were not different in terms of sex and age from non-ad patients. The mean age of the subjects was 68.6 years with 289 (52.4 %) women. CSF biomarker analysis was performed at the end of the first consultation in 37.3 % of patients (n = 225), at the end of the second consultation in 27 % (n = 163), after more than 3 months follow-up in 4.5 % (n = 27), after more than 6 months follow-up in 7.6 % (n = 46) or after 1 year follow-up in 15.4 % (n = 93). Most of the patients (n = 554, 91.7 %) were explored by other exams before LP including neuropsychological or language tests in 88 % (n = 494), brain MRI/CT scan in 94 % (n = 530), FDG PET scan in 23.8 % (n = 134), previous LP in 0.01 % (n = 7) or other explorations in 0.05 % (n = 30). Among all patients, 38 % (n = 230) were classified with AD as initial diagnosis. Pursuant to the numerical scale, the mean clinician s certainty of their initial diagnosis was 6.3 ± 1.72 for all diagnosis. When AD was ticked the mean confidence was 6.2 ± 1.62 and 6.5 ± 1.92 when AD was not considered as an hypothesis. The mean confidence was similar for AD and non-ad patients. Characteristics of the study population after CSF results, are presented in Table 2. Subjects with final AD diagnosis were different in terms of age with a trend for sex difference from non-ad patients. A total of 262 (43.37 %) had N Age, years, mean (SD) 67.3 (10.0) 70.3 (9.0) \0.001 Women, n (%) 148 (48.7) 141 (56.9) 0.06 CSF biomarkers positivity, 32 (10.4) 230 (90.9) \0.001 n (%) Initial suspicion of AD, n (%) 60 (19.5) 153 (60.5) \0.001 an AD and 299 (49.5 %) had a non-ad CSF profile with no difference according to age and sex. The final diagnosis that corresponds to the clinician s final conclusion after all explorations were done. Accordingly, patients were classified as AD (n = 253) and non-ad (n = 308). Among them, 153 (60.4 %) of the AD group had the same diagnosis prior to CSF results, while 100 patients did not have this initial diagnosis. Among these 100 patients, clinicians conclude to AD for 93 patients, meaning that only in seven cases did clinicians keep another diagnosis even if CSF biomarkers were in favour of AD. Among patients with non-ad as final diagnosis (non- AD group), 248 (69.2 %) had the same diagnosis prior to CSF results while 60 patients had a AD diagnosis. Among the 248 patients, 29 patients had an AD biological profile but clinicians kept their non-ad initial diagnosis. In contrast, among the 60 patients with AD as the initial diagnosis, 57 did not have AD biological profile and were reclassified according to CSF biomarker results (Fig. 1). Overall, CSF results were concordant with the initial diagnosis before CSF biomarkers in 65.2 % of cases. There was no difference in this percentage between AD and non- AD groups before the LP. In 34.8 % of patients (n = 195), clinical diagnosis and CSF results were discordant. In this group, clinicians change their initial diagnosis for a diagnosis in agreement with CSF profile in 76.9 % (n = 150) (Figs. 1, 2). When CSF biomarkers were in favour of AD, the final diagnosis was AD in 87 % of cases. On the other hand, when CSF results were not in favour, clinicians concluded to non-ad pathology in 92 %. Overall, the rate of changed diagnosis was 26.7 % (150/561). Figure 2 shows the reclassification according CSF profile. Reclassification occurred predominantly in non-ad final diagnosis when clinician s initial diagnosis was AD but CSF biomarkers results were not in favor of their diagnosis. According to the numerical scale, the mean clinician s certainty of their final diagnosis was for all diagnosis 8.1 ± 1.58 significantly higher than before LP (p \ ) (Fig. 2). When the final diagnosis was AD, the mean confidence was 8.13 ± 1.57 and 8.04 ± 1.54

5 Fig. 1 Flow chart illustrating the repartition of patients before and after the results of CSF AD biomarkers increased by 22.1 % for the total population with 22.9 % in the AD group and 20.5 % in the non-ad group. Discussion Fig. 2 Shows that providing a CSF biomarker analysis lead to an important reclassification of AD and non-ad diagnosis. Patients represented in the blue boxes did not change classification with the help of CSF biomarker results. Red boxes indicate the number of patients who were reclassified despite CSF biomarker results; green boxes indicate patients who were reclassified accordingly with CSF biomarker results. The NRI, which indicates overall reclassification was 39.5 % when non-ad was the final diagnosis. There was no difference between the two groups in term of confidence. Between the initial and the final diagnosis, the confidence In this first large, multicentric, prospective study, we showed that clinicians tend to heavily rely on the CSF AD biomarkers when clinical symptoms and biomarkers results are discordant. The use of these biomarkers improves the confidence in the final diagnosis. These results illustrate the impact of these markers in routine clinical practice. Overall, the clinical diagnosis and CSF biomarker results are concordant in about 2/3 of the cases, a finding close to neuropathological studies [16, 27]. In contrast to neuropathological studies that reclassify the diagnosis after several years of evolution, we showed that a reclassification performed at the beginning of the follow-up could be beneficial for the patient. This finding underlines the good concurrence between CSF and neuropathological studies [31]. Previously, two studies explored the biomarker impact on final diagnosis: the first was a single-site study with 109 patients that explored the impact of CSF biomarkers in routine practice [13]; the second was a multisite analysis and evaluated the impact of amyloid imaging on diagnosis assessing two equals groups of patients [12]. In these two studies and our data, all diagnoses have been made by experienced clinicians and show a significant increased confidence in diagnosis confirming the role of biomarker for the clinician in the management of the patient [13]. The rate of changed diagnosis was 10 % in the Keister s work, 54.6 % in the amyloid study, and 26.7 % in this study. This important difference could be explained by the

6 time of the biomarkers evaluation, the profile of patient and explorations made before the lumbar puncture (LP). In the PET amyloid study [12], all included patients had an uncertain clinical diagnostic, but half of patients did not perform all usual neuropsychological and imaging examinations before the LP. It seems more difficult to estimate an accurate diagnostic before all usual explorations. In the present study, patients have had accomplished set of usual explorations, and the CSF biomarker assessment evaluation was not done only because of uncertainty but according to the French Authority recommendations (detailed above) that encompass more clinical situations. As AD diagnosis is a clinical diagnosis in daily practice, it is not surprising that diagnosis changed more often in the florbetapir study for clinicians did not have all clinical information available. It seems appropriate to consider clinical symptoms as a reference in order to assess the real impact of CSF evaluation in routine practice. Furthermore, florbetapir can only show amyloid pathology that is only one aspect of AD pathology and is not sufficient to confirm diagnosis of AD. CSF results incorporate amyloid and tau results that are needful two major features to diagnose AD. Unfortunately, we cannot analyse the difference with the Keister study because the time of biomarker and decision criteria for performing CSF biomarker were not mentioned [13]. In addition, the multicentric aspect of our data is a reliable way to validate the rate of change compared to a single-site study. We used the NRI that analyse more precisely the reclassification in each group (AD versus non-ad) after CSF results to further analyse the data. These method include concordant and discordant results and whether or not there is a change in each group and in both directions (from AD to non-ad and inversely) [12]. This kind of analysis is more complete and precise (than the rate of changed diagnosis) to evaluate the impact of a biomarker. Other studies did not use the NRI evaluation. To evaluate the external validity of this work, we compared demographical data with data from the National Bank Alzheimer that records all information concerning patients from French memory clinics [15]. We showed that in the study, patients were significantly younger and more frequently diagnosed with AD. These findings are linked to the criteria defined by the French authority for performing CSF biomarker dosage recommended in younger patients or with atypical presentation. However, LP and CSF biomarkers are not feasible and helpful in all patients, suggesting that it is difficult to get comparable groups AD patients in routine clinical practice. Furthermore, CSF AD biomarkers are more discriminant in younger than in older AD patients, reinforcing the tendency to assess younger individuals by clinicians [9]. As in the present study, the prospective design with local clinical and biochemical interpretation made possible a realistic estimation of the effect of CSF biomarker results on the final diagnosis of patients with cognitive decline. Some limitations have to be considered. First, as CSF AD biomarkers appear able to decipher for AD pathology and reliable diagnosis, we consider as valid AD diagnosis each time a clinician checked the AD box. Consequently, we have included mixed dementia (AD? vascular or AD? DLB, etc.) in this group. However, AD diagnosis is the most frequent cause of dementia in all epidemiological studies, and this procedure does not seem introduce a major bias. Second, recruiting questionnaire on voluntary basis may create a bias in the selection of centres because clinicians who participated in our study usually integrated LP in their clinical practice. The reasons for not performing CSF biomarker assessment in a memory center could be technical and/or linked to the type of recruitment. This limitation is difficult to avoid because clinical habits and recruitment differ in various memory centers. Furthermore, as in daily practice, because of refusal or contra indications, LP was not performed in all patients among the involved centers. Thirdly, we did not get postmortem brain autopsies to confirm CSF findings. Previous studies have indicated a good correlation between CSF results and neuropathology, but we cannot presently conclude on the diagnostic value of CSF AD biomarkers in clinical practice [6, 27, 30, 31]. Further studies are necessary to explore it in daily practice. These future studies are fundamental because the use of CSF analysis and the improvement of AD diagnosis will be important in the future when appropriate effective therapies will become available and will be routinely proposed to the patients. Acknowledgments We thank all the patients who were involved in this study and their physicians that accepted to fill in questionnaires that constitute the ISAC network: F. Blanc, MD; B. Crétin MD (University Hospital of Strasbourg, Co-Investigator); L. Volpegillot, MD (Saint-Joseph Hospital, Paris, Co-Investigator); L. Bernard- Bourzeix, MD (University Hospital of Limoges, Co-Investigator); M.N. Cuffi, MD (Castres Hospital; Co-Investigator); A.S. Carret- Rebillat, MD (Leopold Bellan Hospital and Inserm U942, Paris, Co- Investigator); L. Cohen, MD (University Hospital of Grenoble Hospital, Co-Investigator); A. Rollin, MD (University Hospital of Lille, Co-Investigator); F. Lebert MD (University Hospital of Lille, Co- Investigator); M. Sauvée, MD (University Hospital of Nancy, Co- Investigator); O. Vercruysse, MD (University Hospital of Lille, Co- Investigator); O. Hanon MD, PhD; ML. Seux, MD (Univerisity Hospital of Broca APHP, Co-Investigator); E. Dionet MD,(University Hospital of Clermont-Ferrand, Co-Investigator); A. Debart, MD (University Hospital of Reims, Co-Investigator), J. Delrieu, MD (University Hospital of Toulouse, Co-Investigator); O.Dereeper, MD (Calais Hospital, Co-Investigator), J. d Avigneau (Cholet, Co-investigator). We also thank Joel Ménard, Patrick Trunet, Hervé Maisonneuve for critical revision of the manuscript and statistical advices. We thank Amel Rouis and Roland Chevrier for providing statistical data and analysis from BNA. We thank DIU plan Alzheimer for critical revision and advices.

7 Conflicts of interest The study has no sponsor. F. Mouton-Liger, reports no disclosures. D. Wallon received travel expense from Novartis. A.C. Troussiere, Yatimi report no disclosures. J. Dumurgier received travel expense from Novartis. E. Magnin, V. de la Sayette, E. Duron, N. Philippi, E. Beaufils, A. Gabelle, B. Croisile, P. Robert, D. Hannequin report no disclosure. F. Pasquier has received consultancy honorarium from ESAI. Pr. Hugon has received consultancy honorarium from Novartis, Lundbeck. C. Paquet has received travel expense from Novartis. References 1. Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, Gamst A, Holtzman DM, Jagust WJ, Petersen RC, Snyder PJ, Carrillo MC, Thies B, Phelps CH (2011) The diagnosis of mild cognitive impairment 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: Andreasson U, Lautner R, Schott JM, Mattsson N, Hansson O, Herukka SK, Helisalmi S, Ewers M, Hampel H, Wallin A, Minthon L, Hardy J, Blennow K, Zetterberg H (2013) CSF biomarkers for Alzheimer s pathology and the effect size of APOE epsilon4. Mol Psychiatry. doi: /mp Beach TG, Monsell SE, Phillips LE, Kukull W (2012) Accuracy of the clinical diagnosis of Alzheimer disease at National Institute on Aging Alzheimer Disease Centers, J Neuropathol Exp Neurol 71: Bian H, Van Swieten JC, Leight S, Massimo L, Wood E, Forman M, Moore P, de Koning I, Clark CM, Rosso S, Trojanowski J, Lee VM, Grossman M (2008) CSF biomarkers in frontotemporal lobar degeneration with known pathology. Neurology 70: Blennow K, Hampel H, Weiner M, Zetterberg H (2010) Cerebrospinal fluid and plasma biomarkers in Alzheimer disease. Nat Rev Neurol 6: Brunnstrom H, Rawshani N, Zetterberg H, Blennow K, Minthon L, Passant U, Englund E (2010) Cerebrospinal fluid biomarker results in relation to neuropathological dementia diagnoses. Alzheimers Dement 6: De Lepeleire J, Wind AW, Iliffe S, Moniz-Cook ED, Wilcock J, Gonzalez VM, Derksen E, Gianelli MV, Vernooij-Dassen M, Interdem G (2008) The primary care diagnosis of dementia in Europe: an analysis using multidisciplinary, multinational expert groups. Aging Ment Health 12: Derksen E, Vernooij-Dassen M, Gillissen F, Olde Rikkert M, Scheltens P (2006) Impact of diagnostic disclosure in dementia on patients and carers: qualitative case series analysis. Aging Ment Health 10: Dumurgier J, Gabelle A, Vercruysse O, Bombois S, Laplanche JL, Peoc h K, Schraen S, Sablonniere B, Pasquier F, Touchon J, Lehmann S, Hugon J, Paquet C (2013) Exacerbated CSF abnormalities in younger patients with Alzheimer s disease. Neurobiol Dis 54: Dumurgier J, Vercruysse O, Paquet C, Bombois S, Chaulet C, Laplanche JL, Peoc h K, Schraen S, Pasquier F, Touchon J, Hugon J, Lehmann S, Gabelle A (2012) Intersite variability of CSF Alzheimer s disease biomarkers in clinical setting. Alzheimers Dement 9(4): Gabelle A, Dumurgier J, Vercruysse O, Paquet C, Bombois S, Laplanche JL, Peoc h K, Schraen S, Buee L, Pasquier F, Hugon J, Touchon J, Lehmann S (2013) Impact of the French Alzheimer plan on the use of cerebrospinal fluid biomarkers in research memory center: the PLM study. J Alzheimers Dis 34: Grundman M, Pontecorvo MJ, Salloway SP, Doraiswamy PM, Fleisher AS, Sadowsky CH, Nair AK, Siderowf A, Lu M, Arora AK, Agbulos A, Flitter ML, Krautkramer MJ, Skovronsky DM, Mintun MA, Group AS (2013) Potential impact of amyloid imaging on diagnosis and intended management in patients with progressive cognitive decline. Alzheimer Dis Assoc Disord 27: Kester MI, Boelaarts L, Bouwman FH, Vogels RL, Groot ER, van Elk EJ, Blankenstein MA, van der Flier WM, Scheltens P (2010) Diagnostic impact of CSF biomarkers in a local hospital memory clinic. Dement Geriatr Cogn Disord 29: Knopman DS, DeKosky ST, Cummings JL, Chui H, Corey- Bloom J, Relkin N, Small GW, Miller B, Stevens JC (2001) Practice parameter: diagnosis of dementia (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 56: Le Duff F, Develay AE, Quetel J, Lafay P, Schuck S, Pradier C, Robert P, French National Alzheimer d (2012) The French Alzheimer plan: description of the national Alzheimer information system. J Alzheimers Dis 29: Lim A, Tsuang D, Kukull W, Nochlin D, Leverenz J, McCormick W, Bowen J, Teri L, Thompson J, Peskind ER, Raskind M, Larson EB (1999) Clinico-neuropathological correlation of Alzheimer s disease in a community-based case series. J Am Geriatr Soc 47: Mattsson N, Andreasson U, Persson S, Arai H, Batish SD, Bernardini S, Bocchio-Chiavetto L, Blankenstein MA, Carrillo MC, Chalbot S, Coart E, Chiasserini D, Cutler N, Dahlfors G, Duller S, Fagan AM, Forlenza O, Frisoni GB, Galasko D, Galimberti D, Hampel H, Handberg A, Heneka MT, Herskovits AZ, Herukka SK, Holtzman DM, Humpel C, Hyman BT, Iqbal K, Jucker M, Kaeser SA, Kaiser E, Kapaki E, Kidd D, Klivenyi P, Knudsen CS, Kummer MP, Lui J, Llado A, Lewczuk P, Li QX, Martins R, Masters C, McAuliffe J, Mercken M, Moghekar A, Molinuevo JL, Montine TJ, Nowatzke W, O Brien R, Otto M, Paraskevas GP, Parnetti L, Petersen RC, Prvulovic D, de Reus HP, Rissman RA, Scarpini E, Stefani A, Soininen H, Schroder J, Shaw LM, Skinningsrud A, Skrogstad B, Spreer A, Talib L, Teunissen C, Trojanowski JQ, Tumani H, Umek RM, Van Broeck B, Vanderstichele H, Vecsei L, Verbeek MM, Windisch M, Zhang J, Zetterberg H, Blennow K (2011) The Alzheimer s association external quality control program for cerebrospinal fluid biomarkers. Alzheimers Dement 7( ):e Mattsson N, Zetterberg H, Blennow K (2010) Lessons from multicenter studies on CSF biomarkers for Alzheimer s disease. Int J Alzheimers Dis. doi: /2010/ McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM (1984) 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 34: McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR Jr, Kawas CH, Klunk WE, Koroshetz WJ, Manly JJ, Mayeux R, Mohs RC, Morris JC, Rossor MN, Scheltens P, Carrillo MC, Thies B, Weintraub S, Phelps CH (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: Moons KG, de Groot JA, Linnet K, Reitsma JB, Bossuyt PM (2012) Quantifying the added value of a diagnostic test or marker. Clin Chem 58: Paquet C, Latour F, Saulnier I, Hanon O (2012) Multicenter study on lumbar puncture indication, clinical practice and feasibility. Rev Neurol (Paris) 168:28 32

8 23. Petersen RC, Aisen PS, Beckett LA, Donohue MC, Gamst AC, Harvey DJ, Jack CR Jr, Jagust WJ, Shaw LM, Toga AW, Trojanowski JQ, Weiner MW (2010) Alzheimer s disease neuroimaging initiative (ADNI): clinical characterization. Neurology 74: Ranginwala NA, Hynan LS, Weiner MF, White CL 3rd (2008) Clinical criteria for the diagnosis of Alzheimer disease: still good after all these years. Am J Geriatr Psychiatry 16: Rinne JO, Brooks DJ, Rossor MN, Fox NC, Bullock R, Klunk WE, Mathis CA, Blennow K, Barakos J, Okello AA, Martinez Rodriguez, de Liano S, Liu E, Koller M, Gregg KM, Schenk D, Black R, Grundman M (2010) 11C-PiB PET assessment of change in fibrillar amyloid-beta load in patients with Alzheimer s disease treated with bapineuzumab: a phase 2, double-blind, placebo-controlled, ascending-dose study. Lancet Neurol 9: Santos AN, Ewers M, Minthon L, Simm A, Silber RE, Blennow K, Prvulovic D, Hansson O, Hampel H (2012) Amyloid-beta oligomers in cerebrospinal fluid are associated with cognitive decline in patients with Alzheimer s disease. J Alzheimers Dis 29: Schoonenboom NS, Reesink FE, Verwey NA, Kester MI, Teunissen CE, van de Ven PM, Pijnenburg YA, Blankenstein MA, Rozemuller AJ, Scheltens P, van der Flier WM (2012) Cerebrospinal fluid markers for differential dementia diagnosis in a large memory clinic cohort. Neurology 78: Smits LL, Pijnenburg YA, Koedam EL, van der Vlies AE, Reuling IE, Koene T, Teunissen CE, Scheltens P, van der Flier WM (2012) Early onset Alzheimer s disease is associated with a distinct neuropsychological profile. J Alzheimers Dis 30: Sperling RA, Aisen PS, Beckett LA, Bennett DA, Craft S, Fagan AM, Iwatsubo T, Jack CR Jr, Kaye J, Montine TJ, Park DC, Reiman EM, Rowe CC, Siemers E, Stern Y, Yaffe K, Carrillo MC, Thies B, Morrison-Bogorad M, Wagster MV, Phelps CH (2011) Toward defining the preclinical stages of Alzheimer s disease: recommendations from the National Institute on Aging- Alzheimer s Association workgroups on diagnostic guidelines for Alzheimer s disease. Alzheimers Dement 7: Strozyk D, Blennow K, White LR, Launer LJ (2003) CSF Abeta 42 levels correlate with amyloid-neuropathology in a populationbased autopsy study. Neurology 60: Tapiola T, Alafuzoff I, Herukka SK, Parkkinen L, Hartikainen P, Soininen H, Pirttila T (2009) Cerebrospinal fluid (beta)-amyloid 42 and tau proteins as biomarkers of Alzheimer-type pathologic changes in the brain. Arch Neurol 66: Vanderstichele HM, Shaw L, Vandijck M, Jeromin A, Zetterberg H, Blennow K, Teunissen C, Engelborghs S (2013) Alzheimer disease biomarker testing in cerebrospinal fluid: a method to harmonize assay platforms in the absence of an absolute reference standard. Clin Chem 59(4): Warren JD, Fletcher PD, Golden HL (2012) The paradox of syndromic diversity in Alzheimer disease. Nat Rev Neurol 8: