The memory associative test of the district of Seine-Saint-Denis (tma-93) validation as Alzheimer's disease diagnosis test

DOCTORAL THESIS THE MEMORY ASSOCIATIVE TEST OF THE DISTRICT OF SEINESAINT-DENIS (TMA-93) VALIDATION AS ALZHEIMER’S DISEASE DIAGNOSIS TEST Silvia Rodrigo Herrero Director Emilio Franco Macías Tutor Miriam Echevarría Irusta University of Sevilla, 2021 INDEX ABBREVIATIONS ........................................................................................... 1 INTRODUCTION ............................................................................................. 4 1. ALZHEIMER’S DISEASE. HISTORY’S OVERVIEW ......................................................... 5 2. EPIDEMIOLOGY AND CURRENT SITUATION IN OUR AREA ............................................ 7 3. CLINICAL DIAGNOSIS .................................................................................................... 9 3.1 Mild Cognitive Impairment ................................................................................................ 11 3.2 Diagnosis Criteria ............................................................................................................... 12 4. IMAGE BIOMARKERS .................................................................................................. 15 4.1 Pathophysiological image biomarkers ............................................................................... 15 4.1.1 AMYLOID-PET .............................................................................................................. 15 4.1.2 TAU-PET ....................................................................................................................... 16 4.2 Topographic image biomarkers .......................................................................................... 17 4.2.1 MAGNETIC RESONANCE IMAGING (MRI) ...................................................................... 17 4.2.3 18FDG-PET .................................................................................................................... 18 5. BIOMARKERS IN BIOLOGICAL LIQUIDS ..................................................................... 19 5.1 Cerebrospinal Fluid Biomarkers ........................................................................................ 19 5.2 Plasma Biomarkers ............................................................................................................. 20 6. MEMORY EVALUATION IN OUR AREA ......................................................................... 22 6.1 Screening tests .................................................................................................................... 23 6.1.1 INFORMANT QUESTIONNAIRE ON COGNITIVE DECLINE IN THE ELDERLY (IQCODE) .. 25 6.1.2 THE AD8: “EIGHT-ITEM INTERVIEW TO DIFFERENTIATE AGING AND DEMENTIA” ........ 26 6.1.3 MINI-MENTAL STATE EXAM (MMSE) .......................................................................... 27 6.1.4 MINI-COG ...................................................................................................................... 29 6.1.5 COGNITIVE MINI-EXAM (MEC) .................................................................................... 29 6.1.6 MEMORY IMPAIRMENT SCREEN (MIS) .......................................................................... 30 6.1.7 CLOCK DRAWING TEST (CDT) ...................................................................................... 31 6.1.8 SEVEN MINUTES SCREEN .............................................................................................. 32 6.1.9 EUROTEST ...................................................................................................................... 33 6.1.10 SHORT PORTABLE MENTAL STATUS QUESTIONNAIRE (SPMSQ) ............................... 34 6.1.11 PHOTOTEST .................................................................................................................. 35 6.1.12 MEMORY ALTERATION TEST (M@T) .......................................................................... 36 6.2 Specific memory tests ....................................................................................................... 36 6.2.1 VERBAL MEMORY TESTS BY WORD LISTS ...................................................................... 36 6.2.2 LOGICAL MEMORY TESTS .............................................................................................. 38 6.2.3 MEMORY EVALUATION BASED ON SEMANTIC CATEGORIES AND CUED RECALL. FREE AND CUED SELECTIVE REMINDING TEST (FCSRT) ....................................................... 40 6.2.4 VERBAL BINDING TESTS ................................................................................................ 42 6.2.5 VISUAL MEMORY TESTS ................................................................................................ 43 7. DEVELOPMENT OF A DIAGNOSTIC TEST .................................................................... 48 7.1 Definition of the test ........................................................................................................... 48 7.2 Validation studies ............................................................................................................... 48 7.3 Reliability studies ............................................................................................................... 50 7.4 Normative studies ............................................................................................................... 51 7.5 Applicability studies ........................................................................................................... 51 FUNDAMENTS .............................................................................................. 54 RESEARCH JUSTIFICATION ............................................................................................. 56 HYPOTHESIS ................................................................................................. 59 OBJECTIVES ................................................................................................. 62 METHODS AND RESULTS ......................................................................... 65 CHAPTER I. ARTICLE I. ................................................................................................. 68 CHAPTER II. ARTICLE II. ............................................................................................... 75 CHAPTER III. ARTICLE III. ............................................................................................ 85 CHAPTER IV. ARTICLE VI. ............................................................................................ 93 DISCUSSION ................................................................................................ 103 THE CHOSEN TEST: TMA-93 ........................................................................................ 105 CHAPTER I. ARTICLE I. TMA-93 FOR DIAGNOSING AMNESTIC MILD COGNITIVE IMPAIRMENT: A COMPARISON WITH THE FREE AND CUED SELECTIVE REMINDING TEST ............................ 106 CHAPTER II. ARTICLE II. RELIABILITY AND FEASIBILITY OF THE MEMORY ASSOCIATIVE TEST TMA-93 .......................................................................................................................... 108 CHAPTER III. ARTICLE III. NORMS FOR TESTING VISUAL BINDING USING THE MEMORY ASSOCIATIVE TEST (TMA-93) IN OLDER EDUCATIONALLY-DIVERSE ADULTS ..................... 110 CHAPTER IV. ARTICLE VI. TMA-93 VALIDATION BY AD BIOMARKERS. A COMPARISON WITH THE FCSRT ON A BIOBANK SAMPLE ......................................................................... 112 CONCLUSIONS ........................................................................................... 116 REFERENCES .............................................................................................. 120 1 ABBREVIATIONS AA: Alzheimer's Association Aβ42: Amyloid-beta 42 AD: Alzheimer’s disease AD8: Eight-item interview to differentiate aging and dementia ADL: Activities of Daily Living APA: American Psychiatric Association AUC: Area Under the Curve aMCI: Amnestic Mild Cognitive Impairment CERAD: Consortium to Establish a Registry for Alzheimer's Disease CDT: Clock Drawing Test CSF: Cerebrospinal Fluid CVLT: California Verbal Learning Test DMS: Diagnostic and Statistical Manual of Mental Disorders DMS-48: Delayed Matching-to-Sample Task 48 FCSRT: Free and Cued Selective Reminding Test FDG: Fluosdeoxyglucose GDS: Global Dementia Scale GP: General Practitioner 2 HCs: Healthy Controls IDDD: Interview for Deterioration in Daily Living Activities in Dementia IQCODE: Informant Questionnaire on Cognitive Decline in the Elderly IQR: Interquartile Range IWG: International Working Group MCI: Mild Cognitive Impairment MEC: Mini-Exam Cognoscitivo MIS: Memory Impairment Screen MMSE: Mini-Mental State Exam MoCA: Montreal Cognitive Assessment MRI: Magnetic Resonance Imaging MTL: Medial Temporal Lobe NFL: Light Neurophilically Protein NIA: National Institute of Aging NINCDS-ADRDA: National Institute of Neurological and Communicative Disordes and Stroke- Alzheimer NP1: Neuronal Pentraxin 1 NPV: Negative Predictive Value PET: Positron Emission Tomography 3 PPV: Positive Predictive Value P-Tau: Phosphorylated Tau RAVLT: Rey Auditive Verbal Learning Test SMC: Subjective Memory Complaints SPMSQ: Short Portable Mental Status Questionnaire STARD: Standards for Diagnostic Accuracy Reporting T@M: Memory Alteration Test TAVEC: Verbal Learning Test Spain- Complutense TMA-93: Memory Associative Test of the District of Seine-Saint-Denis T-Tau: Total Tau WHO: World Health Organization WMS: Wechsler Memory Scale Introduction Introduction 11 Independence in activities of daily living (ADL) is gradually affected in AD. First, ADL impaired are instrumental and complex activities, such as interpersonal and social functioning, being the self-maintenance the last involved [44]. 3.1 Mild Cognitive Impairment The most frequently applied concept referring to a state between healthy aging and dementia is Mild Cognitive Impairment (MCI). It refers to a transitional state between "normality" and dementia, including individuals with a higher risk situation to develop dementia, especially AD. Although MCI is a widely accepted term, the clinical separation between this MCI and established dementia is uncertain [45-46]. Neuropsychological examination in MCI patients involves deficits in any cognitive domain, compared with expected results by age and educational level, while the patient continues functionally independent [47-48]. However, MCI is not always due to AD. It can be secondary to other causes of dementia or non-neurodegenerative diseases. It is classified as amnesic/non-amnesic and single/multidomain [49]. Amnestic MCI (aMCI) with multidomain involvement is the most likely that evolves into an established AD. Saykin et al. [50] demonstrated that both patients with MCI and those who present cognitive complaints without evident alteration in neuropsychological tests (Subjective Memory Complaints) present a regional loss of density in the gray matter. This pattern suggests a continuum between "normal aging" and MCI. It also leaves the door wide open to the early diagnosis of cognitive decline these densities’ measurement seems more sensitive than volumetric loss or cognitive changes. For the definitive diagnosis of an MCI due to AD (prodromal AD) is mandatory the analysis of pathophysiological biomarkers of the disease (see section 4) because, Introduction 12 although the aMCI is characteristic of AD, in our environment, only 62% of these patients have positive CSF biomarkers for AD [49]. 3.2 Diagnosis criteria Until 2007, the most used diagnostic criteria were those of the NINCDSADRDA group, which established that AD could only be definitively diagnosed retrospectively, after the post-mortem confirmation [22]. With the advances in image techniques and the introduction of biomarkers, the International Working Group (IWG) developed in 2007 new criteria [51], revised again in 2014 (Table 1) [52]. Those criteria combine the episodic memory decline with the positivity of biomarkers and support preclinical and prodromal AD diagnosis. By 2011, the National Institute of Aging, together with the Alzheimer's Association (NIA-AA), published a new update, including the indication of the use of the different biomarkers available (Table 2) [53]. After these updates, Jack et al. (2018) published a new conceptual framework for the investigation of AD [54], in which the disease is defined strictly biologically, based on the biomarker profile, without taking into account the patient's phenotype or the clinical stage (Table 3). The great novelty of this design is its basis in the A / T / N classification, being (A) cerebral amyloidosis, (T) Tau pathology, and (N) Neurodegeneration. According to this classification, all patients with cerebral amyloidosis, regardless of the presence or not of Tau or Neurodegeneration, will be classified within the "Alzheimer's disease continuum." Although this last update is not yet part of the usual clinical practice, it is the one that currently prevails in the investigation's field [54]. Introduction 13 Table 1. Diagnostic criteria for AD. International Working Group Adapted from Dubois et al. 2014. Lancet Neurol [52] AD: Alzheimer’s disease. Introduction 14 Diagnostic category Biomarker Probability of AD etiology Aβ (PET or CSF) Neuronal injury (CSFtau,FDG-PET,MRI) PROBABLE AD dementia • Clinical criteria Uninformative Unavailable, conflicting, or indeterminate Unavailable, conflicting, indeterminate • 3 levels of evidence of AD pathophysiological process Intermediate Unavailable or indeterminate Positive Intermediate Positive Unavailable or indeterminate High Positive Positive Possible AD (atypical presentation) • Based on clinical criteria Uninformative Unavailable, conflicting, or indeterminate Unavailable, conflicting, indeterminate • With evidence of AD pathophysiological process High but does not rule out second etiology Positive Positive • Dementia-unlikely due to AD Lowest Negative Negative Table 2. AD criteria. National Institute on Aging/Alzheimer’s Association, 2011. Adapted from McKhann et al. [53]. AD (Alzheimer’s disease, Aβ Amyloid-beta, PET: positron emission tomography. CSF: Cerebrospinal fluid. FDG: 18fluorodeoxyglucose. MRI: magnetic resonance imaging. Cog. Unimpaired MCI Dementia Biomarker Profile A- T- (N)- Normal AD biomarkers. Cognitively unimpaired Normal AD biomarkers with MCI Normal AD biomarkers with dementia A+ T- (N)- Preclinical Alzh pathologic change Alzh pathologic change with MCI Alzh pathologic change with dementia A+ T+ (N)- Preclinical AD MCI due to AD (Prodromal AD) AD with dementia A+ T+ (N)+ A+ T- (N)+ Alzh and concomitant suspected Non-Alzh pathologic change, cognitively unimpaired Alzh and concomitant suspected Non-Alzh pathologic change with MCI Alzh and concomitant suspected Non-Alzh pathologic change with dementia A- T+ (N)- Non-Alzh pathologic change,cognitively unimpaired Non-Alzh pathologic change with MCI Non-Alzh pathologic change with dementia A- T- (N)+ A- T+ (N)+ Table 3. Cognitive syndrome combined with AD biomarkers. Adapted from Jack et al. (2018) [54]. Cog: Cognitive. MCI: Mild Cognitive Impairment. AD: Alzheimer’s disease. Alzh: Alzheimer Introduction 15 4. IMAGE BIOMARKERS Biomarkers allow in vivo studies of structural and molecular changes related to AD pathophysiologic process, increasing the diagnosis accuracy. Its inclusion in the diagnosis criteria [52-53] has led to an expansion in its use in routine clinical practice [49]. Biomarkers may be classified as pathophysiological and topographic. Pathophysiological ones allow the etiological substrate's identification, while topographic ones are not specific but better define the disease's severity and progression. [55]. 4.1 Pathophysiological image biomarkers 4.1.1 Amyloid PET Amyloid-beta tracers cross the blood-brain barrier and bind to amyloid plaques with high affinity (specifically insoluble fibrillary forms of Amyloid-beta 40 and 42), so they accurately mark the burden of cerebral fibrillar amyloid [56]. Figure 5 A) Negative 11F-Florbetaben Amyloid-PET. B) Positive 11F- Florbetaben Amyloid-PET A B Introduction 16 Amyloid-PET leads diagnostic utility to exclude the disease, with a high Negative Predictive Value [49]. PET scans performed during life have up to a 98% sensitivity and 80% to 95% specificity for detecting neuritic amyloid plaques at autopsy [57]. Additionally, patients with MCI and positive Amyloid-PET are significantly more likely to convert to established AD during the next three-year follow-up, than those with MCI without Amyloid retention [58]. However, there are no significant differences in Amyloid retention as the patients' clinic declines. Amyloid retention reaches a plateau when patients manifest clinical symptoms, suggesting that the deposition precedes the onset of cognitive decline [59]. With the extended use of Amyloid-PET, the Spanish Neurological Society published indications for Amyloid-PET as a diagnostic tool in our country [49], which mainly includes a progressive cognitive deficit of uncertain etiology, atypical or mixed presentation, or early-onset dementia (under 65 years old). 4.1.2 Tau PET Whole-brain tau PET could be a suitable biomarker for identifying AD-related in vivo regional distribution of tau (Figure 6) [60]. Ossenkoppele et al. (2016) suggested that tau pathology, against amyloid theory, could cause clinical manifestations in AD [61]. Furthermore, the correlation between tau and neuroinflammation has been demonstrated even without significant amyloid pathology, suggesting an Aβ-independent process [62-63]. Introduction 17 Figure 6 A) Normal 18F-AV-1451 PET study illustrating a tau-PET scan from an elderly cognitively normal subject. B) Abnormal 18F-AV-1451 PET study illustrating a tau-PET scan images of a patient with mil Alzheimer’s disease. 4.2 Topographic image biomarkers Topographic biomarkers are not specific but provide a good representation of the disease's severity and progression [55]. 4.2.1 Magnetic Resonance Imaging (MRI) MRI classically played a key role in excluding other causes of dementia, such as vascular or neoplastic etiologies. However, thanks to technological advances and improved image quality, it is currently also employed to rate the hippocampal atrophy. In typical AD, MRI shows MTL, hippocampus, and entorhinal cortex atrophy [64]. Figure 7. T1-weighted MRI image showing AD typical atrophy: medial temporal and frontotemporal cortical. Coronal view. B) A) Introduction 18 4.2.2 18FDG-PET The decrease in fluorodeoxyglucose tracer (18FDG) uptake indicates hypometabolism related to neuronal injury and synaptic dysfunction regarding AD pathology [65]. Apart from the severity rate based on posterior cingular and temporoparietal hypometabolism, 18FDG-PET seems to have an outstanding accuracy in discriminating between Alzheimer's dementia and controls. Hence, a normal 18FDG-PET almost excludes a neurodegenerative disease being the patient's cognitive symptoms cause [66-67]. Figure 8. A) Normal 18FDG-PET metabolism in healthy control. B) Characteristic 18FDG-PET hypometabolism in AD pathology. Introduction 19 5. BIOMARKERS IN BIOLOGICAL LIQUIDS Fluid biomarkers are promising tools for clinical staging and personalized patient monitoring, and for establishing a proper treatment [68]. They have the advantage of being easy to implement in clinical trials and allow the analysis of various biomarkers in the same sample [69]. 5.1 Cerebrospinal Fluid biomarkers CSF amyloid-beta 42 (Aβ42), total tau (t-tau), and phosphorylated tau (p-tau), are the most widely recognized as CSF pathophysiological biomarkers [70-71]. Likewise, they are the most accepted inclusion criteria and outcome measures in clinical trials [72]. The decrease in Aβ42 CSF levels is inversely correlated with the amyloid load on brain plaques [73]. Aβ42 remains relatively stable over time in patients with AD dementia, so it only has a limited utility for progression monitoring [74]. Aβ40 is the predominant amyloid peptide in the brain, CSF, and plasma but does not appear to be as pathogenic as Aβ42 [75]. However, the Aβ42/Aβ40 ratio seems to be a better predictor than CSF Aβ42 alone; comparable to the T-tau/Aβ42 and P-tau/Aβ42 ratios [76]. CSF T-tau and P-tau levels have demonstrated a high validity to differentiate AD from healthy controls and other causes of dementia [77] and predict disease progression [78]. In combination with Aβ42, both T-tau/Aβ42 and P-tau/Aβ42 ratios outperform AD diagnosis utility [79]. Introduction 20 Apart from the etiological markers, it has been recently described the usefulness of other associated processes' biomarkers, such as neurodegeneration, synaptic dysfunction, or inflammation [80-81]. Neurodegeneration biomarkers include Light Neurophilically Protein (LNP) [81-82] and YKL-40, which also measures glial activation [83]. Regarding synaptic dysfunction, we can mention Neurogranin Synaptic Protein, NPTX2, or SNAP25 [81, 84]. Some studies also describe the benefit in measuring the concentration’s increase of chemokines, as CCL23 [85]. 5.2 Plasma biomarkers Since blood is more accessible than CSF, biomarkers measurement for AD diagnosis is preferable in this fluid. However, while there is a free exchange of molecules between the brain and CSF, only a small fraction of the Central Nervous System's proteins enters the bloodstream, and they are easily degraded or metabolized [81]. Recent immunoassay studies suggest that Aβ may be a useful plasma biomarker, despite previous contradictory studies [71]. Aβ42 and Aβ42/Aβ40 plasma ratio are correlated with high Aβ deposition levels in CSF and brain PET, and the Aβ42/Aβ40 ratio is associated with the risk of progression to clinical AD in individuals with subjective memory complaints [86]. Ultrasensitive immunoassay techniques have also demonstrated increased Tauplasma levels in AD compared to healthy controls. However, these differences, although significant, are low to be considered diagnostically useful [81]. Alternatively, T-tau or P- tau measurement in exosome preparations could improve its performance as a blood biomarker [87]. Introduction 27 quick to administer (~ 3 minutes). Cutoff points are normal cognition 0-1; impairment in cognition 2 or greater. A Spanish version of AD8 has been validated in our country. The questionnaire showed high diagnostic accuracy to discriminate between patients with cognitive impairment and healthy controls (AUC 0.90; CI 95%, 0.86-0.93). The best cutoff was 3/4 (Sens. 93%; Spec. 81%) [118]. 6.1.3 Mini-Mental State Exam (MMSE) [103] The Mini-mental State Examination (MMSE) was designed by Folstein in 1975 [103] as an instrument for a brief mental state assessment. Widely disseminated, it is the most frequently cited cognitive test on Medline and the one with the largest language versions. It is the most adapted test for standardized cognitive evaluation in clinical settings, having data for screening, staging, and treatment monitoring [119]. MMSE is a brief test structured in eleven questions, able to administer within 5-10 minutes. It attempts to examine various cognitive functions: orientation, immediate and deferred memory, concentration, language, and visuospatial function [103]. The total score must be adjusted to the demographic characteristics of the individual and study population. The test's Sensitivity to the dementia patients' diagnosis is 77.0%, and the Specificity 91.2%. However, the precision in identifying of MCI versus healthy controls is lower, with a Sensitivity of 63.4%, and a Specificity of 65.4%. Its properties also change depending on the cut-off point selected: for 23/24, the Sensitivity is 85%, and Introduction 28 the Specificity 90%, while for a 24/25 cut-off point, Sensitivity is 87%, and Specificity is 82% [119-120]. The MMSE presents a good agreement between General Practitioners and neuropsychologists in specialized memory units. Although the scores obtained by the GPs are generally lower (15.8 vs. 17.4 for specialized units; p <0.01), the Kappa index of the agreement is 0.86 [121]. In Spain, we have several validations and normative studies for the different versions. For its original one, according to Blesa et al. (2001) [122], Cronbach's alpha coefficient was 0.94. For an estimated cut-off point in MCI patients of 24/25 Sensitivity is 87.32%, and Specificity 89.19%. The psychometric test properties vary depending on the population prevalence and the selected cut-off point [119-120]. This statement has also been demonstrated in Spain where, for a general Primary Care population, the recommended cut-off point for dementia patients is 22/23 [123], while for specialized memory clinics ("hyper selected" patients), it is preferable 24/25 [124] or 26/27 [125] as optimal cut points. We can conclude that MMSE meets the psychometric requirements of reliability and reproducibility for distinguishing patients with cognitive impairment from healthy controls. However, it is only modestly accurate, turning it more suitable for its use in the Primary Care setting, needed to combine it with other neuropsychological tests for purposes such as differentiation between MCI and AD. Introduction 29 6.1.4 Mini-Cog [104] The Mini-Cog is a short and easy cognitive test, consisting of a three-element verbal memory task and a simplified assessment of the Clock Drawing Test. It was developed as a brief test to discriminate people with dementia from healthy elders in a typical heterogeneous population of a Primary Care setting. According to its authors, it is also free of educational and cultural biases [104]. The standard scoring system involves assigning 0 to 3 points on the word retrieval task for correct retrieval of 0, 1, 2, or 3 words, respectively. The rating of the clock drawing part is as "normal" or "abnormal" (0 vs. 2 points). The Mini-Cog has been validated in Spain [126]. The test showed high diagnostic accuracy for discriminating between patients with cognitive impairment against healthy controls (AUC ± Sensitivity: 0.88 ± 0.01). The instrument was less useful for screening individuals with low education levels (AUC ± Sensitivity: 0.74 ± 0,05). A cut-off point of 2/3 in the Mini-Cog achieved a Sensitivity of 0.90 and a Specificity of 0.71 [104]. 6.1.5 Cognitive Mini-Exam (MEC) [105] Lobo's MEC is a Spanish adapted and validated version of the MMSE [105]. It is a dementia screening test, also useful in their evolutionary monitoring. There are two versions, 30 (MEC-30) and 35 (MEC-35) points. The items are grouped into five sections in both versions that examine orientation, fixation memory, concentration and calculation, delayed recall, language, and construction. The 35-point version adds a 3- point digit item in the "concentration and calculation" section and another 2-point abstraction item in the "language and construction" section. It is more advisable to use the 30-points version. The cut-off point for dementia is usually set at 24 points. Introduction 30 According to Lobo et al. (1980) [105], this is a useful screening test to discriminate between patients with cognitive decline from healthy controls, with test-retest reliability of 0.87. 6.1.6 Memory Impairment Screen (MIS) [106] Memory Impairment Screen (MIS) is a short and easy-to-administer detection test, which through four words (or drawings in its version for illiterates), evaluates free and cued reminding. It uses controlled learning to specify attention, induce specific semantic processing, and modify the coding specification to improve cognitive decline detection [106]. Although its psychometric data varies depending on the cut-off point, they are excellent, with a Sensitivity between 85-90% and Specificity between 90-98%. As it is a memory-based test, its performance is better for AD or mixed dementia than vascular dementia [106]. MIS also has excellent discriminatory data for MCI and AD for its Spanish versions, similar to the English version. However, in our country, most of the authors’ chosen cut-off point is 4, granting a Sensitivity over 90% and Specificity over 80%. It also has an adequate interobserver (0.85) and test-retest (0.81) Specificity [127-129]. A recent study further demonstrates that MIS Sensitivity is better for non-English speakers and less educated people, minimizing cultural bias in ethnically diverse populations [130]. Together with its excellent diagnostic utility values and its brevity Introduction 31 and ease of administration, those properties make MIS a reasonable alternative to other screening methods. 6.1.7 Clock Drawing Test (CDT) [107] The Clock Drawing Test (CDT) is a short and easy to administer screening test. Introduced at the beginning of the 20th century as an apraxia sign for detecting parietal lesions [107], soon after, it also became used as early detection of cognitive decline. The CDT is one of the most widely used brief cognitive tests. However, it has significant disadvantages. In addition to having a discrete diagnostic utility, it does not explicitly explore episodic memory, is not adapted for low-educated patients, and is not provided by a unified correction method [131]. There are more than fifteen validated variants. Some of them comprise the drawing on a blank sheet, from a previous circle, or combining the spontaneous drawing with a copy [132-136]. Each one of these variants has its scoring system [137]. CDT is a short and easy to administer screening test, with excellent internal consistency and reliability among evaluators. However, due to the high number of different scoring systems, its accuracy is still debated. These problems could be solved when combining the CDT with the MMSE, both variants improving its diagnostic utility [138]. We also have validation and normative values for CDT in Spain, although they give similar results to other countries [139]. Concluding, CDT can reliably and accurately differentiate patients with dementia from healthy controls, providing crucial clinical information. However, its Introduction 32 diagnostic utility is discrete in differentiating MCI from Healthy Controls. This limited utility, adding to the fact that it is not suitable for illiterate patients, and does not include any specific memory examination, turns it into a second option as a screening test for a memory outpatient clinic [131]. 6.1.8 Seven Minutes Screen [108] The 7 Minute Screen Test (7MS) was designed by Solomon et al. (1998) [108] as a screening instrument for dementia, especially for the Primary Care setting. Its main contribution is the inclusion in a single instrument of several tests that had previously demonstrated excellent diagnostic performance [108]. The 7MS includes four subtests that examine specific affected areas early affected in AD (temporal orientation, episodic memory, visuospatial and visuoconstructive capacity, and semantic memory). In its Spanish version [140], the tests included are the Benton Temporal Orientation Test, a version of Buschke's Free and Cued Selective Reminding Test (FCSRT), the CDT, and a category fluency test. The Benton Temporal Orientation Test includes five temporal orientation questions and evaluates the error and the deviation degree from the correct answer. The FCSRT variation of the one proposed by Buschke et al. examines episodic memory (free and facilitated recall) after conducting the semantic processing of 16 drawings. By establishing guided learning with a semantic cue, it minimizes the interference due to distraction or anxiety. The Spanish adaptation of the original FCSRT employs drawings rather than words, allowing testing on low-educated patients. The Clock Drawing Test assesses visuospatial and visuoconstructive skills through a Introduction 33 simplified version. Finally, semantic memory is explored using a category fluency test (animals in 1 minute). According to the original study [108], the test is highly sensitive (92%) and specific (96%) on diagnosing dementia, in addition to specifying excellent test-retest reliability (0.91). This high ability to discriminate between AD patients and healthy controls has been confirmed internationally [141] and in its Spanish version [140, 142]. However, it seems that gender, age, and educational level have a higher impact on results than described initially. 7MS is an excellent instrument for dementia diagnostic screening. However, despite being short, it is needed a mean of 12.4 minutes (between 8 and 22) for its complete administration [143], so it is not suitable for a short memory outpatient consult. 6.1.9 Eurotest [109] The Eurotest is an instrument designed in Spain, based on the money knowledge and management, adapted from the Money Test. It is easy to administer, useful, and applicable to illiterate and low-educated patients [109]. Among its psychometric properties, it stands out with a 93% Sensitivity and 87% Specificity on differentiating dementia from Healthy Controls, similar values to MMSE and 7MS, but with less administration time needed. The Eurotest ecological validity is guaranteed by the day-to-day nature of the tasks and materials, and the proper construct validity is guaranteed by the significant correlation between the Eurotest score and the Global Dementia Scale [144]. Introduction 34 Therefore, the Eurotest is a rapid, easy, and useful screening test for dementia in routine clinical practice. It is not influenced by sociodemographic variables such as educational level, which is advantageous over other available screening tests [145- 146]. 6.1.10 Short Portable Mental Status Questionnaire (SPMSQ) [110] The Short Portable Mental Status Pfeiffer (SPMSQ) was developed in 1975 to differentiate organic and functional disorders [110]. Through ten questions and a serial math subtraction, briefly evaluate short and long-term memory, orientation, information on daily events, and executive function. Its main advantage is its easy administration since it does not require any specific tool for its completion, and it applies to low-educated patients. The scoring encompasses four categories: 0-2 (cognitively intact), 3-4 (medium damaged), 5-7 (moderately disabled), and 8-10 (severely disabled) [147]. It is a sensitive and specific screening test for moderate to severe dementia in the community and in-hospital patients. However, its diagnostic utility for the MCI diagnosis is limited [148-149]. For its Spanish adaptation, results on differentiating dementia from Healthy Controls are similar to the original version, with an AUC of 0.892 and at outstanding reliability, interobserver (k = 0.734) and intraobserver (k = 0.925) [150]. The most recommendable cut-off point in our setting is three mistakes, the same recommended by Pfeiffer et al. [110], or 4 in the case of illiterates. [150]. Introduction 35 SPMSQ is a relatively sensitive screening test for dementia both in the community and in pluripathological inpatients. It is easy to administer, even in illiterate or elders. However, its diagnostic utility for MCI is limited, so this evaluation must be completed with other neuropsychological tests or batteries. 6.1.11 Phototest [111] The Phototest, developed in Spain, is a brief and easy-to-administer instrument feasible for illiterates. It evaluates visual recognition and denomination, verbal fluency, and memory, evaluating the cue efficiency [111]. A complete diagnostic validation of this test has been carried out, including phases I [111], II [151], and III [152], in addition to a normative study [153]. Phototest shows good test-retest and interobserver reliability, and cutoff scores of 26/27 and 28/29 points give adequate discriminatory validity for dementia and cognitive decline, respectively [154]. Phototest seems to be more precise and less expensive than MMSE. It has similar diagnostic effectiveness to MIS, being also applicable to illiterate patients [154]. Phototest is a brief and easy-to-administer screening test with good diagnostic accuracy for dementia and cognitive decline. It is influenced by age, gender, and educational level, but it is suitable for illiterates. Those properties make it feasible for the primary care setting and general neurology outpatient clinic. Introduction 36 6.1.12 Memory Alteration Test (M@T) [112] M@T is another screening test developed in Spain, brief, easy-to-administer, and score. It covers tasks evaluating temporal orientation and episodic and semantic memories [112]. It provides useful and valid discrimination between patients with Subjective Memory Complaints (SMC) and aMCI, SMC and mild AD, and between aMCI and mild AD [155-156]. M@T constitutes a short and reliable screening test that could be applicable by GPs in primary care clinics. However, it has not normative values, and it should not be used singly to define dementia, as it only evaluates memory and temporal orientation, not valid to detect atypical AD or other dementias. 6.2 Specific memory tests In our country, there is no standardization in the use of memory tests. Some of the most widely used are word lists, logical memory tests, or those evaluating the semantic cue's efficiency. The choice usually depends on the available time and suspected disease. 6.2.1 Verbal memory tests by word lists Some of the most used in our environment are: “The California Verbal Learning Test” [157], “The Rey Auditive Verbal Learning Test” [158], “Word-List memory subtest of the CERAD” [159], and de Spanish “Verbal Learning Test Spain- Complutense (TAVEC)” [160]. The California Verbal Learning Test (CVLT) is one of the most commonly used tests to assess older adults' verbal episodic memory. It evaluates the free and cued Introduction 43 test format. Questionnaires assessing binding appear to have better sensitivity than the FCSRT, especially in longitudinal studies, thus facilitating the measurement of minor memory decline and the detection of very early AD [185]. One of the most employed verbal binding tests is the Associated Pairs subtest of the Wechsler Scale [167]. This subtest assesses immediate recall with a list of between 10 and 14 “easy” (E.g., North/South) and “difficult” (E.g., School/Cellar) pairs of words presented orally. Assess long-term recall with semantic cues, as well as verbal recognition. As we mentioned before, we do have a Spanish adaptation of the fourth edition, the Wechsler-IV memory scale [169-170]. The most widely used associative memory test is The Memory Binding Test (MBT) [186]. It is based on the specificity of coding and the evaluation of free memory strategies and with a semantic key, through the memorization of two lists with items of the same semantic categories in both lists; this is how the two lists' interference is evaluated. The MBT owns a reasonable validity for aMCI discrimination of healthy controls [187], and we also have a validation study in our country [188]. 6.2.5 Visual Memory tests. An advantage of tests that evaluate memory using images is that they are adequate for illiterate and low-educated patients [189]. Introduction 44 The most used ones have been the Copy and reproduction of the Rey Complex Figure [187] and the Visual Reproduction subtest of the Wechsler Scale [167]. The task of copying the Rey figure is a widely used test in routine clinical practice. It consists of reproducing a meaningless figure of high geometric complexity due to its richness of details. A second phase can be performed, which assesses visual memory at 3 and 30 minutes. It usually takes about 10 minutes to administer [181]. One of its drawbacks is that it has several scoring systems, although the best known is the 36-point system developed by Osterrieth [191]. The visual reproduction subtest of the Wechsler Scale evaluates the immediate and delayed recall of simple geometric figures, presented for 10 seconds. Evaluates the copy, recognizing the drawings presented among others presented as distractions, and their memory. We might highlight The Delayed Matching-to-Sample Task 48 (DMS48) [192] within the visual recognition tests. This type of visual recognition test is severely impaired in the perirhinal cortex lesions, compared to hippocampal lesions, where it is mildly affected or even intact [193], due to neurofibrillary tangles are initially deposited in the perirhinal cortex rather than in the entorhinal cortex or the hippocampal formation [194]. The DMS48 test presents 48 visual stimuli, colored drawings divided into three types of elements: abstract, paired, and unique. It consists of a coding phase and three recognition tasks of the drawings seen in the first phase, together with another distractor [192]. Introduction 45 In addition to being suitable for illiterate or low-educated patients, it appears relatively independent of attentional processes. Although it requires 30 minutes between sets 1 and 2, the administration of each of them takes around 5 minutes [192]. The biggest drawback is that we do not have normative or validation studies in our setting, having to be guided by the references of the French [195] and Chinese [196] studies. In its French normalization, it shows a good balance between sensitivity and specificity, both for immediate recall (Sensitivity 70.6%; Specificity 79.6%) and delayed recall (Sensitivity 79.4%; Specificity 72.9%). Furthermore, it has a high negative predictive value, around 98.5% [195]. The Free and Cued Selective Reminding Test (FCSRT), already discussed above (Headland 6.2.3), also has an available pictorial version. It is essential to know that their verbal and pictorial versions are not equivalent. The scores are higher in the pictorial one since both in MCI and healthy adults, the information presented graphically improves coding performance [197]. This better performance observed in memory tests that use images instead of words could be related to the “dual coding theory,” which proposes that images are more beneficial than words because images evoke both verbal and image codes, while words only trigger an abstract verbal code [198]. Both verbal and pictorial versions demonstrate appropriate discriminant validity between Healthy controls and MCI patients, with a Sensitivity over 90% for the Free Recall (FR), over 85% for the Total Recall (TR), and a Specificity over 90 % for both FR and TR [198]. Following the line of the Memory Binding Test [186] and its relationship with hippocampal injury [188], we find a new test assessing memory by association is the Introduction 46 "Memory Associative Test of the District of Seine-Saint-Denis" TMA-93 [199], which examines binding by images: drawings of familiar objects of everyday life. By using images, this test overcomes the difficulty of testing memory in low educated individuals. In the original paper, the test demonstrated optimal diagnostic accuracy to differentiate AD patients from healthy controls in a poorly educated and culturally diverse population [199]. The TMA93 evaluates binding memory by ten semantically related pairs of daily life objects. Those pairs of objects are shown in the encoding phase (Figure 10a), while in the recall phase (Figure 10b), only one of the two items has to recall the missing one. Figure10. TMA-93. Example sheets of the TMA-93. Maillet et al, 2017. A) Sheets for the coding task. B) Sheets for the recall task. A B) Introduction 47 In its original validation study, the TMA-93 total score identified AD patients with high sensitivity (88%) and specificity (97%), with the total score strongly correlated with the FCSRT free memory and total memory scores [199]. Introduction 48 7. DEVELOPMENT OF A DIAGNOSTIC TEST Similar to treatments, new diagnostic methods need a rigorous evaluation process before their introduction in routine clinical practice [200]. 7.1 Definition of the test The first step is accurately defining the test's purpose because this will influence many of the later steps in the selection process. Considerations may include the disease or condition to be diagnosed, whether a single test or diagnostic algorithm is required, and whether the test should or could provide a qualitative or quantitative result. It is also crucial to correctly define its potential final user (for example, specialized neurologist or neuropsychologist, or a primary care worker). It will also be vital to determine its clinical use for detection (sensitive test) or for diagnostic confirmation (a very high specificity will be the choice) [201-202]. After selecting the test that we are preparing to develop, we must review other similar tests available for that same condition. 7.2 Validation studies After selecting our objective test, the first is to perform a preliminary validation (I) [200], a cross-sectional or case-control study with a convenience sample. This preliminary validation aims to confirm differences in results between patients and Healthy Controls (HCs). The belonging to patients or HCs group must be determined by a gold-standard applied before the test to validate. Introduction 49 The formed groups must be different concerning the diagnosis and not regarding other variables (age, gender, educational level). The new test's diagnostic accuracy is analyzed using ROC curves and estimated by the area under the curve, using the gold standard as a reference criterion. The diagnostic accuracy (or diagnostic utility) is considered Good if the area under the curve is higher than 0.80 and optimal if the area under the curve is higher than 0.90. The ideal cut-off is calculated by using the Youden index, which optimizes sensitivity and specificity. Phase II consists of a cross-sectional design, but the sample includes a broader representation of the process to be diagnosed, with doubtful cases and different process stages, in the same proportion as it appears in the test conditions is theoretically going to be applied. The critical characteristic of phase III is that the test to be validated is administered before the gold-standard, constituting a cohort. This design allows for evaluating the predictive-diagnostic capacity of the test. When conducting discriminative validation studies, it is necessary to follow the Standards for Diagnostic Accuracy Reporting (STARD), updated in 2015 [204-206], which guide how to improve the quality of reports, in order to avoid failures in study design, data collection, or test interpretation, among others. The STARD Statement lists 30 essential recommendations to include in any validation study to minimize biases and improve the results' generalizability and applicability [207]. Introduction 50 Assessments should be performed in the target population in which the test will be used, as this will provide clinical precision data that is appropriate for the prevalence of the disease locally and for other context-specific factors that could influence the precision, like common comorbidities. 7.3 Reliability studies Reliability studies include two concepts: the tool's internal consistency and the evaluation of the measurement's precision: the result does not change according to the observer (inter-observer reliability), and the result is reproducible in repeated studies (test-retest reliability). Internal consistency analyzes whether the different parts of the test measure the evaluated construct homogeneously, without low correlation or redundant items. It is estimated mainly by Cronbach's alpha: a greater value than 0.70 is considered acceptable, optimal between 0.90 and 0.95, but redundant if higher than 0.95. Interobserver variability studies test the agreement between two or more raters on test scores. Ideally, they include healthy and cases with less or greater severity. Test-retest variability studies are essential to assess precision, mainly if the measure will test longitudinal evaluations. There are typically included Healthy Controls. The period between first and second evaluations, conducted by the same rater, should take between 2 and 4 months. If the interval is shorter than two months, it could appear a "practical effect": if longer than four, we could find a potential diagnostic status change. Introduction 51 The interobserver and test-retest reliability studies' statistic of choice is the Intraclass Correlation Coefficient for continuous variables. 7.4 Normative studies Normative studies are carried out on a healthy reference population, with a broad representation, to assess how sociodemographic variables (age, gender, educational level) influence the test score [208]. The effect of sociodemographic variables on the test score is analyzed statistically. Depending on the result variables distribution, regressions or an approximation based on percentile references are followed to establish tables that allow knowing which scores correspond to 1.5 SD or 5th - 10th percentiles according to strata of combinations of sociodemographic variables. They will be the threshold that determines the pathological score cutoff. 7.5 Applicability studies After evaluating the precision, diagnostic evaluation studies should demonstrate the beneficial effects and potential harm derived from its implementation, the real utility, through studies of effect and applicability [209]. This validation phase needs a random assignment to determine whether participants take the index test or not, and the results are evaluated in terms of health, quality of life, or costs. We could also include here feasibility studies, which aim to analyze whether the test we are about to validate applies to our target population. Feasibility studies also consider the time it takes to administer our test, which is very important in our clinical Introduction 52 context, in where we have a limited time per patient, and a test may have an excellent potential diagnostic but not be suitable for our outpatient clinic context. There are only a few applicability studies due to the methodological design's complexity and the results' interpretation. Hypotheses Hypotheses 60 The hypotheses of this Doctoral Thesis are: Chapter I 1. TMA-93 is as discriminative as the FCSRT for diagnosing aMCI patients. Chapter II 2. TMA-93 has good reliability (internal consistency and inter-rater and test-retest reliability) and feasibility (task-tolerability, short-time, and simple administration and scoring). Chapter III 3. TMA-93 total score does not depend on educational attainment. Chapter IV 4. TMA-93 has high sensitivity for AD diagnosis and improves the biomarker’s prediction when added to the FCSRT results. Hypotheses 61 Objectives Objectives 63 The general aim of this Doctoral Thesis is to properly validate a new diagnostic tool for amnestic Mild Cognitive Impairment, suitable for illiterate and low-educated patients. The specific objectives of each chapter are the following: Objectives of chapter I: 1. To compare the diagnostic accuracy of the TMA-93 against the FCSRT to differentiate patients with amnestic Mild Cognitive Impairment from Healthy Controls in a sample including low-educated patients. Objectives of chapter II: 2. To study the reliability (internal consistency and inter-rater and test-retest reliability) of the TMA-93. 3. To study the TMA-93 feasibility, through a register of the participants' percentage who completed the test, and measuring the administration time. Objectives of chapter III: 4. To provide normative values for the TMA-93 in cognitively unimpaired older educationally-diverse Spanish population. Objectives of chapter IV: 5. To validate the TMA-93 using Alzheimer's Disease biomarkers as gold-standard. 6. To compare TMA-93 diagnostic characteristics against FCSRT ones on a Biobank sample of patients who initially consulted concerning memory complaints. Objectives 64 Methods and Results Methods and Results 66 Chapter I. Article I. Rodrigo-Herrero. S, Carnero-Pardo. C, Méndez-Barrio. C, Miguel-Tristancho. M, Graciani-Cantisán. E, Bernal Sánchez-Arjona. M, Maillet. D, Jiménez- Hernández. MD, Franco-Macías. E (2019) TMA-93 for Diagnosing Amnestic Mild Cognitive Imapirment: A Comparison With the Free and Cued Selective Reminding Test. American Journal of Alzheimer’s Disease & Other Dementias 34(5): 322-328 _______________________________________ Chapter II. Article II. Franco-Macías. E, Rodrigo-Herrero. S, Luque-Tirado. A, Méndez-Barrio. C, Medina-Rodríguez. M, Graciani-Cantisán. E, Bernal Sánchez-Arjona. M, Maillet. D (2020) Reliability and Feasibility of the Memory Associative Test TMA-93. Journal of Alzheimer’s Disease Reports 4: 431-440 _______________________________________ Chapter III. Article III. Rodrigo-Herrero. S, Sánchez-Benavides. G, Ainz-Gómez. L, Luque-Tirado. A, Graciani-Cantisán. E, Bernal Sánzhez-Arjona. M, Maillet. D, Jiménez-Hernández. MD, Franco-Macías. E. (2020) Norms for Testing Visual Binding Using the Memory Associative Test (TMA-93) in Older Educationally-Diverse Adults. Journal of Alzheimer’s Disease 75 (2020) 871-878 _______________________________________ Methods and Results 67 Chapter IV. Article IV. Rodrigo-Herrero. S, Luque-Tirado. A, Méndez-Barrio. C, García-Solís. D, Bernal Sánchez-Arjona. M, Oropesa-Ruiz. JM, Maillet. D, Franco-Macías. E (2021) TMA-93 validation by AD biomarkers. A comparison with the FCSRT on a Biobank sample. Manuscript under review, submitted to the Journal of Alzjeimer’s Disease. Track number JAD 21-0115 Methods and Results 68 Chapter I. Article I. Methods and Results 75 Chapter II. Article II. Methods and Results 76 Methods and Results 77 Methods and Results 78 Methods and Results 79 Methods and Results 80 Methods and Results 81 Methods and Results 82 Methods and Results 83 Methods and Results 84 Methods and Results 92 Journal of Alzheimer’s Disease 82 (2021) 401–410 DOI 10.3233/JAD-210115 IOS Press 401 TMA-93 Validation by Alzheimer’s Disease Biomarkers: A Comparison with the Free and Cued Selective Reminding Test on a Biobank Sample Silvia Rodrigo-Herreroa,∗, Andrea Luque-Tiradob, Carlota M´ endez-Barrioa, David Garc´ ıa-Sol´ ısc, Mar´ ıa Bernal S´ anchez-Arjonab, Juan Manuel Oropesa-Ruiza, Didier Mailletd and Emilio Franco-Mac´ ıasb aMemory Unit, Juan Ram´on Jim´enez University Hospital, Huelva, Spain bMemory Unit, Virgen del Rocio University Hospital, Seville, Spain cNuclear Medicine Unit, Virgen del Rocio University Hospital, Seville, Spain dNeurology Service, Saint-Louis Hospital (AP-HP), Paris, France Accepted 17 April 2021 Pre-press 19 May 2021 Abstract. Background: The Memory Associative Test TMA-93 examines visual relational binding, characteristically affected in early-AD stages. Objective: We aim to validate the TMA-93 by biomarkers determination and compare its diagnostic characteristics with the Free and Cued Selective Reminding Test (FCSRT). Methods: Retrospective analysis of a Biobank database. Patients’ records initially consulted for memory complaints, scored MMSE ≥22, had TMA-93 and FCSRT tested, and AD biomarker determination (Amyloid-PET or CSF), either positive or negative, were selected. As cutoffs, we considered the 10-percentile for TMA-93 (P10/TMA-93), and “total free recall” (TFR) 21/22, total recall (TR) 43/44, and Cued Index < 0.77 for FCSRT from previous Spanish validation and normative studies. Diagnostic utilities were calculated using ROC curves and compared by the DeLong method. We studied if one test improved the other test’s prediction, following a forward stepwise logistic regression model. Results: We selected 105 records: 64 “positive” and 41 “negative” biomarkers. TMA-93 total score diagnostic utility (AUC = 0.72; 95%CI:0.62–0.82) was higher than those of the FCSRT: TFR (AUC = 0.70; 95%CI: 0.60–0.80), TR (AUC = 0.63; 95%CI:0.53–0.74), and Cued Index (AUC = 0.62; 95%CI:0.52–0.73). The P10/TMA-93 cutoff showed 86% sensitivity, similar to that of the most sensitive FCSRT cutoff (TFR21/22, 89%) and 29% specificity, lower than that of the most specific FCSRT cutoff (Cued Index < 0.77, 57%). 32.8% of the positive-biomarker group scored above CI/0.77 but below p10TMA-93. The addition of TMA-93 total score to FCSRT variables improved significantly the biomarkers results’ prediction. Conclusion: TMA-93 demonstrated “reasonable” diagnostic utility, similar to FCSRT, for discriminating AD biomarker groups. TMA-93 total score improved the AD biomarker result prediction when added to FCSRT variables. Keywords: Alzheimer’s disease, amyloid-PET, biomarker, cerebrospinal fluid, free and cued selective reminding test, TMA-93 ∗Correspondence to: Silvia Rodrigo-Herrero, Neurologist and Neuropsychologist consultant, Atlantic Senior Fellow for Equity and Brain Health, Global Brain Health Institute, Juan Ramon Jimenez University Hospital, Ronda Exterior Norte s/n, First floor, Secretary of Neurology, 21005 Huelva, Spain. E-mail: silvia. [email protected]. INTRODUCTION In the past few years, the development of new diagnostic criteria for Alzheimer’s disease (AD) has ISSN 1387-2877/$35.00 © 2021 – IOS Press. All rights reserved. 402 S. Rodrigo-Herrero et al. / TMA-93 validation by Alzheimer’s Disease Biomarkers allowed an earlier diagnosis of the disease and, consequently, earlier interventions [1, 2]. These criteria, including neuropsychological data and cerebrospinal fluid (CSF) and amyloid-positron emission tomography (PET) biomarkers, have allowed the identification of prodromal and mild dementia AD patients [3, 4]. Both stages are mainly characterized by a specific episodic memory deficit [5]. The Free and Cued Selective Reminding Test (FCSRT) [6] is the gold standard to test episodic memory, being able to predict the conversion from MCI to clinical dementia, based on cue testing [7]. The test controls for a successful encoding (achieved by cued recall), and it facilitates retrieval processing with the same semantic cues. Despite cueing, a low total recall has an excellent specificity for AD [8]. However, FCSRT may not be sensitive enough due to the ceiling effect [9]. Ongoing research focuses on developing new tests designed to target other fundamental memory components, including relational binding. In cognition, binding is the function that supports the integration of multiple elements together. Relational binding is the ability to remember novel associations between words or pictures. Associative binding tasks seem to be more sensitive to capture AD- related memory impairment than those not assessing binding [10–12]. In this sense, the “Memory Associative Test of the district of Seine-Saint-Denis” (TMA-93) is a recently developed test that examines “visual relational binding”. By using images, this test overcomes the difficulty of testing memory in low educated individuals. In the original paper, the test demonstrated optimal diagnostic accuracy to differentiate AD patients from healthy controls in a poorly educated and culturally diverse population [13]. However, the TMA-93 is broader in scope than low-educated patients. A recent preliminary validation study demonstrated that the test is as sensitive as the picture version of the FCSRT in discriminating between amnestic mild cognitive impairment (aMCI) patients and healthy controls [14]. Another advantage for the TMA-93 is its short administration time (6 min for aMCI patients), turning the test feasible to be administered in the busy Primary Care and General Neurology Outpatient settings, with limited face-to-face time per patient [15]. Our study aims to validate the TMA-93 using biomarkers as gold-standard and to compare its diagnostic characteristics with those of the FCSRT on a Biobank sample of patients who initially consulted with memory complaints and finally had a positive or negative AD biomarker result. METHODS Design This study is an observational, cross-sectional, and retrospective study. We compared the performance of two memory tests (TMA-93 and FCSRT) for AD diagnosis in a patients’ sample with a Biobank register and an AD-biomarkers available result. TMA-93 was compared to FCSRT concerning diagnostic accuracy, sensitivity, and specificity, diagnostic agreement, and predictive ability according to the AD biomarker result. Study population Patients had been evaluated between 2017 and 2020 at the Outpatient Memory Clinic of a tertiary Hospital in Southern Spain: Virgen del Rocio University Hospital. They had been referred to our clinic for memory complaints and had undergone complete clinical, neuropsychological, analytic, and neuroimaging studies. Biomarkers (amyloid PET or CSF) were done to diagnose early AD following recommendations from the Spanish Neurology Society [16]. After informed consent, all data were registered in the Biobank collection “Alzheimer and other dementias” (C330022), belonging to the Andalusian Biobank. For this study, we selected from the database the records with the following criteria: 1) initial consultation for memory complaints corroborated by a relative; 2) Mini-Mental State Examination (MMSE) score between 22 and 30; 3) as part of the neuropsychological examination, they had performed both FCSRT and TMA-93; 4) available result, positive or negative, for specific AD biomarkers. TMA-93: procedure and cutoffs The TMA-93 had been used during the diagnosis, following the instructions given by its authors [13]. It examines binding by images: drawings of familiar objects of everyday life. During the encoding phase (Fig. 1A), the patient is shown ten semantically related pairs of daily life objects (tree/bird, bed/bedside-lamp, boat/fish, dog/sheep, foot/trousers, knife/apple, glasses/book, hand/watch, car/key, flower/sun). In the recall phase (Fig. 1B), the patient is shown only one of the two items, having to recall the missing one. After each answer or up to 5 s, the examiner displays the pair again as a recall. The maximum score of 30 points is granted when S. Rodrigo-Herrero et al. / TMA-93 validation by Alzheimer’s Disease Biomarkers 403 Fig. 1. Example of drawing TMA-93. A) Encoding phase B) Recall phase. the patient produces 10 out of 10 correct responses in this first trial, in which case the second and the third trials are omitted. If the patient fails any item, s/he is scored from 0 to 9 based on correct answers in this first trial and is administered a second similar trial with the same 10 pairs of drawings. If the patient correctly recalls the 10 missing objects in this second trial, s/he is given 20 points: 10 points corresponding to the second trial and 10 more corresponding to the third trial, which is canceled. Otherwise, the third trial is undertaken, and its score is added to those of the previous two trials. For this study, as cutoffs, we chose the 5th and 10th percentiles for the TMA-93 total scores (P5/TMA-93 and P10/TMA-93, respectively), obtained from the Spanish normative study for the test [17]. For the P5/TMA-93 cutoff, a total TMA-93 score was considered pathological if equal or lower than the score that marks the 5-percentile. This score varies from 19 out of 30 for the older and less educated group to 27 out of 30 for the younger and more educated one, according to the Spanish normative study [17]. For the P10/TMA-93 cutoff, a total TMA-93 score was considered pathological if equal or lower than the score that marks the 10-percentile. This score varies from 20 out of 30 for the older and less educated group to 28 out of 30 for the younger and more educated one, according to the same study [17]. FCSRT: procedure and cutoffs During patients’ diagnosis, we had used the picture version of the FCSRT with Immediate Recall following the instructions given by its authors [6]. It was administered on a different session and different day, with a maximum of 2 weeks delay, to avoid interference from the TMA-93. The test begins with an encoding phase, in which participants must examine 16 easily recognizable drawings, each one from a unique semantic category. The 16 items are presented in groups of four, in four different cards. The patient is asked to point to and name each item after its semantic cue. After the correct identification, the card is removed, and the immediate cued recall of those four items presented before is tested. The participant is reminded of failed items to retrieve by presenting the cue and the item together. Once immediate recall of each card is completed, the patient is asked for 3 recall trials, each one preceded by 20 s of backward counting as interference. Each recall trial has three parts. First, each participant has up to 2 min to freely recall as many items as possible. Afterward, the semantic cues are verbally provided for the not retrieved items. Lastly, if the patient fails to retrieve the item with the category cue, s/he is reminded by presenting the cue and the item together (selective reminding). We recorded three variables: total free recall (TFR), the sum of free recall from the three memory trials); total recall (TR), the sum of free and facilitated recall of the three memory trials); Cued Index, the result of the TFR-TR / TFR-48 ratio. Forthisstudy, we chose the best cutoffs obtained by the Spanish validation study for the picture version of the FCSRT: TFR 21/22; TR 43/44; Cued Index < 0.77 [14]. In other words, a total score equal or lower 21 for TFR, a total score equal or lower 43 for TR, or a total score lower than 0.77 for Cued Index, were considered as pathological. AD biomarkers status: procedures and the positive or negative result Patients had a positive or negative result for AD diagnosis by amyloid PET or CSF biomarkers. For amyloid-PET, the cortical deposition of amyloid-!had been assessed by florbetaben during 404 S. Rodrigo-Herrero et al. / TMA-93 validation by Alzheimer’s Disease Biomarkers the diagnosis. Patients underwent PET between 90 and 110 min after intravenous injection of 300 MBq of florbetaben, SUVs were calculated using the cerebellar cortex as a reference region. The acquisition time was 20 min. Images were assessed by an expert on Nuclear Medicine (DGS) following the European Association of Nuclear Medicine recommendations [18] with an “Amyloid positive” (A+) or “Amyloid negative” (A−) final result. For CSF AD biomarkers, a lumbar puncture had been done in the intervertebral region L3/L4 or L4/ L5. A sample was taken in polypropylene test tubes with storage in the IBIS Biobank C330022 collection. Its conservation was at < 80◦C, proceeding only with thawing for the determination, following the international consensus recommendations [19]. CSF A!42, A!40, T-Tau, and P-Tau levels were determined following the manufacturers’ recommendations by the Euroimmun automated method [20]. CSF biomarker result was classified as positive or negative depending on the A!42/A!40 ratio. If lower than 0.095 was considered “Amyloid positive” (A+), and if equal to or greater than 0.095, “Amyloid negative” (A−). To complete the AT(N) classification, if a P-tau resulted higher than 61 pg/ml was considered “Tau positive” (T+), and if a T-tau resulted higher than 452 pg/ml, “Neurodegeneration positive” (N+). Values of P-tau ≤61 pg/ml and T-tau ≤452 pg/ml were considered “Tau negative” (T−) and “Neurodegeneration negative” (N−), respectively. Ethics This study was approved by the Hospital Virgen del Rocio (Seville, Spain) ethics committee and conducted according to the World Medical Association Declaration of Helsinki. Written informed consent was obtained from all patients. Statistical study We present descriptive results with frequencies for categorical variables; mean, standard deviation, and range for quantitative and normal distribution variables; and median, interquartile range, and range for quantitative non-normal distribution variables. Comparisons between groups for continuous variables were made with Student’s t-test if the variable had a normal distribution and the Mann-Whitney U test if it had a non-normal distribution. Comparisons between groups for categorical variables were made with the Chi-square test. The diagnostic utility of the two tests was analyzed using ROC curves and estimated by the area under the curve, considering: “excellent” (> 0.90), “good” (> 0.80), “reasonable” (> 0.70), or “low” (< 0.70). Comparisons between ROC curves were made according to the DeLong method. Crosstabs were made up for sensitivity and specificity analysis. We considered the two cutoff points chosen for the total score on TMA-93 (P5/TMA- 93, P10/TMA-93) and the three cutoffs for the total scores on FCSRT (TFR 21/22, TR 43, Cued Index < 0.77). We displayed values as dichotomized results, both in the rows (according to a result above or below the cutoffs chosen for each test) and columns (positive or negative AD biomarker result). For the study of diagnostic concordance between the tests, we considered the P10/TMA-93- TFR 21/ 22, TFR 21/22- Cued Index < 0.77, and P10/TMA- 93- Cued Index < 0.77 pairs. We constructed 3 crosstabs, also displayed as dichotomized values. The logistic regression analysis was performed following a forward stepwise model. We first calculated whether the addition of the TMA-93 total score to the models made up of age and each of the FCSRT total scores (TFR, TR, Cued Index) improved the prediction biomarker outcome. Afterward, we tested the inverse hypothesis: if the addition of the FCSRT total scores improved the model’s prediction that already included the TMA-93 total score and age. We studied the difference in chi-square, statistical significance, and the difference in Nagelkerke’s R2 for each step. We also calculated the diagnostic utility of the models by ROC curve analysis. All statistical analyzes were run in SPSS version 25 (IBM, USA), and statistical significance was established at p< 0.05. RESULTS Biobank database analysis selected 105 registers that met the inclusion criteria and were finally included in the statistical analysis. 16 patients had been diagnosed by amyloid PET and 89 by CSF biomarkers. By amyloid PET, there were 14 A+ patients and 2 A−patients. By CSF, there were 50 A+ patients [A+T+N+(n= 38), A + T−N−(n= 11), A+T−N+ (n= 1)] and 39 A−patients [A−T−N− (n= 38), A−T−N+ (n= 1)]. We formed the two final comparison groups by adding the A+ and the A−results obtained by amyloid PET and CSF. In total, 64 patients had S. Rodrigo-Herrero et al. / TMA-93 validation by Alzheimer’s Disease Biomarkers 405 Table 1 Sociodemographic data of the sample, split according to positivity or negativity of AD biomarker profile Positive AD (n=64) NegativeAD(n= 41) p Age 70.27 ±5.71 (56–82) 63.63 ±9.18 (46–83) <0.001 Gender 38 Males (59.4%) 17 Males (41.5%) 0.093 26 Females (40.6%) 24 Females (58.5%) Education Median = 6 Median = 8 0.937 IQR = 5 – 9 IQR = 5 – 9.5 Range = 0 – 19 Range = 0 – 13 MMSE Median = 24 Median = 25 0.078 IQR=23–26 IQR=23–27 Range = 22–30 Range = 22–30 TMA-93 Median = 18.50 Median = 23 <0.001 IQR = 10 – 21.75 IQR = 19 – 25 Range = 0 – 30 Range = 0 – 29 FCSRT •TFR Median = 15.50 Median = 20 IQR = 9 – 18.75 IQR = 14 – 24 <0.001 Range = 1 – 34 Range = 7 – 33 •TR Median = 39 Median = 44 IQR = 33 – 45 IQR = 37.50 – 46 Range = 11 – 48 Range = 28 – 48 0.022 •Cued Index Median = 0.73 Median = 0.81 IQR = 0.55 – 0.89 IQR = 0.67 – 0.93 Range = 0.10 – 1 Range = 0.40 – 1 0.032 Amyloid PET 14 2 CSF 50 39 <0.001 A!42 Median = 410.00 Median = 938.00 IQR = 301.77–605.25 IQR = 630.75–1185.50 Range = 188.80–950.90 Range = 319.80–1662.60 A!42/A!40 Median = 0.059 Median = 0,134 IQR = 0.044 – 0.072 IQR = 0.100–0.145 Range = 0.030 – 0.091 Range = 0.097 – 0.161 T-Tau Median = 488.95 Median = 212.80 IQR = 379.27 – 592.05 IQR = 167.00–284.40 Range = 180.20–2320 Range = 20–455.40 P-Tau Median = 99.10 Median = 28 IQR = 60–156.57 IQR = 18.20–40.80 Range = 6.80–734 Range = 5–55.40 AD, Alzheimer’s disease; MMSE, Mini-Mental State Examination; FCSRT, Free and Cued Selective Reminding Test; TFR, Total Free Recall; TR, Total Recall; CSF, cerebrospinal fluid. “positive biomarkers” and 41 “negative biomarkers” (Table 1). Age was significantly higher in the “positive biomarkers” group. The overall educational level of the sample was low, with 25% of patients with education equal to or less than 5 years, although with no differences between groups. The total MMSE score was similar in both groups. Total scores on memory tests were significantly lower in the “positive biomarker” group (Table 1). We analyze the positive group by CSF and compare the A + T + N+ and the A+T−N−subgroups on memory tests. There were not significant differences in total scores for TMA- 93 (p= 0.072), TFR FCSRT (p= 0.931), TR FCSRT (p= 0.557), and Cued Index FCSRT (p= 0.970). The diagnostic utility of TMA-93 to discriminate between the “positive biomarkers” and “negative biomarkers” groups (AUC = 0.72; 95% CI: 0.62–0.82, p< 0.001) was higher than those of the three vari- 406 S. Rodrigo-Herrero et al. / TMA-93 validation by Alzheimer’s Disease Biomarkers Fig. 2. Diagnostic utility of TMA-93 and each variable of FCSRT (TFR, TR. and Cued Index). Represented by DeLong method. FCSRT, Free and Cued Selective Reminding Test; TFR, Total Free Recall; TR, Total Recall. ables resulting from FCSRT: TFR (AUC = 0.70; 95% CI: 0.60–0.80, p= 0.001), TR (AUC = 0.63; 95% CI: 0.53–0.74, p= 0.022), Cued Index (AUC = 0.62; 95% CI: 0.52–0.73, p= 0.033), although the differences were not statistically significant according to DeLong method (Fig. 2). For TMA-93, the diagnostic utility was analyzed according to education after segmenting the database by the median of 6 years of education. The diagnostic utility was not different for discriminating “positive biomarkers” and “negative biomarkers” by education equal or lower than 6 years (AUC = 0.72; 95% CI: 0.58–0.86) or education greater than 6 years (AUC = 0.72; 95% CI: 0.57–0.86). P5/TMA-93 and P10/TMA-93 cutoffs showed a sensitivity of 75% and 86%, respectively, and specificity of 41% and 29%, according to biomarkers’ results. TFR 21/22 showed higher sensitivity (89%) and a medium specificity (33%), closer to that of the P10/TMA-93, whereas TR 43/44 and Cued Index < 0.77 showed lower sensitivity (62 and 55%, respectively) and higher specificity (54% and 57%, respectively). Regarding tests’ diagnostic agreement, 79.7% of “positive biomarkers” group patients showed scores under P10/TMA-93 and TFR 21/22 cutoffs (Table 2). TFR 21/22 detected 6 biomarkers positive cases not detected by P10/TMA-3. On the other hand, P10/TMA-93 detected 4 positive cases not detected by TFR 21/22. Three patients of the “positive biomarkers” group (4.7%) scored over both cutoffs (Table 2). These subjects were younger than 70 years old, had at least completed primary studies, Table 2 Agreement between P10/TMA-93, and TFR 21, and Cued Index of the FCSRT. Split according to Positivity or Negativity on AD biomarkers result “Positive” “Negative” ‘AD AD (n= 64) (n= 41) TMA-93 > P10 / TFR > 21 3 (4.7%) 6 (14.6%) TMA-93 ≤P10 / TFR > 21 4 (6.3%) 9 (22%) TMA-93 > P10/ TFR ≤21 6 (9.4%) 5 (12.2%) TMA-93 ≤P10/ TFR ≤21 51 (79.7%) 21 (51.2%) TFR > 21/Cued Index ≥0.77 7 (10.9%) 12 (29.3%) TFR ≤21/Cued Index ≥0.77 22 (34.4%) 12 (29.3%) TFR > 21/Cued Index < 0.77 0 2 (4.9%) TFR ≤21/Cued Index < 0.77 35 (54.7%) 15 (36.6%) TMA-93 > P10/Cued Index ≥0.77 7 (10.9%) 11 (26.8%) TMA-93 ≤P10/Cued Index ≥0.77 21 (32.8%) 12 (29.3%) TMA-93 > P10/Cued Index < 0.77 2 (3.1%) 1 (2.4%) TMA-93 ≤P10/Cued Index < 0.77 34 (53.1%) 17 (41.5%) AD, Alzheimer’s disease; TFR, Total Free Recall; TR, Total Recall. and their ranges of total scores in the tests were: MMSE (26–30), TMA-93 (28–30), TFR (28–34), TR (46–47), and Cued Index (0.85–0.95). 34.4% and 32.8% of “positive biomarkers” group patients scored above the Cued Index < 0.77 cutoff point but below the TFR 21/22 and p10TMA-93 cutoffs, respectively (Table 2). Logistic regression models showed that the addition of TMA-93 total score to the FCSRT models (TFR, TR, and Cued Index) improved the biomarker’s prediction result. Each step was significant, improving the model’s fit and registering an increase in Nagelkerke’s R2. The resulting models showed a “reasonable” AUC to discriminate biomarker results. Otherwise, adding each of the FCSRT total scores to the model that already included the TMA-93 total score did not improve the prediction (Table 3). DISCUSSION In light of ongoing efforts to develop anti-dementia drugs, identifying AD patients at the early stages is increasingly a clinical priority. Early diagnosis and intervention have become political priorities in national and international dementia strategies [21–22]. In this context, cognitive tests to detect patients suffering from MCI and Mild dementia with underlying AD pathology are essential. This study is the first validation with biomarkers of the TMA-93. The TMA-93 total score demonstrated a “reasonable” diagnostic utility for discriminating between S. Rodrigo-Herrero et al. / TMA-93 validation by Alzheimer’s Disease Biomarkers 407 Table 3 Results of logistic regression analyses for models with different sets of predictors MODELS -2LL DIF χ2pNagelkerke R2AUC 1A (TFR) 115.696 0.287 1B (TFR + TMA-93) 111.524 3.982 0.046 0.327 0.783 (0.693–0.874) p< 0.001 2A (TR) 114.824 0.294 2B (TR + TMA-93) 110.682 4.142 0.042 0.335 0.793(0.705–0.881) p< 0.001 3A (Cued Index) 115.753 0.284 3B (Cued Index + TMA-93) 111.020 4.733 0.030 0.332 0.791 (0.703–0.879) p< 0.001 4A (TMA-93) 111.901 0.323 0.784 (0.693–0.875) p< 0.001 4B (TMA-93 + TFR) 111.524 0.378 0.539 0.327 4C (TMA-93 + TR) 110.682 1.219 0.270 0.335 4D (TMA-93 + Cued Index) 111.020 0.881 0.348 0.332 Model 1A with age and TFR. Model 1B with age, TFR, and TMA-93. Model 2A with age and TR. Model 2B with age, TR, and TMA-93. Model 3A with age and Cued Index. Model 3B with age, Cued Index, and TMA-93. Model 4A with age and TMA-93. Model 4B with age, TMA-93, and TFR. Model 4C with age, TMA-93, and TR. Model 4D with age, TMA-93, and Cued Index. Dif, difference; AUC, Area Under the Curve; FCSRT, Free and Cued Selective Reminding Test; TFR, Total Free Recall; TR, Total Recall. “positive” and “negative” biomarker groups. This utility was higher, although not significantly, than those of the FCSRT. The test has already shown an optimal diagnostic utility for distinguishing aMCI patients from healthy controls [14]. Here, the requirement was greater as we demanded the test to discriminate among patients with memory complaints and MMSE total score ≥22 the belonging to a positive or negative AD biomarker group. The result, statistically only “reasonable”, is substantial for this design, even more, if it is taken into account that the “positive” and “negative” groups were not different regarding the MMSE total score. In this study, the total sample was globally low educated, consistent with limited access to education for many elderly Spanish people in the aftermath of the Spanish Civil War (1936-1939). The TMA-93 and FCSRT’s picture version have been demonstrated feasible and discriminative for low-educated patients [15]. The TMA-93 diagnostic utility was not different when segmented the database by six years of education. However, this point should be addressed in future TMA-93 validation studies, including more educated populations. We validated the 5th and 10th percentile scores, stratified by age and educational level, obtained from a recent normative Spanish study for the TMA-93. In patients with memory complaints and MMSE ≥22, total scores equal or under the 5th and 10th percentiles showed 75 and 86% sensitivity and 41 and 29% specificity, respectively, for diagnosis with biomarkers. The 86% sensitivity obtained by scores equal or under the 10th percentile combined with the 6 minutes the test averaged in a previous feasibility study in aMCI patients [15] makes the TMA-93 a potentially useful screening tool for AD diagnosis. The test could also detect cognitive impairment causes other than AD, conditioning the low specificity for AD. In a second step, the complementary use of biomarkers would provide the AD diagnosis’s necessary specificity. Regarding the concordance between the most sensitive cutoffs for both tests, P10/TMA-93 and TFR 21/22, there was a reasonable agreement as both cutoffs share similar diagnostic characteristics. TFR 21/22 resulted slightly more sensitive and specific than P10/TMA-93, and this means a discrete advantage over TMA-93 as a screener. The advantage for P10/TMA-93 is again its short administration time, less than half of that for FCSRT [15]. The sequential use of both tests at different sessions and possibly in different contexts with more or less time availability would increase the sensitivity up to 95,3%, and this approach could be taken into account in clinical practice guidelines. On the contrary, this diagnostic agreement analysis also revealed that 4.7% of patients with corroborated memory complaints and MMSE ≥22 scored above TFR 21/22 and P10/TMA-93, despite showing “positive biomarkers.” This small percentage represents a challenge in clinical practice. It would be necessary to define better this group of patients and examine them more in-depth with higher demanding memory tests with a reduced ceiling effect. In this study, the not-detected three AD patients were under 70 years, had completed 408 S. Rodrigo-Herrero et al. / TMA-93 validation by Alzheimer’s Disease Biomarkers at least primary education, and scored between 26 and 30 on the MMSE. There was disagreement between both results with the cutoff of the Cued Index (< 0.77). More than 40% of patients with positive biomarkers maintained the cued efficiency (Cued Index ≥0.77) in this Biobank sample, resulting in false negatives for AD pathology. FCSRT is currently the gold-standard test for screening in clinical trials of anti-amyloid molecules for AD pathology. In this context, a lower than TFR 21/22 ensures a higher sensitivity, and, finally, a lower than Cued Index < 0.77 assures a higher specificity. Operationally, the latter runs as the limiting step. Following this rule, more than 40% of biomarkerpositive patients analyzed in the present study could hypothetically fail screenings requiring a Cued Index lower than 0.77. The P10/TMA-93 reduced this percentage of false negatives to 11%, so this result positions the test as a potentially useful tool in clinical trials’ screenings. The TMA-93 improved the prediction of biomarker results when added to FCSRT models. We found this result when adding the TMA-93 total score to the FCSRT model, including the TFR, TR, or Cued Index parameters. We never found an improved prediction when adding the FCSRT variables to the model with the TMA-93 total score variable. These findings would allow us to conclude that the binding test improved biomarker’s prediction when added to a test based on coding with a semantic clue. In this way, memory tools testing “binding” are currently being incorporated into the trial procedures to select AD patients at an earlier stage [10]. TMA-93, examining “visual relational binding” could be useful for that purpose. The test could complement FCSRT (particularly when the cue efficiency is still preserved) to predict biomarker results. Limitations The main limitation of this study is its retrospective nature. Results should be confirmed in a prospective study with a systematic selection procedure, avoiding selection bias. Although heterogeneous, our sample follows the normal population distribution for Memory Outpatient clinics in our area. Participants’ selection was mainly based on memory complaints corroborated by an informant and MMSE ≥22. This selection was not conditioned by any memory test result, allowing us to get some surprising results, such as the 4.7% of positive biomarkers patients who scored above both FCSRT and TMA-93 cutoffs. A second limitation is that the biomarkers, positive or negative, were studied by different methods (Amyloid-PET and CSF). However, the final result, dichotomized into positive or negative groups, allowed us to analyze diagnostic utility, diagnostic agreement, and tests’ prediction. Considering the positive group, A+T+N+patients’ performance on both TMA-93 and FCSRT was not significantly different from the A + T−N−ones. However, TMA- 93 total score was relatively close to a significant difference. A future study focused on TMA-93 discrimination between “Alzheimer pathologic change” and “biological AD” diagnosis with an appropriate sample size could confirm a significant difference between groups [23]. A third limitation was the lack of diagnostic characterization for the “negative biomarkers” group. Patients diagnosed as “negative” by CSF constituted a relatively homogenous A−T−N−group. Considering the whole negative group (negative CSF and negative Amyloid-PET), more than 50% scored lower than P10/TMA-93 and TFR 21/22. This result suggests a hippocampal injury but without AD as substrate. Among possible underlying etiologies, we could consider primary age-related tauopathy, argyrophilic grain disease, hippocampal sclerosis, or cerebral age-related TDP-43 and sclerosis [24]. There are no specific biomarkers for these diseases, so we have to monitor the patients’ clinical course and preserve blood/CSF samples to facilitate future research. CONCLUSION In conclusion, TMA-93 demonstrated a “reasonable” diagnostic utility, similar to FCSRT, to discriminate AD biomarkers among patients with memory complaints and MMSE ≥22. The test showed good sensitivity for the AD diagnosis and improved the biomarker’s prediction when added to the FCSRT results. These results combined with other properties previously demonstrated as its optimal diagnostic utility for distinguishing aMCI patients from controls and good reliability and feasibility position the test as an excellent alternative to examine “relational binding” for screening purposes, including individuals with a low educational level and contexts with limited face-to-face time per patient. Prospective phase III validation studies, reaching areas such as Primary Care, will be the next step in the test development. S. Rodrigo-Herrero et al. / TMA-93 validation by Alzheimer’s Disease Biomarkers 409 ACKNOWLEDGMENTS For assistance with recruitment, we thank Neurologists at the Memory Department from Virgen del Rocio University Hospital. We are also extremely grateful to all participants in our study. The study was sponsored by the Department of Health (Andalusia, Spain). No additional private funding. Authors’ disclosures available online (https:// www.j-alz.com/manuscript-disclosures/21-0115r2). REFERENCES [1] Jack CR Jr, Albert MS, Knopman DS, McKhann GM, Sperling RA, Carrillo MC, Thies B, Phelps CH (2011) Introduction to the recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 257-262. [2] Jameson JL, Longo DL (2015) Precision medicine— personalized, problematic, and promising. N Engl J Med 372, 2229–2234. [3] Dubois B, Feldman HH, Jacova C, Hampel H, Molinuevo JL, Blennow K, Dekosky ST, Gauthier S, Selkoe D, Bateman R, Cappa S, Crutch S, Engelborghs S, Frisoni GB, Fox NC, Galasko D, Habert MO, Jicha GA, Nordberg A, Pasquier F, Rabinovici G, Robert P, Rowe C, Salloway S, Sarazin M, Epelbaum S, de Souza LC, Vellas B, Visser PJ, Schneider L, Stern Y, Scheltens P, Cummings JL (2014) Advancing research diagnostic criteria for Alzheimer’s disease: The IWG-2 criteria. Lancet Neurol 13, 614-629. [4] Albert MS, Dekosky ST, Dickson DW, 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, 270-279. [5] Dubois B, Feldman HH, Jacova C, Cummings JL, Dekosky ST, Barberger-Gateau P, Delacourte A, Frisoni G, Fox NC, Galasko D, Gauthier S, Hampel H, Jicha GA, Meguro K, O’Brien J, Pasquier F, Robert P, Rossor M, Salloway S, Sarazin M, de Souza LC, Stern Y, Visser PJ, Scheltens P (2010) Revising the definition of Alzheimer’s disease: A new lexicon. Lancet Neurol 9, 1118-1127. [6] Buschke H (1984) Cued recall in amnesia. J Clin Neuropsychol 6, 433-440. [7] Grande G, Vanacore N, Vetrano DL, Cova I, Rizzuto D, Mayer F, Maggiore L, Ghiretti R, Cucumo V, Mariani C, Cappa SF, Pomati S (2018) Free and cued selective reminding test predicts progression to Alzheimer’s disease in people with mild cognitive impairment. Neurol Sci 39, 1867-1875. [8] Tounsi H, Deweer B, Ergis AM, Van der Linden M, Pillon B, Michon A, Dubois B (1999) Sensitivity to semantic cuing: An index of episodic memory dysfunction in early Alzheimer disease. Alzheimer Dis Assoc Disord 13, 38-46. [9] Gramunt N, S´ enchez-Benavides G, Buschke H, Lipton RB, Masramon X, Gispert JD, Pe˜ na-Casanova J, Fauria K, Molinuevo JL (2016) Psychometric properties of the memory binding test: Test-retest reliability and convergent validity. J Alzheimers Dis 50, 999-1010. [10] Pavisic IM, Suarez-Gonzalez A, Pertzov Y (2020) Translating visual short-term memory binding tasks to clinical practice: From theory to practice. Front Neurol 11, 458. [11] Rentz DM, Parra Rodriguez MA, Amariglio R, Stern Y, Sperling R, Ferris S (2013) Promising developments in neuropsychological approaches for the detection of preclinical Alzheimer’s disease: A selective review. Alzheimers Res Therapy 5, 58. [12] Parra MA, Abrahams S, Fabi K, Logie R, Luzzi S, Della Sala S (2009) Short-term memory binding deficits in Alzheimer’s disease. Brain 132, 1057-1066. [13] Maillet D, Narme P, Amieva H, Matharan F, Bailon O, Le Cl´ esiau H, Belin C (2017) The TMA-93: A new memory test for Alzheimer’s disease in illiterate and less educated people. Am J Alzheimers Dis Other Dement 32, 461-467. [14] Rodrigo-Herrero S, Carnero-Pardo C, M´ endez-Barrio, C, De Miguel-Tristancho M, Graciani-Cantis´ en E, S´ enchez- Arjona MB, Maillet D, Jim´ enez-Hern´ endez MD, Franco- Mac´ ıas E (2019) TMA-93 for diagnosing amnestic mild cognitive impairment: A comparison with the Free and Cued Selective Reminding Test. Am J Alzheimers Dis Other Dement 34, 322-328. [15] Franco-Mac´ ıas E, Rodrigo-Herrero S, Luque-Tirado A, M´ endez-Barrio C, Medina-Rodriguez M, Graciani- Cantis´ en E, Bernal Sanchez-Arjona M, Maillet D (2020) Reliability and feasibility of the Memory Associative Test TMA-93. J Alzheimers Dis Rep 4, 431-440. [16] Ministerio de Sanidad, consumo y bienestar social espa˜ nol (2019) Plan Integral de Alzheimer y otras Demencias, 2019- 2023. ED. Ministerio de Sanidad online: 731-19-108-2 https://www.mscbs.gob.es/profesionales/saludPublica/docs/ Plan Integral Alhzeimer Octubre 2019.pdf Updated 2019. Accessed on September 10, 2020. [17] Rodrigo-Herrero S, S´ anchez-Benavides G, Ainz-G´ omez L, Luque-Tirado A, Graciani-Cantisan E, Bernal Sanchez- Arjona M, Maillet D, Jim´ enez-Hern´ andez MD, Franco- Macias E (2020) Norms for testing visual binding using the Memory Associative Test (TMA-93) in older educationallydiverse adults. J Alzheimers Dis 75, 871-878. [18] Minoshima S, Drzezga AE, Barthel H, Bohnen N, Djekidel M, Lewis DH, Mathis CA, McConathy J, Nordberg A, Sabri O, Seibyl J, Stokes MK, Van Laere K (2016) SNMMI procedure standard/EANM practice guideline for amyloid PET imaging of the brain 1.0. J Nucl Med 57, 1316-1322. [19] Hampel H, Frank R, Broich K, Teipel SJ, Katz RG, Hardy J, Herholz K, Bokde ALW, Jessen F, Hoessler YC, Sanhai WR, Zetterberg H, Woodcock J, Blennow K (2010) Biomarkers for Alzheimer’s disease: Academic, industry and regulatory perspectives. Nat Rev Drug Discov 9, 560–574. [20] Chiasserini D, Biscetti L, Farotti L, Eusebi P, Salvadori N, Lisetti V, Baschieri F, Chipi E, Frattini G, Stoops E, Vanderstichele H, Calabresi P, Parnetti L (2016) Performance evaluation of an automated ELISA system for Alzheimer’s disease detection in clinical routine. J Alzheimers Dis 54, 55-67. [21] World Health Organization. Global action plan on the public health response to dementia 2017-2025. https://apps.who. int/iris/bitstream/handle/10665/259615/9789241513487- Discussion 108 Reliability study. Article II. "Reliability and Feasibility of the Memory Associative Test TMA-93" In this development phase, we aimed to study the reliability and feasibility of the TMA-93. This work is the first reliability and feasibility study for the TMA-93. Reliability Within the reliability study, we valued internal consistency and inter-rater and testretest reliability. The TMA-93 internal consistency between the ten pairs of semantically related pictures was "optimal" (Cronbach's alpha = 0.936). This consistency implies that the 10 test items are highly correlated, so they similarly measure the interest construct (the "visual relational binding"). Comparing, the FCSRT internal consistency has been described just as "acceptable" (Cronbach's alpha = 0.810) [221]. The "corrected item-total correlation" was at least 0.40 for each of the TMA-93 items. Cronbach's alpha did not increase when eliminating any of the ten pairs, discarding any redundancy in the ten drawing pairs. We found a strong correlation between the two halves of the TMA-93, indicating that HCs and aMCI patients performed equally well (or as poorly) on both halves of the test. In the HCs group, the TMA-93 showed a “good” test-retest reliability at 2-4 months [ICC = 0.802 (95% CI = 0.653 - 0.887)], suggesting stability in the over-time performance. The 2-4 months' interval time chosen seems to be short enough to prevent the effect of an eventual cognitive impairment on the sample, particularly from participants with lower scores, and long enough to prevent a practice effect. Discussion 109 By comparison, this reliability is similar to that reported for the "Mini-Mental State Examination" (MMSE) [222]. This property enables the test for use in longitudinal studies. The inter-rater TMA-93 reliability resulted “optimal” for the total score [ICC = 0.999, 95% CI 0.999 - 1], number of errors [ICC = 0.996, 95% CI 0.993 - 0.998], and number of intrusions [ ICC = 0.985, 95% CI 0.974-0.992]. It was “good” for the number of perseverations [ICC = 0.853, 95% CI 0.738 - 0.918]. Administration and scoring are relatively simple, but classifying incorrect responses in errors, intrusions, or perseverations can lead to disagreements between examiners and require some training. Feasibility To demonstrate the test's feasibility, we recorded the percentage of participants who completed the test and the employed administration time. All participants, including those in the mild-dementia stage, completed the test. The task's tolerability was good, including those patients with lower scores or longer administration-time recorded. We found significant differences in the administration time according to the diagnosis: the average time required to complete the test was 2 to 3 minutes for HCs (IQR = 2.0 - 4.0), 6 minutes for aMCI patients (IQR = 4.7 - 7.8), and 7 minutes for AD- like dementia patients (IQR = 5.9 - 9.4). However, there were no significant differences in administration time regarding educational level. An average time of 6 minutes in aMCI Discussion 110 Normative study. Article III "Norms for Testing Visual Binding Using the Memory Associative Test (TMA-93) in Older Educationally-Diverse Adults" This one is the first TMA-93 Spanish normative study. Through it, we provide normative percentiles data. Following international recommendations [208], we describe the healthy population's reference scores, with a broad representation of our region's community in which we are applying the test. We study how sociodemographic variables (age, gender, and educational level) influence the TMA-93 total score. To carry out this study, we followed a systematized recruitment strategy for partners of patients who came to the Memory Outpatient Clinic. Included cases did not suffer from memory difficulties and were cognitively unimpaired. Following routine clinical practice conditions, we did not exclude any participants due to their educational level, reaching a total sample of 1131 participants. TMA-93 total score was influenced by age and educational level, but not by gender. The non-normal distribution of the TMA-93 total score led to a percentile results approximation, with age and educational level stratification. This normative study showed wide variations of the TMA-93 total scores for 5th and 10th percentile by education and age, lower for the older and less educated groups. This distribution suggests that the ability to learn by visual association is lower and more sensitive to aging in the low-educated group. Discussion 111 Compared to its French equivalent (original normative and validation study) [199], the test ran similarly. In both populations, the test works with a ceiling effect mitigated by age and educational level. Regarding the whole sample, the 5th percentile score was slightly higher in the French study. This finding cannot be explained because of a younger sample or a higher educational level. It could be due to a cultural effect. Potential cultural differences corroborate the need for normative studies for each reference population. Discussion 112 Biomarkers’ validation. Article IV "TMA-93 validation by AD biomarkers. A comparison with the FCSRT on a Biobank sample" During the last stage of this doctoral work, we carried out the TMA-93 validation with biomarkers. This work represents the first validation of the test with biomarkers. We performed a retrospective analysis of patient records on a biobank database. The patients included had memory complaints corroborated by an informant, a total score on MMSE equal to or higher than 22, the neuropsychological examination with the TMA-93 and the pictorial FCSRT included; and they had performed a biomarker test (CSF or Amyloid-PET), either with a positive or negative result. To evaluate the diagnostic accuracy and the biomarkers' predictive value (CSF or Amyloid-PET), we compared the TMA-93 and FCSRT variables with the positive or negative biomarkers' result, considered as the Gold-Standard. The TMA-93 total score demonstrated "reasonable" diagnostic utility in discriminating between "positive" and "negative" biomarker groups (AUC = 0.72; 95% CI: 0.62 - 0.82, p <.001). This diagnostic utility was higher than that of the FCSRT variables. According to the DeLong method, we found no significant differences between the TMA-93 and pictorial FCSRT variables. It shows that TMA-93 is as useful as the international Gold-Standard FCSRT to discriminate either patients with memory impairment and MMSE ≥ 22 have positive or negative biomarkers. Discussion 113 The study involved the first validation with biomarkers of the 5th and 10th percentile cutoffs obtained in the previous normative study according to age and educational level. Memory impaired patients with an MMSE ≥ 22 and TMA-93 total score ≤ 5th and 10th percentiles showed 75 and 86% sensitivity and 41 and 29% specificity, respectively, for AD biological diagnosis. These high sensitivity values position the TMA-93 as a good memory screening test, particularly for limited face-to- face time settings. The low specificity is possibly due to the binding component in other non-AD entities, as Argyrophilic grain disease, TDP-43 limbic-predominant age-related encephalopathy, hippocampal sclerosis, and neurofibrillary tangle dementia. Therefore, we propose the patients' memory examination to start with the TMA-93 (as screening), followed by the AD-pathology confirmation with the most specific test at present: biomarkers. The sequential use of the TMA-93 after the pictorial FCSRT increased the diagnostic sensitivity up to 95.3%. However, 4.7% of the evaluated patients obtained a TMA-93 total score above the 10th percentile and FCSRT - TFR over the 21/22 cutoff, a "positive" biomarker result. This 4.7% with positive biomarkers and that was not detected with the cutoff points of TMA-93 and FCSRT represents the real challenge in our daily clinical practice. A high cognitive reserve could play an essential role in those cases, perhaps requiring more demanding memory tests, studying semantic interference, such as the Memory Binding Test, or "conjunctive binding" tests. The main limitation of this validation study was the retrospective analysis. The case study from a Biobank database could have incurred a selection bias. To verify this Discussion 114 hypothesis, a prospective study with systematic recruitment of patients would have to be designed. TMA-93 improved the prediction of biomarker outcomes when added to the FCSRT variables. We could extrapolate that a memory test evaluating binding can improve the biomarker's prediction when added to another test based on coding by semantic clue. Discussion 115 Conclusions Conclusions 117 FINAL CONCLUSIONS 1. The TMA-93 associative visual learning test is highly discriminative to distinguish patients with amnestic Mild Cognitive Impairment without excluding loweducated individuals. 2. TMA-93 has a high internal consistency. All its items measure the interest's construct ("visual relational binding") homogeneously. None of the ten items is redundant. 3. The test is precise, with high interobserver and good test-retest reliability. The good test-retest reliability makes the TMA-93 suitable for longitudinal studies. 4. The TMA-93 administration spends an average of 3 minutes in healthy controls, 6 minutes in amnestic Mild Cognitive Impaired patients, and 7 minutes in Mild Demented patients. The test is suitable for General Medicine and General Neurology outpatient clinics. 5. TMA-93 total score varies with sociodemographic variables. It must be evaluated according to age and educational level. The normative data obtained throughout this doctoral work allow its acceptable use by health staff in Spain. 6. In patients with memory difficulties and MMSE ≥ 22, the TMA-93 is as accurate as the pictorial FCSRT to discriminate between positive and negative AD biomarkers' results. Biomarkers' prediction improves by adding the TMA-93 total score to the FCSRT variables. References 124 25. Bushcke.H, Sliwinski.MJ, Kuslansky.G, Lipton.RB. (1997). Diagnosis of early dementia by the double memory test: encoding specificity improves diagnostic sensitivity and specificity. Neurology. 48(4): 989-997 [DOI: 10.1212/wnl.48.4.989] 26. Cummings JL. (1986) Subcortical Dementia: Neuropsychology, neuropsychiatry, and pathophysiology. Br J Psychiatry. 149, 682-697 [DOI: 10.1192/bjp.149.6.682] 27. Graham. NL, Emery.T and Hodges JR. (2004) Distinctive cognitive profiles in Alzheimer’s disease and subcortical vascular dementia. J Neurol Neurosurg Psychiatry; 75(1):61-71. [PMID: 14707310] 28. Liang.Y, Pertzov.Y, Nicholas JM, Henley SMD, Crutch.S, Woodward.F, Leung.K, Fox.NC, Husain.M. (2016) Visual short-term memory binding deficit in familial Alzheimer’s disese. Cortex 78: 150-164. [DOI: 10.1016/j.cortex.2016.015] 29. Parra. MA, Abrahams.S, Fabi.K, Logie.R, Luzzi.S and Della Sala.S. (2009) Short-term memory binding deficits in Alzheimer’s disease. Brain 132(Pt4): 1057-1066 [DOI: 10.1093/brain/awp036] 30. Shinagawa S (2016) [Language Symptoms of Alzheimer’s disease] Brain Nerve 68(5): 551-557. [DOI: 10.11477/mf.1416200437] 31. Mueller KD, Hermann.B, Mecollari.J, Turkstra.LS. (2018) Connected speech and language in mild cognitive impairment and Alzheimer’s disease: a review of picture description tasks. J Clin Exp. Neuropsychol. 40 (9): 917-939. [DOI: 10.1080/13803395.2018.1446513] 32. Nutter-Upham. KE, Saykin. A, Rabin. L, Roth. R, Wishart HA, Pare. N, Flashman LA. (2008) Verbal fluency performance in amnestic MCI and older adults with cognitive complaints. Arch. Clin Neuropsych. 23(3): 229-241. [DOI: 1016/j.acn.2008.01.005] 33. Lafleche G, Albert MS. (1995) Executive function deficits in Mild Alzheimer’s Disese Neurosychology 9(3): 313-320. References 125 34. Perry RJ, Hodges JR. (1999) Attention and executive deficits in Alzheimer’s disease: A critical review. Brain 122(Pt 3):383-404. [DOI: 10.1093/brain/122.3.383] 35. Suárez-González.A, Henley. S, Walton J, Crutch SJ. (2015) Posterior cortical atrophy: an atypical variant of Alzheimer disease. Psychiatr Clin Nort Am 38(2): 211-220. [DOI: 10.1016/j.psc.2015.01.009] 36. Schott JM and Crutch SJ. (2019) Posterior Cortical Atrophy. Continuum (Minneap Minn) 25 (1): 52-75. [DOI: 10.1212/CON.0000000000000696] 37. Gallagher.D, Coen R, Kilroy. D, Belinski K, Bruce I, Coakley. D, Walsh B, Cunningham C, Lawlor.B. (2011) Anxiety and behavioural disturbance as markers of prodromal Alzheimer’s disease in patients with mild cognitive impairment. Int J Geriatr Psychiatry. 26(2): 166-172. [DOI: 10.1002/gps.2509] 38. Seo. EH, Kim H, Choi KY, Lee KH, Choo IH. (2017) Association of subjective memory complaint and depressive symptoms wit objective cognitive functions in prodromal Alzheimer’s disease including pre-mild cognitive impairment. J. Affect Disord 217: 24- 28. [DOI: 10.1016/j.jad.2017.03.062] 39. Novais F, Starkstein.S (2015) Phenomenology of Depression in Alzheimer’s Disease. J. Alzheimers Dis 47(4): 845-855. [DOI: 10.3233/JAD-148004] 40. Wolinsky D, Drake K, Bostwick J. (2018) Diagnosis and Management of Neuropsychiatric Symptoms in Alzheimer’s Disease. Curr Psychiatry Rep 20(12): 117. [DOI: 10.1007/s11920-018-0978-8] 41. Kawakatsu. S, Kobayashi. R, Hayashi H. (2017) Typical and atpypical appearance of early-onset Alzheimer’s disease: A clinical, neuroimaging and neuropathological study. Neuropathology 37(2): 150-173. [DOI: 10.1111/neup.12364] References 126 42. Tagai. K, Nagata. T, Shinagawa. S, Shigeta. M. (2020) Anosognosia in patients with Alzheimer’s disease: current perspectives. Psychogeriatrics 20(3): 345-352 [DOI: 10.1111/psyg.12507] 43. Starkstein. SE. (2014) Anosognosia in Alzheimer’s disease: diagnosis, frequency, mechanism and clinical correlates. Cortex 61:64-73. [DOI: 10.1016/ j.cortex.2014.07.019] 44. Slachevsky. A, Forno. G, Barraza. P, Mioshi. E, Delgado. C, Lillo. P, Herniquez. F, Bravo. E, Farias. M, Muñoz-Neira. C, Ibañez. A, Parra. MA, HOrnberger.M (2019). Mapping the neuroanatomy of functional decline in Alzheimer’s disease from basic to advanced activities of daily living. J Neurol 266(6), 1310-1322 [DOI: 10.1007/s00415- 019-09260-w] 45. Sanford.AM. (2017) Mild Cognitive Impairment. Clin Geriatr Med. 33 (3): 325-337. [DOI: 10.1016/j.cger.2017.02.005] 46. Petersen. RC (2016) Mild Cognitive Impairment. Continuum (Minneap Minn) 22 (2Dementia): 404-418. [DOI: 10.1212/CON.0000000000000313] 47. Albert. MS, DeKosky. ST, Dickson. D, Dubois B, Feldman. HH, Fox NC, Gamst A, Holtzman. DM, Jagust WJ, Pettersen. 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(3): 270-279 [DOI: 10.1016/j.alz.2011.03.008] 48. Saunders NLJ, Summers. MJ. (2011). Longitudinal deficits to attention, executive, and working memory in subtypes of mild cognitive impairment. Neuropsychology, 25(2), 237-248. [DOI: 10.1037/a0021134] References 127 49. Guías diagnósticas y terapéuticas de la Sociedad Española de Neurología. (2018). Coordinators Sagrario Manzano. M, Fortea.J, Villarejo.A and Sánchez del Valle.R. Ed. Luzan 5. ISBN: 978-84-17372-34-7. PP.11-72 50. Saykin. AJ, Wishart. HA, Rabin.LA, Santulli. RB, Flashman. LA, West. JD, McHugh. TL, Mamourian. AC (2006). Older adults with cognitive complaints show brain atrophy similar to that of amnestic MCI. Neurology 67(5): 834-842. [DOI: 10.1212/ 01.wnl.0000234032.77541.a2] 51. Dubois.B, Feldman.HH, Jacova.C, Dekosky.ST, Barberger-Gateau.P, Cummings.J, Delacourte.A, Galasko.D, Gauthier.S, Jicha.G, Meguro.K, O’brien.J, Pasquier.F, Robert.P, Rossor.M, Salloway.S, Stem.Y, Visser.PJ and Scheltens.P. (2007). Research criteria for the diagnosis of Alzheimer’s disease: revising the NINCDS-ADRDA criteria. Lancet Neurol 6(8): 734-46. [DOI: 10.1016/S1474-4422(07)70178-3] 52. Dubois.B, Feldman.HH, Jacova.C, Hampel.H, Molinuevo.JL, Blennow.K, DeKosky.ST, Gauthier.S, Selkoe.D, Bateman.R, Cappa.S, Crutch.S, Engelborghs.S, Frisoni.GB, Fox.NC, Galasko.D, Habert.MO, Jicha.GA, Nordberg.A, Pasquier.F, Rabinovici.G, Robert.P, Rowe.C, Salloway.S, Sarazin.M, Epelbaum.S, de Souza.LC, Vellas.B, Visser.PJ, Schneider.L, Stern.Y, Scheltens.P and Cummings.JL. (2014). Advancing research diagnostic criteria for Alzheimer’s disease: the IWG-2 criteria. Lancet Neurol 13(6): 614-29. [DOI: 10.1016/S1474-4422(14)70090-0] 53. McKhann.GM, Knopman.DS, Chertkow.H, Hyman.BT, Jack.CR, 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. 7:263-9. [DOI: 10.1016/j.alz.2011.03.005] References 128 54. Jack.CR, Bennett.DA, Blennow.K, Carrillo.MC, Dunn.B, Haeberlein.SB, Holtzman.DM, Jagust.W, Jessen.F, Karlawish.J, Liu.E, Molinuevo.JL, Montine.T, Phelps.C, Rankin.KP, Rowe CC, Shcletens.P, Siemers.E, Snyder.HM and Sperling.R (2018). NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement. 14(4): 535-562. [DOI: 10.1016/j.alz.2018.02.018] 55. Jack. CR, Bennett. DA, Blennow. K, Carrillo. MC, Feldman. HH, Frisoni. GB, Hampel.H, Jagust. WJ, Johnson. KA, Knopman. DS, Petersen. RC, Scheltens. P, Sperling .RA, Dubois. B (2016) A/T/N: An unbiased descriptive classification scheme for Alzheimer disease biomarkers. Neurology 2; 87(5): 539-547.m[Doi: 10.1212/WNL. 0000000000002923] 56. Sabri. O, Sabbagh. MN, Seibyl. J, Barthel. H, Akatsu. H., Ouchi. Y, Senda. K, Murayama. S, Ishii.K, Takao. M, Beach. T.G., Rowe. C.C., Leverenz. J.B., Ghetti.B, Ironside. J.I, Catafau.AM, Stephens. AW. Mueller. A, Koglin. N, Hoffmann. A, Roth. K, Reininger.C, Schulz-Schaeffer. WJ, Florbetaben Phase 3 Study Group. (2015) Florbetaben PET imaging to detect amyloid beta plaques in Alzheimer’s disease: phase 3 study. Alzheimers Dement. 11(8):964-974. [DOI:10.1016/j.jalz.2015.02.00] 57. Rabinovici.GD, Gatsonis.C, Apgar.C, Chaudhary.K, Gareen.I, Hanna.L, Hendrix.J, Hillner.B, Olson.C, Lesman-Segev-OH, Romanoff.J, Siegel.BA, Whitmer.RA and Carrillo.MC. (2019) Association of Amyloid Positron Emission Tomography with subsequent change in clinical management among Medicare beneficiaries with Mild Cognitive Impairment of Dementia. JAMA. 321 (13): 1286-1294. [DOI: 10.1001/jama.2019.2000] 58. Okello. A, Koivunen. J, Edison. P, Archer. HA, Turkheimer. FE, Naàgren. K, Bullock. R, Walker. Z, Kennedy. A, Fox. NC, Rossor. MN, Rinne. JO, Brooks. DJ (2009) References 129 Conversion of Amyloid positive and negative MCI to AD over 3 years: an 11C-PIB PET study. Neurology 8; 73(10): 754-760. [DOI: 10.1212/WNL.0b013e3181b23564] 59. Jack CR, Barrio.JR, Kepe.V (2013) Cerebral amyloid PET imaging in Alzheimer’s disease. Acta Neuropathol 126, 643-657 [DOI: 10.1007/s00401-013-1185-7] 60. Hall. B, Mak. E, Cervenka. S, Aigbirhio. FI, Rowe. JB, O’Brien. JT (2017). In vivo tau PET imaging in dementia: pathophysiology, radiotracer quantification, and a systematic review of clinical findings. Aeging Res Rev 36: 50-63 [DOI: 10.1016/j.arr.2017.03.002] 61. Ossenkoppele. R, Schonhaut. DR, Schöll. M, Lockhart. SN, Ayakta. N, Baker. SL, O’Neil. JP, Janabi. M, Lazaris. A, Cantwell. A, Vogel. J, Santos. M, Miller. ZA, Bettcher. BM, Vossel. KA, Kramer. JH, Gorno-Tempini. ML, Miller. BL, Jagust. WJ, Rabinovici. GD (2016) Tau PET patterns mirror clinical and neuroanatomical variability in Alzheimer’s disease. Brain 139 (5): 1551-1567. [DOI: 10.1093/brain/aww027] 62. Jack. CR, Wiste. HJ, Schwarz. CG, Lowe. VJ, Senjem. ML, Vemuri. P, Wigand. SD, Therneau. TM, Knopman. DS, Gunter. JL, Jones. DT, Graff-Radford. J, Kantarci. K, Roberts. RO, Mielke. MM, Machulda. MM, Petersen. RC (2018). Longitudinal tau PET in ageing and Alzheimer’s disese. Brain 141(5): 1517-1528. [DOI: 10.1093/brain/awy059] 63. Harrison. TM, Maass. A, Adams. JN, Du. R, Baker. SL, Jagust. WJ (2019). Tau deposition is associated with funcitonal isolation of the hippocampus in aging. Nat Commun 10(1): 4900. [DOI: 10.1038/s41467-019-12921-z] 64. Lombardi. G, Crescioli. G, Cavedo. E, Lucenteforte. E, Casazza. G, Bellatorre. AG, Lista. C, Constantino. G, Frisoni. G, Virgili. G, Filippini. G (2020). Structural magnetic resonance imaging for the early diagnosis of dementia due to Alzheimer’s disease in References 130 people with mild cognitive impairment. Cochrane Database Sys Rev 3: CD009628. [DOI: 10.1002/14651858.CD009628.pub2] 65. Kato. T, Inui. Y, Nakamura. A, Ito. K. (2016) Brain fluorodeoxyglucose (FDG) PET in dementia. Ageing Res Rev 30: 73-84. [DOI: 10.1016/j.arr.2016.02.003] 66. Rubi. S, Noguera. A, Tarongí. S, Oporto. M, García. A, Vico. H, Espino. A, Picado. MJ, Mas. A, Peña. C, Amer. G (2018). Concordance between bran 18F-FDG PET and cerebrospinal fluid biomarkers in diagnosing Alzheimer’s disease. Rev Esp Med Nucl Imagen Mol 37 (1): 3-8. [DOI: 10.1016/j.remn.2017.05.003] 67. Saint-Aubert. L, Lemoine. L, Chiotis. K, Leuzy. A, Rodriguez-Vieitez. E, Nordberg. A (2017). Tau PET imaging: present and future directions. Molecular Neurodegeneration 12:19. [DOI: 10.1186/s13024-017-0162-3] 68. Scheltens. P, Blennow. K, Breteler. MM, de Strooper. B, Frisoni. GB, Salloway. S, Van der Flier. WM. (2016). Alzheimer’s disease. Lancet: 388: 505–517. [DOI: 10.1016/s0140-6736(15)01124-1] 69. Hampel. H, Lista. S, Teipel. SJ, Garaci. F, Nisticó. R, Blennow. K, Zetterberg. H, Bertram. L, Duyckaerts. C, Bakardjian. H, Drzezga. A, Colliot. O, Epelbaum. S, Broich. K, Lehéricy. S, Brice. A, Khachaturian. ZS, Aisen. PS, Dubois. B (2014) Perspective on future role of biological markers in clinical therapy trials of Alzheimer’s disease: a long-range point of view beyond. Biochem Pharmacol. 88 (4): 426–449. [DOI: 10.1016/j.bcp.2013.11.009] 70. Blennow. K, Hampel. H, Weiner. M, Zetterberg. H. (2010) Cerebrospinal fluid and plasma biomarkers in Alzheimer disease. Nat Rev Neurol. 6(3): 131–144. [DOI: 10.1038/nrneurol.2010.4] 71. Molinuevo. JL, Ayton. S, Batrla. R, Bednar. M, Bittner. T, Cummings. J, Fagan. AM, Hampel. H, Mielke. MM, Mikulskis. A, O’Bryant. S, Scheltens. P, Sevigny. J, Shaw. References 131 LM, Soares. HD, Tong. G, Trojanowski. JQ, Zetterberg. H, Blennow. K (2018) Current state of Alzheimer’s fluid biomarkers. Acta Neuropathol. 136(6): 821-853 [Doi: 10.1007/s00401-018-1932-x 72. Ritter. A, Cummings. J. (2015) Fluid biomarkers in clinical trials of Alzheimer’s disease therapeutics. Front Neurol. 6:186. [DOI: 10.3389/fneur.2015.00186] 73. Tapiola. T, Alafuzoff. I, Herukka. SK, Parkkinen. L, Hartikainen. P, Soininen. H, Pirttilä. T (2009) Cerebrospinal fluid β-amyloid 42 and tau proteins as biomarkers of Alzheimer-type pathologic changes in the brain. Arch Neurol. 66(3): 382–389. [DOI: 10.1001/archneurol.2008.596] 74. Toledo. JB, Xie. SX, Trojanowski. JQ, Shaw. LM. (2013) Longitudinal change in CSF tau and Aβ biomarkers for up to 48 months in ADNI. Acta Neuropathol. 126 (5): 659– 670. [DOI: 10.1007/s00401-013-1151-4] 75. McGowan. E, Pickford. F, Kim. J, Onstead. L, Eriksen. J, Yu. C, Skipper. L, Murphy. MP, Beard. J, Das. P, Jansen. K, DeLucia. M, Lin. WL, Dolios. G, Wang. R, Eckman. CB, Dickson. DW, Hutton. M, Hardy. J, Golde.T. (2005) Aβ42 is essential for parenchymal and vascular amyloid deposition in mice. Neuron. 47(2):191–199. [DOI: 10.1016/j.neuron.2005.06.030] 76. Pannee. J, Portelius. E, Minthon. L, Gobom. J, Andreasson. U, Zetterberg. H, Hansson. O, Blennow. K (2016) Reference measurement procedure for CSF amyloid beta (Aβ)1- 42 and the CSF Aβ1-42/Aβ1-40 ratio - a cross-validation study against amyloid PET. J Neurochem. 139(4): 651–658. [DOI: 10.1111/jnc.13838] 77. Blennow. K, Hampel. H, Weiner. M, Zetterberg. H. (2010) Cerebrospinal fluid and plasma biomarkers in Alzheimer disease. Nat Rev Neurol. 6(3): 131–144. [DOI: 10.1038/nrneurol.2010.4] References 132 78. Hampel. H, Blennow. K, Shaw. LM, Hoessler. YC, Zetterberg. H, Trojanowski. JQ (2010) Total and Phosphorylated Tau protein as biological markers of Alzheimer’s disease. Exp Gerontol 45(1): 30 [Doi: 10.1016/j.exger.2009.10.010] 79. Fagan. AM, Shaw. LM, Xiong. C, Vanderstichele. H, Mintun. MA, Trojanowski. JQ, Coart. E, Morris. JC, Holtzman. DM. (2011) Comparison 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(9): 1137–1144. [DOI: 10.1001/archneurol.2011.105] 80. Olsson. B, Lautner. R, Andreasson. U, Öhrfelt. A, Portelius. E, Bjerke. M, Hölttä. M, Rosén. C, Olsson. C, Strobel. G, Wu. E, Dakin. K, Petzold. M, Blennow. K, Zetterberg. H (2016). CSF and blood biomarkers for the diagnosis of Alzheimer's disease: a systematic review and meta-analysis. Lancet Neurol. 15(7): 673-684. [DOI: 10.1016/S1474-4422(16)00070-3] 81. Blennow.K, Zetterberg.H (2018) The past and the future of Alzheimer’s disease fluid biomarkers. J Alzheimers Dis 62(3): 1125-1140 [DOI: 10.3233/JAD-170773] 82. RL Friede, T Samorajski (1970) Axon caliber related to neurofilaments and microtubules in sciatic nerve fibers of rats and mice. Anat Rec, 167:379-387 [DOI: 10.1002/ar.1091670402] 83. D Bonneh-Barkay. D, Bissel. SJ, Wang. G, fish. KN, Nicholl. GCB, Darko. SW, Medina-Flores. R, Murphey-Corb. M, Rajakumar. PA, Nyaundi. J, Mellors. JW, Bowser. R, Wiley. CA. (2008) YKL-40, a marker of simian immunodeficiency virus encephalitis, modulates the biological activity of basic fibroblast growth factor. Am J Pathol, 173(1): 130-143 [DOI: 10.2353/ajpath.2008.080045] 84. Galasko. D, Xiao. M, Xu. D, Smirnov. D, Salmon. DP, Dewit. N, Vanbrabant. J, Jacobs. D, Vanderstichele. H, Vanmechelen. E, Alzheimer’s Disease Neuroimaging Initiative References 133 (ADNI), Worley. P (2019) Synaptic biomarkers in CSF aid in diagnosis, correlate with cognition and predict progression in MCI and Alzheimer’s disease. Alzheimers Dement (NY) 5: 871-882 [DOI: 10.1016/j.trci.2019.11.002] 85. Faura. J, Bustamante. A, Penalba. A, Giralt. D, Simats. A, Martínez-Sáez. E, Alcolea. D, fortea. J, Lleó. A, Teunissen. CE, Van der Flier. WM, Ibañez. L, Harari. O, Cruchaga. C, Hernández-Guillamón. M, Delgado. P, Montaner. J. (2020) CCL23: A chemokine associated with progression from Mild Cognitive Impairment to Alzheimer’s Disease. J Alz Dis. 73(4): 1585-1595. [DOI: 10.3233/JAD-190753] 86. Gao. H, Liu. M, Zhao. Z, Yang. C, Zhu. L, Cai. Y, Yang. Y, Hu. Z (2020). Diagnosis of Mild Cognitive Impairment and Alzheimer’s Disease by the plasma and serum Amyloid-beta 42 assay through highly sensitive peptoid nanosheet sensor. ACS Appl Mater Interfaces 12(8): 9693-9700. [DOI: 10.1021/acsami.0c00370] 87. Fiandaca. MS, Kapogiannis. D, Mapstone. M, Boxer. A, Eitan. E, Schwartz. JB, Abner. EL, Petersen. RC, Federoff. HJ, Miller. BL, Goetzl. EJ (2015) Identification of preclinical Alzheimer’s disease by a profile of pathogenic proteins in neurally derived blood exosomes: A case-control study. Alzheimers Dement 11: 600–607 [DOI: 10.1016/j.jalz.2014.06.008] 88. Mattsson. N, Andreasson. U, Zetterberg. H, Blennow. K. Alzheimer’s Disease Neuroimaging Initiative (2017) Association of plasma Neurofilament Light With Neurodegeneration in patients with Alzhiemer Disease. JAMA Neurol 74(5): 557-566. [DOI: 10.1001/jamaneurol.2016.6117] 89. Li. D. Hagen. C, Fett. AR, Bui. HH, Knopman. D, Vemuri. P, Machulda. MM, Jack Jr. CR, Petersen. RC, Mielke. MM (2019) Longitudinal association between phosphatidylcholines, neuroimaging measures of Alzheimer’s disease pathophysiology, References 140 dementia severity. J Am Geriatr Soc 37(8): 725-729 [DOI: 10.1111/j.1532- 5415.1989.tb02233.x] 133. Wolf-Klein. GP, Silverstone. FA, Levy. AP, Brod.MS (1989) Screening for Alzheimer’s disease by clock drawing. J Am Geriatr Soc 37:730-734 [DOI: 10.1111/j.1532-5415.1989.tb02234.x] 134. Royall. DR, Cordes. JA, Polk. M (1998) CLOX: an executive clock drawing task. J Neurol Neurosurg Psychiatry. 64(5): 588-594 [Doi: 10.1136/jnnp.64.5.588] 135. Cacho-Gutiérrez. LJ, Garcia-Garcia. R, Arcaya-Navarro. JL., Vicente-Villardón. JL, Lantada-Puebla. N. (1999). Una propuesta de aplicacion y puntuacion del test del reloj en la enfermedad de Alzheimer. Rev Neurol 28:648-55. 136. Crodaty. H, Moore. CM. (1997) The clock drawing test for dementia of the Alzheimer’s type: a comparison of three scoring methods in a memory diseorders clinic. Intern J Geriatr Psychiatry 12(6): 619-627 [PMID: 9215942] 137. Shulman. KI. (2000) Clock drawing: is it the ideal cognitive Screening test? Int J Geriatr Psychiatry 15(6): 548-561 [DOI: 10.1002/1099-1166(200006)15:6<548::aidgps242>3.0.co;2-u] 138. Cacho. J, Benito-Leon. J, Garcia-Garcia. R, Fernandez-Calvo. B, Vicente-Villardon. JL, Mitchell. AJ. (2010) Does the combination of the MMSE and clock drawing test improve the detection of mild Alzheimer's disease and mild cognitive impairment? J Alz Dis. 22(3): 889-896 [DOI: 10.3233/JAD-2010-101182] 139. del Ser Quijano. T, Garcia de Yebenes. MJ, Sanchez-Sanchez. F, Frades Payo. B, Rodriguez Laso. A, Bartolomé Martínez. MP, Otero Puime. A (2004). Cognitive assessment in the elderly. Normative data of a Spanish population simple oder tan 70 years. Med Clin (Barc). 122(19): 727-740 [Doi. 10.1157/13062190] References 141 140. del Ser Quijano. T, Sanchez-Sanchez. F, Garcia de Yebenes. MJ, Otero Puime. A, Zunzunegui. MV, Munoz. DG (2004) Spanish versión of the 7 Minute screening neurocognitive battery. Normative data of an elderly population simple over 70. Neurologia 19(7): 344-358 [PMID: 15273881] 141. Ijuin. M, Homma. A, Mimura. M, Kitamura. S (2008) Validation of the 7-Minute Screen for the Detection of Early Stage Alzheimer’s Disease. Dement and Geriat Dis 25(3): 248-255 [DOI: 10.1159/000115972] 142. del Ser. T, Sanchez-Sanchez. F, Garcia de Yebenes. MJ, Otero. A, Munoz. DG. (2006) Validation of the seven-minute screen neurocognitive battery for the diagnosis of dementia in a Spanish population-based sample. Dement Geriatr Cogn Disord. 22(5-6): 454-464 [DOI: 10.1159/000095858] 143. Meulen. EFJ, Schmand. B, Van Campen. JP, de Koning. SJ, Ponds. RW, Scheltens. P, Verhey. FR (2004) The seven minute screen: a neurocognitive screening test highly sensitive to various types of dementia. J Neurology, Neurosurgery & Psychiatry, 75: 700-705. [DOI: 10.1136/jnnp.2003.021055] 144. Carnero-Pardo. C, Gurpegui. M, Sanchez-Cantalejo. E, Frank. A, Mola. S, Barquero. MS, Montoro-Rios. M, Trans-EUROTEST Group (2006) Diagnostic accuracy of the Eurotest for dementia: a naturalistic, multicenter phase II study. BMC Neurol 6: 15. [DOI: 10.1186/1471-3477-6-15] 145. Carnero-Pardo. C, Gurpegui. M, Sánchez-Cantalejo. E, Camello. AG, Santervás. PG, Ríos. MT, (2007) Reliability assessment of the EUROTEST. Neurologia 22(3): 153- 158. [PMID: 17364253] 146. Pesset. V, Puig. C, Martínez-Ruiz. E, Mazón. J, Menéndez. B, Lacruz Ballester. L, Lluch. V, Lominchar-Espada. J (2007) Experiences with the Eurotest in References 142 neuropsychological examinations. A screening test. Rev Neurol 45(1): 22-26 [PMID: 17620262] 147. Smyer. MA, Hofland. BF, Jonas. EA (1979) Validity study of the short portable mental status questionnaire for the elderly. J Am Geriatr Soc 27(6): 263-269 [DOI: 10.1111/j.1532-5415.1979.tb06128.x] 148. Wolber. G, Romaniuk. M, Eastman. E, Robinson.C (1984) Validity of the short portable Mental Status Questionnaire with Elderly Psychiatric Patients. J Consulting and Clinical psychology. 52(4): 712-713 [DOI: 10.1037//0022-006x.52.4.712] 149. Erkinjuntti. T, Sulkava. R, Wikström. J, Autio. L (1987) Short Portalbe Mental Status Questionnaire as a Screening Test for Dementia and Delirium Among the Elderly. J Am Ger Society 35(5). [Doi: 10.1111/j.1532-5415.1987.tb04662.x] 150. Martinez de la Iglesia. J, Duenas Herrero. R, Onis Vilches. MC, Aguado Taberné. C, Albert Colomerc C, Luque Luque. R. (2001) Spanish language adaptation and validation of the Pfeiffer’s questionnaire (SPMSQ) to detect cognitive deterioration in people over 65 years of age. Med Clin (Barc) 2001;117(4): 129-134 [DOI: 10.1016/s0025-7753(01)72040-4] 151. Carnero Pardo. C, Sáez-Zea. C, Montiel Navarro. L, Del Sazo. P, Feria Vilar. I, Pérez Navarro. MJ, Ruiz Jiménez. J, Vilchez Carrillo. R Montoro Ríos. MT (2007) Diagnostic accuracy of the Phototest for cognitive impairment and dementia. Neurologia. 22(10): 860-869 [PMID: 18040903] 152. Carnero-Pardo. C, Espejo-Martinez. B, Lopez-Alcalde. S, Espinosa-Garcia. M, Saez- Zea. C, Hernandez-Torres. E, Navarro-Espigares. JL, Vilchez-Carrillo. R (2011) Diagnostic accuracy, effectiveness and cost for cognitive impairment and dementia screening of three short cognitive tests applicable to illiterates. PLoS One, 6(11):e27069 [DOI: 10.1371/journal.pone.0027069] References 143 153. Carnero-Pardo. C, Saez-Zea. C, Montiel-Navarro. L, Feria-Vilar. I, Gurpegui. M (2011) Normative and reliability study of fototest. Neurologia, 26(1):20-25 [DOI: 10.1016/j.nrl.2010.09.021] 154. Carnero-Pardo. C, Espejo-Martínez. B, López-Alcalde. S, Espinosa-García. M, Saez- Zea. C, Vilchez-Carrillo. R, Hernández-Torres. E, Navarro-Espigares. JL (2011) Effectiveness and costs of phototest in dementia and cognitive impairment screening. BMC Neurol 11: 92 [DOI: 10.1186/1471-2377-11-92] 155. Rami. L, Bosch. B, Sanchez-Valle. R, Molinuevo. JL (2010) The Memory Alteration Test (M@T) discriminates between subjective memory complaints, mild cognitive impairment and Alzheimer’s disease. Arch Gerontol Geritr 50(2): 171-174 [DOI: 10.1016/j.archger.2009.03.005] 156. Custodio. N, Lira. D, Herrera-Perez. E, Montesinos. R, Castro-Suarez. S, Cuenca- Alfaro. J, Valeriano-Lorenzo. L (2017) Memory Alteration Test to detect Amnestic Mild Cognitive Impairment and Early Alzheimer’s Dementia in Population with Low Educational Level. Front Aging Neurosci 9: 278 [DOI: 10.3389/fnagi.2017.00278] 157. Delis. DC, Kramer. JH, Kaplan. E, Ober. BA (1987) California Verbal Learning Test: adult version. Manual. Psychological Corporation, San Antonio, TX. 158. Rey. A (1958). L’examen clinique en psychologie. Paris: Presse Universitaire de France. 159. Morris. JC, Heyman. A, Mohs. RC, Hughes. JP, Van Belle. G, Fillenbaum. G, Mellits. ED, Clark. C (1989) The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part I. Clinical and neuropsychological assessment of Alzheimer’s disease. Neurology 39(9): 1159-1165 [DOI: 10.1212/wnl.29.9.1159] References 144 160. Benedet. MJ, Alejandre. MA. (2014) Test de Aprendizaje Verbal España-Complutense. Ed. TEA Ediciones, SAU. Fray Bernardino Sahagún, 24. 28036 Madrid. España. ISBN. 978-84-16231-00-1. 161. Elwood. RW (1995) The California Verbal Learning Test: Psychometric Characteristics and Clinical Application. Neuropsychology Review, 5(3) 173-201 [DOI: 10.1007/BF02214761] 162. Woods. SP, Delis. DC, Scott. JC, Kramer. JH, Holdnack. JA (2006) The California Verbal Learning Test – second edition: Test-retest reliability, practice effects, and reliable change indices for the standard and alternate forms. Arc Clin Neuropsychology. 21(5): 413-420. [DOI: 10.1016/j.acn.2006.06.002] 163. Campos-Magdaleno. M, Facal. D, Juncos-Rabadán. O, Braña. T, Pereiro. AX (2014) Cluster subtypes of the Spanish version of the California Verbal Learning Test in a simple of adults with subjective memory complaints. J Clin & Experimental Neuropsychology. 36(3): 317-333 [DOI: 10.1080/13803395.2014.890698] 164. Estévez-González. A, Kulisevsky. J, Boltes. A, Otermín. P, García-Sánchez. C (2003) Rey verbal learning test is a useful tool for differential diagnosis in the preclinical phase of Alzheimer’s disease: comparison with mild cognitive impairment and normal aging. Int J Geriatric Psychiatry 18: 1021-1028 [Doi: 10.1002/gps.1010] 165. Strauss. ME, Fritsch. T (2004) Factor structure of the CERAD neuropsychological battery. J Int Neuopsychol Soc Jul; 10(4): 559-565 [DOI: 10.1017/ S1355617704104098] 166. Sudo. FK, De Souza. AS, Drummond. C, Assuncao. N, Teldeschi. A, Oliveira. N, Rodrigues. F, Santiago-Bravo.G, Calil. V, Lima. G, Erthal. P, Bernardes. G, Monteiro. M, Tovar-Moll. F, Mattos. P (2019) Inter-method and anatomical correlates of episodic References 145 memory tests in the Alzheimer’s Disease spectrum. PloS One 14(10: e0223731 [DOI: 10.1371/journal.pone.0223731] 167. Wechsler, D (1945) A standardized memory scale for clinical use. The Journal of Psychology: Interdisciplinary and Applied, 19(1), 87-95 [DOI: 10.1080/00223980.1945.9917223] 168. Hartman. DE (2009) Wechsler Adult Interlligence Scale IV (WAIS IV): return of the gold standard. Appl Neuropsychol. 16(1): 85-87 [DOI: 10.1080/09084280802644466] 169. Wechsler. D (2013) WMS-IV. Escala de memoria de Wechsler-IV Manual técnico y de interpretación. Madrid: NCS Pearson, Inc. (Ed. Original, 2008, NCS Pearson Inc.). 170. Wechsler. D (2013) WMS-IV. Escala de memoria de Wechsler-IV Manual de administración y corrección. Madrid: NCS Pearson, Inc. (Ed. Original, 2008, NCS Pearson Inc.). 171. Peña-Casanova. J (1990) Programa integrado en la exploración neuropsicológica. Test Barcelona. Manual. Barcelona: Masson. 172. Peña-Casanova. J (1991) Normalidad, semiología y patología neuropsicológicas. Programa Integrado de Exploración Neuropsicológica. Test Barcelona. Barcelona: Masson. 173. Quintana Aparicio. M (2009) Test Barcelona Abreviado: Datos normativos. Aproximación desde la Teoría de Respuesta a los ítems y redes neuronales artificiales en el deterioro cognitivo leve y enfermedad de Alzheimer. Tesis Doctoral Universitat Autònoma de Barcelona. Facultad de Medicina. Departamento de Psiquiatría y Medicina Legal. Director: Dr. Jordi Peña-Casanova. 174. Buschke H. (1984) Cued recall in amnesia. J Clin Exp Neuropsychol 6(4): 433– 440 [DOI: 10.1080/01688638408401233] References 146 175. Grober. E, Ocepek-Welikson. K, Teresi. JA. (2009) The Free and Cued Selective Reminding Test: Evidence of psychometric adequacy. Psychol Sci Q. 51. 266-28 176. Ivnik. RJ, Smith. GE, Lucas. JA, Tangalos. EG, Kokmen. E, Petersen. RC (1997) Free and cued selective reminding test: Moans norms. J Clin and Experimental Neuropsychology. 19(5): 676-691 [Doi: 10.1080/01688639708403753] 177. Peña-Casanova. J, Gramunt-Fombuena. N, Quiñones-Úbeda. S, Sánchez-Benavides. G, Aguilar. M, Badenes. D, Molinuevo. JL, Robles. A, Sagrario Barquero. M, Payno. M, Antúnez. C, Martínez-Parra. C, Frank-García. A, Fernández. M, Alfonso. V, Sol. JM, Blesa. R. NEURONORMA Study Team (2009) Spanish Multicenter Normative Studies (NEURONORMA Project): Norms for the Rey-Osterrieth Complex Figure (Copy and Memory), and Free and Cued Selective Reminding Test. Archives of Clinical Neuropsychology, 24(4): 371-393. [Doi: 10.1093/arclin/acp041] 178. Lemos. R, Simoes. MR, Santiago. B, Santana. I (2014) The Free and Cued Selective Remiding Test: validation for mild cognitive impairment and Alzheimer’s disease. J Neuropsychol 9(2): 242-257 [Doi: 10.1111/jnp.12048] 179. Grober. E, Merling. A, Heimlich. T, Lipton. RB (1997) Free and cued selective reminding and selective reminding in the elderly. J Clin Experimental Neuopsychology. 19(5): 643-654. [Doi: 10.1080/01688639708403750] 180. Grober. E, Lipton. RB, Hall. C, Crystal. H (2000) Memory impairment of free and cued selective reminding predicts dementia. Neurology 54(4). [Doi: 10.1212/WNL.54.4.827] 181. Peña-Casanova. J, Gramunt Fombuena. N, Gich Fullà. J (2004) Test neuropsicológicos. Fundamentos para una neuropsicología clínica basada en evidencias. Ed. Masson. ISBN: 13:978-458-1468-0 References 147 182. Naveh-Benjamin. M. (2000) Adult age differences in memory performance: Tests of an associative deficit hypothesis. Journal of Experimental Psychology: Learning, Memory, and Cognition, 26(5), 1170–1187 [Doi: 10.1037/0278-7393.26.5.1170]. 183. Naveh-Benjamin. M., Guez. J., Kilb. A, Reedy. S (2004) The Associative Memory Deficit of Older Adults: Further Support Using Face-Name Associations. Psychology and Aging, 19(3), 541–546 [Doi: 10.1037/0882-7974.19.3.541]. 184. Parra. MA, Abrahams. S, Fabi. K, Logie. R, Luzzi. S, Della Sala. S (2009) Short-term memory binding déficits in Alzheimer’s disease. Brain 2009 132 (4) 1057-1066 [DOI: 10.1093/brain/awp036] 185. Gramunt. N, Sánchez-Benavides. G, Buschke. H, Lipton. RB, Masramon. X, Gispert. JD, Peña-Casanova. J, Fauria. K, Molinuevo. JL (2016) Psychometric Properties of the Memory Binding Test: Test-Retest Reliability and Convergent Validity. J Alz Dis. 50: 999-1010. [PubMed: 26836167] 186. Buschke. H (2014) The rationale of the Memory Binding Test. Dementia and Memory. Nilsson LG., Ohta. N, eds. Psychology Press. New York, 55-71. 187. Mowrey. WB, Lipton. RB, Katz. MJ, Ramratan. WS, Loewenstein. DA, Zimmerman. ME, Buschke. H (2018) Memory Binding Test predicts incident dementia: results from the Einstein aging study J Alzheimers Dis 62. 293-304 [DOI: 10.3233/JAD-170714] 188. Gramunt. N, Sánchez-Benavides. G, Buschke. H, Diéguez-Vide. F, Peña-Casanova. J, Masramon. X, Fauria. K, Gispert. JD, Molinuevo. JL (2016) The Memory Binding Test: development of two alternate forms into Spanish and Catalan. J Alz Dis 52: 283-293 [Doi: 10.3233/JAD-151175] 189. Paddick. SM, Gray. WK, McGuire. J, Richardson. J, Dotchin. C, Walker. RW (2017) Cognitive screening tools for identification of dementia in illiterate and low-educated References 148 older adults, a systematicreview and metaanalysis. International psychogeriatrics. 29 (6): 897-929 [DOI: 10.1017/S1041610216001976] 190. Rey. A (1941) L’examen psychologique dans les cas d’encephalopathie traumatique. Archives de Psychologie, 28: 286-340. 191. Osterrieth. PA (1944) Le test de copie d’une figure complexe. Archives de Psychologie. 30: 206-356. 192. Barbeau. E, Didic. M, Tramoni. E, Felician. O, Joubert. S, Sontheimer. A, Ceccaldi. M, Poncet. M (2004) Evaluation of visual recognition memory in MCI patients. Neurology 62(8): 1317-1322 [Doi: 10.1212/01.wnl.0000120548.24298.db] 193. Meunier. M, Bachevalier. J, Mishkin. M, Murray. EA. (1993) Effects on visualrecognition of combined and separate ablations of the entorhinal andperirhinal cortex in rhesus monkeys. J Neurosci. 13(12): 5418–5432 [DOI: 10.1523/JNEUROSCI.13-12-05418.1993] 194. Delacourte. A, David. JP, Sergeant. N, Buée. L, Wattez. A, Vermersch. P, Ghozali. F, Fallet-Bianco. C, Pasquier. F, Lebert. F, Petit. H, Di Menza. C (1999). The biochemical pathway ofneurofibrillary degeneration in aging and Alzheimer’s disease. Neurology.52(6): 1158–1165 [DOI: 10.1212/wnl.52.6.1158] 195. Rullier. L, Matharan. F, Barbeau. EJ, Mokri. H, Dartigues. JF, Pérès. K, Amieva. H (2014) [The DMS 48: norms and diagnostic properties for Alzheimer’s disease in elderly population from the AMI cohort sutdy] Geriatr Psychol Neuropsychiatr Vieil. 12(3): 321 – 330 [Doi: 10.1684/pnv.2014.0486] 196. Feng. X, Zhou. A, Liu. Z, Li. F, Wei. C, Zhang. G, Jia. J (2018) Validation of the Delayed Matching-to-Sample Task 48 (DMS48) in Elderly Chinese. J Alz Dis 61: 1611- 1618 [Doi: 10.3233/JAD-170530] References 149 197. Zimmerman. ME, Katz. MJ, Wang. C, Burns. LC, Berman. RM, Derby. CA, LÍtalien. G, Budd. D, Lipton. RB (2012) Comparison of “Word” vs. “Picture” version of the Free and Cued Selective Reminding Test (FCSRT) in older adults. Cognitive & Behavioral Assessment. 1(1): 94-100. [Doi: 10.1016/j.dadm.2014.11.006] 198. Delgado. C; Muñoz-Neira. C, Soto. A, Martínez. M, Henríquez. F, Flores. P (2016) Comparison of the psychometric properties of the “Word” and “Picture” versions of the Free and Cued Selective Reminding Test in a Spanish-Speaking Cohort of Patients with Mild Alzheimer’s Disease and cognitively healthy controls. Arch Clin Neuropsychology. 31(2): 165-175. [Doi: 10.1093/arclin/acv107] 199. Maillet. D, Narme. P, Amieva. H, Matharan. F, Bailon. O, Le Clésiau. H, Belin. C (2017) The TMA-93: A New Memory Test for Alzheimer’s Disease in Illiterate and Less Educated People. Am J Alzheimers Dis Other Dement, 32(8): 461-467 [DOI: 10.1177/1533317517722630] 200. Kosack. CS, Page. AL, Klatser. P (2017) A guide to aid the selection of diagnostic tests. Bulletin of the World Health Organization. 95: 639-645 [Doi: 10.2471/BLT.16.187468] 201. Bossuyt. PMM, Reitsma. JB, Linnet. K, Moons. KGM. (2012) Beyond diagnostic accuracy: the clinical utility of diagnostic tests. Clin Chem. 58(12): 1636–43. [DOI: 10.1373/clinchem.2012.182576 pmid: 22730450] 202. Sackett. DL, Haynes. RB, Guyatt. GH, Tugwell. P (1991) The selection of diagnostic tests. In: Sackett D, editor. Clinical epidemiology.Boston, Toronto, and London: Little, Brown and Company: 47–57. 203. Carnero-Pardo. C (2005) Evaluación de las pruebas diagnósticas. Rev Neurol 2005; 40(11): 641-643. [DOI: 10.33588/rn.4011.2005288] 204. Bossuyt. PM, Reitsma. JB, Bruns. DE, Gatsonis. CA, Glasziou. PP, Irwig. LM, Lijmer. JG, Moher. D, Rennie. D, de Vet. HCW. Standars for Reporting of Diagnostic Accuracy