VALUTAZIONE DELL'ACCURATEZZA DIAGNOSTICA DI TRE TEST PER LA MALATTIA DI ALZHEIMER: MEMORY BINDING, FREE AND CUED SELECTIVE REMINDING E DUAL TASK

UNIVERSITÀ DEGLI STUDI DI MILANO Facoltà di Medicina e Chirurgia Dipartimento di Scienze Cliniche “Luigi Sacco” SCUOLA DI DOTTORATO Scienze fisiopatologiche, neuropsicologiche e assistenziali del ciclo di vita-Ciclo XXVII Tesi di Dottorato di Ricerca: Valutazione dell’accuratezza diagnostica di tre test per la Malattia di Alzheimer: Memory Binding, Free and Cued Selective Reminding Test e Dual Task Settore Scientifico Disciplinare Med – 26 Dottorando: Dr.ssa Elena SARTORIO Matricola: R09555 Tutor: Chiar.mo Prof. Claudio MARIANI Coordinatore: Chiar.mo Prof. Roberto L. WEINSTEIN Anno Accademco 2013/2014 Background and hyphothesis The main focus of current Alzheimer’s Disease (AD) research is on early diagnosis. This emphasis has led to the proposal of new research diagnostic criteria, which stress the potential role of biomarkers[Dubois B, 2014; Jack C.R. Jr, 2011 ] in diagnosing the disease in a prodromal, or even in a preclinical stage. While a considerable amount of effort is now dedicated to crucial issues about the use of biomarkers, including the standardization of CSF analysis[Molinuevo J.L, 2014] and of neuroimaging techniques [Cure F, 2014], studies on the cognitive markers are relatively scant. This is surprising, considering that the only clinical criterion supporting the use of biomarkers in subjects with cognitive complaints is the presence of an objective cognitive dysfunction, as assessed by neuropsychological tests. Of paramount importance is the assessment of memory, as the impairment in episodic memory is most commonly seen in subjects with Mild Cognitive Impairment (MCI) who subsequently progress to a diagnosis of AD dementia[Albert M.S, 2011]. Notwithstanding the obvious relevance of this point for any validation of the proposed biomarkers, it is unfortunate that no "gold standard" is available for memory testing of AD. There is clearly the need for tests that are specific for the memory dysfunction of prodromal and early AD, and that are predictive of progression to dementia. We have identified from the literature and from our own preliminary work three tests, which appear to be promising candidates to become "gold standards" for the cognitive assessment of AD. Two tests, Short Term Memory Binding Task (STMBT) and Dual Task (DT) were developed by Human Cognitive Neuroscience, Centre for Cognitive Ageing and Cognitive Epidemiology, Psychology, University of Edinburgh, UK. The first is the STMBT. The impairment of the mechanisms responsible for holding integrated objects in verbal short-term memory was shown to be selectively impaired in mild and moderate AD, in comparison to healthy ageing [Brockmole, J.R., 2008] and chronic depression in the elderly [Parra MA, 2010]. The second is the DT [Baddeley A, 1986; Baddeley A, 1991, Della Sala S, 2010]. Performance on the DT has been shown to effectively discriminate AD from healthy aging and depression [Kaschel R]. Task performance appears to be preserved in subjects with a diagnosis of MCI, suggesting that the test is suitable to map disease progression, rather than for early diagnosis [Foley J.A, 2011]. The third test is the FCSRT [Grober E,1988]. This traditional memory test has been proposed to have high predictive value for the progression of isolated disorders of memory towards AD. In particular, Dubois and colleagues [Dubois B, 2007] have claimed that episodic memory loss of the hippocampal type (characterised by a free recall deficit on testing not normalised with cueing) may be the defining feature of prodromal AD. This hypothesis is supported by a study, which indicated the predictive role of defective performance on the FCSRT on progression to dementia in subjects fulfilling the criteria for MCI [Sarazin M., 2007]. Metodologies and statistical analyses Screening procedure: All participants underwent a diagnostic protocol that included: -medical and neurological examinations, -a functional examination with Activities of Daily Living (ADL), Instrumental Acivities of Daily Living (IADL) and Clinical Dementia Rating to assess the dementia severity staging (CDR); - a depression scale Geriatric Depression Scale (GDS) for MCI group and controls; -a neuropsychological examination with the following standardized measures: 1. Global cognitive function: Minimental State Examination (MMSE) 2. Language: Token Test and semantic fluency 3. Memory: Forward Digit Span Test and story recall. 4. Executive function tests: attentional matrices and Frontal Behavioural Inventory. The Exclusion criteria are: a) organic brain pathology or organic illness affecting the brain according to the ICD-10; b)significant history of head injury; c) major systemic illnesses or medical complications, including vitamin deficiency states, thyroid disorders, and sensory disorders (i.e., blindness or deafness); d) history of drug or alcohol dependence; e) structural brain alterations that included mass lesions and hydrocephalus. The inclusion criterion is the diagnosis of dementia/Major Neurocognitive Disorder [APA, 1994] for AD sample and the diagnosis of amnestic (single or multiple-domain) MCI [Winblad B, 2004] for MCI. The samples will be composed as follows: 21 patients with mild (MMSE score equal to or higher than 21) AD dementia 21 with moderate (MMSE score 14-20) AD dementia with a typical amnestic onset [APA, 1994] and CDR > 0.5; 99 patients with MCI (24 amnestic MCI; 15 single domain MCI and 60 multi-domain MCI) with ADL 6/6; IADL lost functions ≤ 2, CDR=0.5; 52 healthy controls, aged over 65 years. Experimental procedures Participants to the study will undergo the experimental tests. The FCSRT is a paper-and-pencil test which assesses the sensitivity to cueing [Frasson P, 2011]. The STMBT (which uses the change detection paradigm) [Parra M.A, 2009] and the DT (based on the dual task paradigm) [Della Sala S, 2010] are computerized tests. Statistical analysis and results The scores on the neuropsychological battery were compared across groups using one-way ANOVA followed by Bonferroni-corrected post hoc tests. The anagraphic and neuropsychological characteristics of the studied populations are presented as follows: Subjects (N) : Control 52, MCI 99, AD 42; Age, mean (±DS): Controls 72.4 (5.3), MCI 75.5 (6.2), AD 78.5 (6.2), p <0.001; Gender, female number (%):Controls 28 (53.8), MCI 57(57.6), AD 23 (54.8), ns; Education, mean (±DS): Controls 11.4 (4.9), MCI 8.9 (4.6), AD 8.9 (4.9), p= 0,006; MMSE score, mean (±DS): Controls 28.1 (1.9), MCI 24.8 (2.9), AD 20.8 (3.4), p<0.001; Depression, n (%): Controls 7 (13.5), MCI 34 (34.3), AD 11 (26.2), p=0.023; CDR score, mean (±DS): Controls 0.0 (0.0), MCI 0.5 (0.0), AD 1.1 (0.4), p<0.001; ADL lost functions, mean (±DS):Controls 0.0 (0.1), MCI 0.2 (0.4), AD 1.0 (1.5), p<0.001; IADL lost functions, mean (±DS): Controls 0.1 (0.3),MCI 0.4 (0.8), AD 3.2 (2.1), p<0.001. We could not use the DT data in this statistical analysis because of a software errors in management of results but we are trying to retrieve the data. MCI subjects performed similarly to healthy controls but significantly better than AD patients in the condition assessing memory for shape only (STM for shape accuracy: controls 0.92 ±0.09, MCI 0.83±0.14, AD 0.72 ±0.16. Post-hoc analysis: controls vs. MCI non significant; controls vs. AD p=0.003; MCI vs. AD p=0.045) . By contrast, MCI subjects performed similarly to AD patients but significantly worse than healthy controls in the condition assessing the binding of colors and shape in memory (STM binding accuracy: controls 0.83±0.15, MCI 0.65±0.16, AD 0.58±0.12. Post-hoc analysis: controls vs. MCI p=0,006; controls vs. AD p=0,011; MCI vs. AD non significant). MCI subjects performed better than AD patients (p<0.001) and worse than healthy controls p<0.001) performing each FCSRT subitem. Discussion and conclusions STM binding is impaired in MCI. MCI subjects could not remember the temporary binding of shapes and colours to the same extent as the controls. Indeed in the condition “Memory for shape-colour binding” they performed similarly to AD patients. This raise the question as to whether STM binding deficits, which have been demonstrated to be a preclinical marker of familial AD, might also be an early marker of sporadic AD. The novelty consists in assessing the best cognitive markers for early AD with a view of identifying which, if any, could complement the currently available biomarkers. Cognitive test batteries are time-consuming [Connor D.J, 2008]. Biomarkers are still experimental, invasive (CSF analysis) or expensive (amyloid-PET) and are not available in Primary Care. The ideal marker would be targeted at cognitive impairments in AD, could be used in Primary Care and in intervention trials with minimal training, would be non-invasive, quick to administer and inexpensive. Such a test would have economic and health benefits because fewer patients would be referred for inappropriate assessments and the efficacy of interventions would be more accurately assessed. Moreover, even if FCSRT is considered a sensitive marker of early AD and it works more effi ciently than other psychometric tests in identifying those MCI subjects who will progress to dementia [Ghiretti R, 2014], it does not give any information about how different groups of patients perform the STMBT. However, while we have the FCSRT normative data for Italian population, currently they are not available yet for STMBT and we need more studies to better detect the potentiality of the test. Bibliography Albert M.S. et al. Alzheimers Dement. 2011; 7,270-9. Americal Psychiatric Association, DSM, 4th edition ed. 1994, Washington: APA. Baddeley A. et al. Q J Exp Psychol A. 1986; 38:603-18. Baddeley A.D. et al., Brain 1991; 114:2521-42. Boustani M. et al. Ann Intern Med. 2003; 138:927-37. Brockmole, J.R., et al., Psychon Bull Rev, 2008. 15:543-7. Connor D.J. and M.N. Sabbagh, J Alzheimers Dis. 2008; 15:461-4. Cure F.. et al. J Alzheimers Dis. 2014;42(1):169-82. Della Sala S. et al. Arch Clin Neuropsychol. 2010; 25:410-9. Dubois B. et al. Lancet Neurol. 2007; 6:34-46. Dubois B et al., Lancet Neurol. 2014; 13(8):757. Foley J.A. et al. Arch Clin Neuropsychol. 2011; 26:340-8. Folstein M.F et al. Journal of Psychiatric Research 1975; 12:189-198 Frasson P et al. Neurol Sci. 2011; 32:1057-1062 Grober E. et al. Neurology 1988; 38:900-3. Ghiretti R. et al. J Alzheimers Dis. 2014; 41:S30. Jack C.R. Jr. et al. Alzheimers Dement. 2011; 7:257-62. Molinuevo J.L. et al Alzheimers Dement. 2014; Kaschel R. et al. J Neurol. 2009; 256:860-8. Parra M.A. et al. Brain 2009; 132:1057-66. Parra M.A. et al. J Neurol. 2010; 257:1160-9. Parra M.A. et al. Brain 2010; 133:2702-13. Sarazin M. et al. Neurology 2007; 69:1859-67. Winblad, B., et al., J Intern Med, 2004. 256,240-6.