Neuropsychological and behavioural correlates of CSF biomarkers in dementia

https://doi.org/10.1016/j.neuint.2005.11.002Get rights and content

Abstract

To improve clinical, neuropsychological and behavioural characterisation of the cerebrospinal fluid (CSF) biomarkers β-amyloid(1-42) protein (Aβ42), protein tau (tau) and tau phosphorylated at threonine 181 (P-tau181) across diagnostic dementia categories, a prospective study was set up. Patients with probable Alzheimer's disease (AD) (n = 201), AD with cerebrovascular disease (CVD) (AD + CVD) (n = 33), frontotemporal dementia (FTD) (n = 27), dementia with Lewy bodies (DLB) (n = 22) and healthy controls (n = 148) were included. All patients underwent neuropsychological examination and behavioural assessment by means of a battery of behavioural assessment scales. CSF was obtained by lumbar puncture and levels of Aβ42, tau and P-tau181 were determined with commercially available ELISA kits.

Negative correlations between CSF Aβ42 levels and aggressiveness (Spearman: r = −0.223; p = 0.002) and positive correlations with age at inclusion (r = 0.195; p = 0.006), age at onset (r = 0.205; p = 0.003) and MMSE scores (r = 0.198; p = 0.005) were found in AD. In AD + CVD, CSF Aβ42 levels were correlated with MMSE (r = 0.482; p = 0.006), Hierarchic Dementia Scale (r = 0.503; p = 0.017) and Boston Naming Test (r = 0.516; p = 0.012) scores. In controls, age was positively correlated with CSF tau (r = 0.465; p < 0.001) and P-tau181 levels (r = 0.312; p < 0.001). CSF tau and P-tau181 levels correlated significantly in all groups, whereas CSF Aβ42 correlated with tau and P-tau181 levels in healthy controls only.

Negative correlations between CSF Aβ42 levels and aggressiveness were found in AD patients. CSF Aβ42 seems to be a stage marker for AD (+/−CVD) given the positive correlations with neuropsychological test results suggesting that CSF Aβ42 might be of help for monitoring disease progression. Different correlations between age and CSF biomarker levels were obtained in healthy controls compared to AD patients, indicating that AD-induced pathophysiological processes change age-dependent regulation of CSF biomarker levels.

Introduction

Following the identification of protein tau (tau) in cerebrospinal fluid (CSF) as a marker for Alzheimer's disease (AD) (Vandermeeren et al., 1993), several biomarkers for different neurodegenerative disorders resulting in dementia have been investigated. The combined assessment of cerebrospinal fluid levels of β-amyloid(1-42) protein (Aβ42) and protein tau (tau) helps discriminating Alzheimer's disease (AD) from age-associated memory impairment, depression and Parkinson's disease in routine clinical practice (Hulstaert et al., 1999, Andreasen et al., 2001, Hampel et al., 2004b). The combined assessment of CSF Aβ42, tau and phosphorylated tau (P-tau) levels could increase specificity for discriminating AD from other (degenerative) dementias (Blennow and Vanmechelen, 2003, Schoonenboom et al., 2004, Andreasen and Blennow, 2005) although its added diagnostic value remains to be established in prospective studies including patients with neuropathologically confirmed clinical diagnoses.

Recent studies suggested new potential applications for Aβ42, tau and P-tau such as predicting the development of AD in mild cognitive impairment (MCI) patients, monitoring of AD disease progression and monitoring of the efficacy of disease-modifying drugs for AD (Tapiola et al., 2000, Riemenschneider et al., 2002a, Andreasen et al., 2003, Stefanova et al., 2003, Wahlund and Blennow, 2003, Zetterberg et al., 2003, Andreasen and Blennow, 2005). These potential new applications urge the need for a better clinical characterisation of the above-mentioned biomarkers.

Although behavioural correlates of CSF biomarkers have not been studied, some publications dealing with disease severity, disease progression and neuropsychological characteristics contributed to the clinical characterisation of CSF biomarkers. In a population of 28 MCI patients, low CSF Aβ42 levels predicted fast progression to dementia, whereas CSF tau levels correlated with memory performance (Ivanoiu and Sindic, 2005). In a longitudinal study including MCI and AD patients, CSF Aβ42 levels correlated positively with baseline brain volumes and negatively with baseline ventricular volumes, suggesting that CSF Aβ42 levels reflect disease stage (Wahlund and Blennow, 2003). In accordance with these results, Okamura et al. (1999) showed that CSF Aβ42 levels correlated positively with global and, in particular, temporal lobe glucose metabolism by means of positron emission tomography in both AD and a small group of non-AD dementias. However, conflicting data have been published with regard to possible associations between dementia severity and CSF Aβ42 levels in AD as some authors found inverse correlations (Nitsch et al., 1995, Hock et al., 1998, Jensen et al., 1999, Samuels et al., 1999, Andreasen et al., 2001, Csernansky et al., 2002, Riemenschneider et al., 2002b, Ganzer et al., 2003), whereas others did not (Andreasen et al., 1998, Andreasen et al., 1999, Mehta et al., 2001, Verbeek et al., 2003, Ivanoiu and Sindic, 2005). Hulstaert et al. (1999) reported a significant correlation between CSF Aβ42 levels and MMSE scores in men but not in female subjects with AD.

Conflicting data have been published concerning possible associations between CSF tau levels and dementia severity as some authors found positive correlations in AD (Parnetti et al., 2001, Csernansky et al., 2002, Schönknecht et al., 2003), whereas others did not (Verbeek et al., 2003, Lewczuk et al., 2004, Grossman et al., 2005, Ivanoiu and Sindic, 2005). One study that as well included FTD patients, did not find any significant associations between CSF tau levels and dementia severity either (Grossman et al., 2005). CSF P-tau231 levels were correlated with cognitive decline in AD (Hampel et al., 2001), whereas conflicting reports were published with regard to correlations between dementia severity and CSF P-tau181 levels as Schönknecht et al. (2003) found a positive association while the majority of studies did not find such correlations (Itoh et al., 2001, Maddalena et al., 2003, Hampel et al., 2004a, Lewczuk et al., 2004). Annual change in ventricular volume (as a measure of progression of brain atrophy or ventricular widening) was found to correlate with CSF levels of tau and tau phosphorylated at threonine 181 (P-tau181) in AD and MCI patients, suggesting that these markers reflect the intensity of the disease process with higher levels of these CSF biomarkers in case of a more aggressive disease course (Wahlund and Blennow, 2003).

Given the many conflicting data, we set up a prospective study including clinically, neuropsychologically and behaviourally well-characterised patients with various degenerative dementias (with or without cerebrovascular disease) aiming to improve the neuropsychological and behavioural characterisation of the CSF biomarkers Aβ42, tau and P-tau181 across diagnostic categories. Given the nature of the study, a correlation study, primary outcome measures were defined as significant correlations between CSF biomarker levels (analysed using commercially available sandwich enzyme-linked immunosorbent assay (ELISA) kits) and neuropsychological and behavioural data (measured by means of a standard battery of neuropsychological tests and behavioural assessment scales). Secondary outcome measures were defined as significant correlations between CSF biomarkers levels and age, disease duration and estimated rate of cognitive decline. Demographic, neuropsychological and behavioural data of each disease group are presented and compared to the AD group. To test whether eventual associations with age were disease-specific, CSF biomarker levels were correlated with age within a healthy control group too. CSF levels of Aβ42, tau and P-tau181 of each disease group are compared to the AD and control groups.

Section snippets

Study population

The study population consisted of 283 patients who were recruited from our Memory Clinic and who (1) gave informed consent for participation to the study and (2) had a diagnosis of degenerative dementia (with or without CVD) following diagnostic work-up that consisted of a general physical and neurological examination, blood screening, structural neuroimaging consisting of brain computerised tomography scan and/or magnetic resonance imaging, functional brain imaging (single photon emission

Demographic, clinical and neuropsychological data

Distribution of gender ratios was significantly different between DLB and AD patients (Table 1). Compared to AD patients, FTD patients were significantly younger at inclusion, whereas FTD and DLB patients were significantly younger at disease onset.

The control group consisted of 148 subjects (male/female: 99/49) resulting in a significant different distribution of gender ratios compared to the AD group. With a mean age of 54.6 ± 18.2 years (range: 17–88 years), the control group was significantly

Discussion

We presented time-linked behavioural and neuropsychological data from a prospective Belgian study of patients with various degenerative forms of dementia, allowing calculation of correlations with CSF biomarker levels. To the best of our knowledge, this is the first study correlating behavioural data with CSF biomarker levels. Moreover, the large number of clinically, neuropsychologically and behaviourally well-characterised patients that were diagnosed according to strict diagnostic criteria

Conclusions

The main findings derived from this large prospective study on behavioural and neuropsychological correlates of frequently used CSF biomarkers across diagnostic dementia categories can be summarised as follows:

  • (1)

    CSF Aβ42 seems to be a stage marker in dementia disorders with AD pathology given the positive correlations with neuropsychological test results in AD (with or without CVD) resulting in lower CSF Aβ42 levels in case of more pronounced neuropsychological deficits. These findings suggest

Acknowledgements

This research was supported by the International Alzheimer's Research Foundation, the Special Research Fund of the University of Antwerp (BOF-UA), the Institute Born-Bunge, the agreement between the Institute Born-Bunge and the University of Antwerp, Medical Research Foundation Antwerp, Neurosearch Antwerp, the Research Foundation—Flanders (FWO-F) and the Institute for Promotion of Innovation through Science and Technology in Flanders (IWT-Vlaanderen). S.E. is a postdoctoral fellow of the

References (65)

  • T. Sunderland et al.

    Cerebrospinal fluid β-amyloid1-42 in control subjects at risk for Alzheimer's disease: the effect of APOE ɛ4 allele

    Biol. Psychiatry

    (2004)
  • T. Tapiola et al.

    Three-year follow-up of cerebrospinal fluid tau, β-amyloid 42 and 40 concentrations in Alzheimer's disease

    Neurosci. Lett.

    (2000)
  • E. Vanmechelen et al.

    Cerebrospinal fluid τ and β-amyloid(1-42) in dementia disorders

    Mech. Ageing Dev.

    (2001)
  • L.-O. Wahlund et al.

    Cerebrospinal fluid biomarkers for disease stage and intensity in cognitively impaired patients

    Neurosci. Lett.

    (2003)
  • H. Zetterberg et al.

    Cerebrospinal fluid markers for prediction of Alzheimer's disease

    Neurosci. Lett.

    (2003)
  • American Psychiatric Association

    DSM-IV: Diagnostic and Statistical Manual of Mental Disorders

    (1994)
  • N. Andreasen et al.

    Cerebrospinal fluid tau protein as a biochemical marker for Alzheimer's disease: a community based follow up study

    J. Neurol. Neurosurg. Psychiatry

    (1998)
  • N. Andreasen et al.

    Cerebrospinal fluid beta-amyloid(1-42) in Alzheimer disease: differences between early- and late-onset Alzheimer disease and stability during the course of the disease

    Arch. Neurol.

    (1999)
  • N. Andreasen et al.

    Evaluation of CSF-tau and CSF-Aβ42 as diagnostic markers for Alzheimer's disease in clinical practice

    Arch. Neurol.

    (2001)
  • N. Andreasen et al.

    Cerebrospinal fluid levels of total-tau, phospho-tau and Aβ42 predicts development of Alzheimer's disease in patients with mild cognitive impairment

    Acta Neurol. Scand.

    (2003)
  • F. Assal et al.

    Neuropsychiatric symptoms in the dementias

    Curr. Opin. Neurol.

    (2002)
  • J. Cohen-Mansfield et al.

    A description of agitation in a nursing home

    J. Gerontol.

    (1989)
  • J.G. Csernansky et al.

    Relationships among cerebrospinal fluid biomarkers in dementia of the Alzheimer type

    Alzheimer Dis. Assoc. Disord.

    (2002)
  • J.F. Demonet et al.

    Standardization of a modular and hierarchic cognitive evaluation scale applicable to dementia. A French version of the Hierarchic Dementia Scale

    Rev. Neurol.

    (1990)
  • P.P. De Deyn et al.

    The concept memory clinic: historical perspective and objectives

  • P.P. De Deyn et al.

    The Middelheim Frontality Score: a behavioural assessment scale that discriminates frontotemporal dementia from Alzheimer's disease

    Int. J. Geriatr. Psychiatry

    (2005)
  • K. De Vreese et al.

    Discrimination of dementia with Lewy bodies from Alzheimer's disease via determination of the phospho-tau concentration in cerebrospinal fluid

  • S. Engelborghs et al.

    Evidence for an association between the CSF HVA:5-HIAA ratio and aggressiveness in patients with frontotemporal dementia but not in Alzheimer's disease

    J. Neurol. Neurosurg. Psychiatry

    (2004)
  • S. Engelborghs et al.

    Neuropsychiatric symptoms of dementia. Cross-sectional analysis from a prospective, longitudinal Belgian study

    Int. J. Geriatr. Psychiatry

    (2005)
  • M. Folstein et al.

    Mini-Mental State. A practical method for grading the cognitive state of patients for the clinician

    J. Psychiatr. Res.

    (1975)
  • S. Ganzer et al.

    CSF-tau, CSF-Abeta1-42, ApoE-genotype and clinical parameters in the diagnosis of Alzheimer's disease: combination of CSF-tau and MMSE yields highest sensitivity and specificity

    J. Neural. Trans.

    (2003)
  • M. Grossman et al.

    Cerebrospinal fluid profile in frontotemporal dementia and Alzheimer's disease

    Ann. Neurol.

    (2005)
  • Cited by (0)

    1

    These authors contributed equally to this work.

    View full text