Cognitive subtypes of probable Alzheimer's disease robustly identified in four cohorts.

INTRODUCTION: Patients with Alzheimer's disease (AD) show heterogeneity in profile of cognitive impairment. We aimed to identify cognitive subtypes in four large AD cohorts using a data-driven clustering approach. METHODS: We included probable AD dementia patients from the Amsterdam Dementia Cohort (n = 496), Alzheimer's Disease Neuroimaging Initiative (n = 376), German Dementia Competence Network (n = 521), and University of California, San Francisco (n = 589). Neuropsychological data were clustered using nonnegative matrix factorization. We explored clinical and neurobiological characteristics of identified clusters. RESULTS: In each cohort, a two-clusters solution best fitted the data (cophenetic correlation >0.9): one cluster was memory-impaired and the other relatively memory spared. Pooled analyses showed that the memory-spared clusters (29%-52% of patients) were younger, more often apolipoprotein E (APOE) ɛ4 negative, and had more severe posterior atrophy compared with the memory-impaired clusters (all P < .05). CONCLUSIONS: We could identify two robust cognitive clusters in four independent large cohorts with distinct clinical characteristics.

Alzheimer's disease (AD) is a heterogeneous disorder. We identified two cognitive AD subtypes in four cohorts with a data-driven approach. Nonamnestic AD is associated with distinct neurobiological characteristics.

Lower cerebral blood flow is associated with impairment in multiple cognitive domains in Alzheimer's disease.

INTRODUCTION: We examined the association between decreased cerebral blood flow (CBF) and cognitive impairment in Alzheimer's disease (AD), mild cognitive impairment (MCI), and subjective cognitive decline (SCD). METHODS: We included 161 AD, 95 MCI, and 143 SCD patients from the Amsterdam Dementia Cohort. We used 3-T pseudo-continuous arterial spin labeling to estimate whole-brain and regional partial volume-corrected CBF. Neuropsychological tests covered global cognition and five cognitive domains. Associations were investigated using linear regression analyses. RESULTS: In the whole sample, reduced overall and regional CBF was associated with impairment in all cognitive domains. We found significant interactions between diagnosis and CBF for language and between diagnosis and parietal CBF for global cognition and executive functioning. Stratification showed that decreased CBF was associated with worse performance in AD patients but not in MCI or SCD. DISCUSSION: Our results suggest that CBF may have potential as a functional marker of disease severity.

The identification of cognitive subtypes in Alzheimer's disease dementia using latent class analysis.

OBJECTIVE: Alzheimer's disease (AD) is a heterogeneous disorder with complex underlying neuropathology that is still not completely understood. For better understanding of this heterogeneity, we aimed to identify cognitive subtypes using latent class analysis (LCA) in a large sample of patients with AD dementia. In addition, we explored the relationship between the identified cognitive subtypes, and their demographical and neurobiological characteristics. METHODS: We performed LCA based on neuropsychological test results of 938 consecutive probable patients with AD dementia using Mini-Mental State Examination as the covariate. Subsequently, we performed multinomial logistic regression analysis with cluster membership as dependent variable and dichotomised demographics, APOE genotype, cerebrospinal fluid biomarkers and MRI characteristics as independent variables. RESULTS: LCA revealed eight clusters characterised by distinct cognitive profile and disease severity. Memory-impaired clusters-mild-memory (MILD-MEM) and moderate-memory (MOD-MEM)-included 43% of patients. Memory-spared clusters mild-visuospatial-language (MILD-VILA), mild-executive (MILD-EXE) and moderate-visuospatial (MOD-VISP) -included 29% of patients. Memory-indifferent clusters mild-diffuse (MILD-DIFF), moderate-language (MOD-LAN) and severe-diffuse (SEV-DIFF) -included 28% of patients. Cognitive clusters were associated with distinct demographical and neurobiological characteristics. In particular, the memory-spared MOD-VISP cluster was associated with younger age, APOE e4 negative genotype and prominent atrophy of the posterior cortex. CONCLUSIONS: Using LCA, we identified eight distinct cognitive subtypes in a large sample of patients with AD dementia. Cognitive clusters were associated with distinct demographical and neurobiological characteristics.

White Matter Hyperintensities Relate to Clinical Progression in Subjective Cognitive Decline.

BACKGROUND AND PURPOSE: In patients with subjective cognitive decline, we assessed whether small vessel disease was associated with clinical progression and cognitive decline. METHODS: We included 334 patients with subjective cognitive decline. Follow-up was 3±2 years. RESULTS: Fifty-three (16%) patients progressed clinically to mild cognitive impairment or dementia. White matter hyperintensities were associated with clinical progression and with annual decline on memory, attention, executive functioning, and global cognition. Microbleeds and lacunes were not associated with clinical progression or cognitive decline. CONCLUSIONS: In patients with subjective cognitive decline, patients with white matter hyperintensities are at increased risk of clinical progression and cognitive decline.

Regional atrophy is associated with impairment in distinct cognitive domains in Alzheimer's disease.

BACKGROUND: In Alzheimer's disease (AD), some patients present with cognitive impairment other than episodic memory disturbances. We evaluated whether occurrence of posterior atrophy (PA) and medial temporal lobe atrophy (MTA) could account for differences in cognitive domains affected. METHODS: In 329 patients with AD, we assessed five cognitive domains: memory, language, visuospatial functioning, executive functioning, and attention. Magnetic resonance imaging (MRI) was rated visually for the presence of MTA and PA. Two-way analyses of variance were performed with MTA and PA as independent variables, and cognitive domains as dependent variables. Gender, age, and education were covariates. As PA is often encountered in younger patients, analyses were repeated after stratification for age of onset (early onset, ≤65 years). RESULTS: The mean age of the participants was 67 years, 175 (53%) were female, and the mean Mini-Mental State Examination (score±standard deviation) was 20±5 points. Based on dichotomized magnetic resonance imaging ratings, 84 patients (26%) had MTA and PA, 98 (30%) had MTA, 57 (17%) had PA, and 90 (27%) had neither. MTA was associated with worse performance on memory, language, and attention (all, P<.05), and PA was associated with worse performance on visuospatial and executive functioning (both, P<.05). Stratification for age showed in patients with late-onset AD (n=173) associations between MTA and impairment on memory, language, visuospatial functioning, and attention (all, P<.05); in early-onset AD (n=156), patients with PA tended to perform worse on visuospatial functioning. CONCLUSIONS: Regional atrophy is related to impairment in specific cognitive domains in AD. The prevalence of PA in a large set of patients with AD and its association with cognitive functioning provides support for the usefulness of this visual rating scale in the diagnostic evaluation of AD.

Episodic memory and the medial temporal lobe: not all it seems. Evidence from the temporal variants of frontotemporal dementia.

BACKGROUND: Disproportionate medial temporal lobe atrophy (MTA) is an early finding in Alzheimer's disease (AD). Episodic memory impairment in AD is associated with the degree of MTA. Episodic memory impairment and MTA are also found in semantic dementia (SD) and in right temporal lobe atrophy (RTLA), the temporal variants of frontotemporal dementia, but their relationship is unclear. OBJECTIVE: To compare episodic memory impairment among patients with these temporal variants of frontotemporal dementia with that of patients with AD with the same degree of MTA. METHODS: Episodic memory was tested with the visual association test, and semantic memory (SM) with animal fluency and the visual association naming test. MTA was measured using a visual rating scale. Each patient with SD or RTLA was matched for MTA with two patients with AD. Comparisons of episodic memory and SM were made for patients with SD versus matched patients with AD; patients with RTLA versus matched patients with AD and for SD, RTLA and all patients with AD. RESULTS: 27 patients with SD and 11 with RTLA were matched with 54 and 22 patients with AD, respectively. Episodic memory was less impaired in patients with SD than in those with AD (8 versus 2; p<0.001) and in patients with RTLA than in those with AD (10 versus 4.5; p=0.009). Semantic memory was more affected in patients with SD than in those with AD, and the Mini Mental State Examination score was higher in patients with RTLA than in those with AD. Comparison of the three diagnostic groups showed that episodic memory was most impaired in AD, whereas SM was most impaired in SD. CONCLUSION: Since episodic memory impairment is more severe in AD than in SD and RTLA, despite a comparable degree of MTA, atrophy of the medial temporal lobe alone cannot account for episodic memory dysfunction.

EEG abnormalities are associated with different cognitive profiles in Alzheimer's disease.

OBJECTIVE: Our purpose was to investigate associations between different cognitive profiles and their underlying functional brain changes as measured by electroencephalogram (EEG) in Alzheimer's disease (AD). METHODS: EEG was obtained and neuropsychological performance assessed in 254 patients with AD. The EEGs were visually assessed for the presence of focal and/or diffuse abnormalities. Multivariate analysis of variance for repeated measures was performed with presence of focal and/or diffuse abnormalities as between-subjects factor and neuropsychological tests as within-subject factor. Age, sex and education were entered as covariates. RESULTS: Twenty-eight percent of the patients had a normal EEG, 32% had focal abnormalities, 14% diffuse abnormalities and 26% had both focal and diffuse abnormalities. Patients with a normal EEG presented with a cognitive profile in which memory was mostly impaired. Patients with focal and diffuse EEG abnormalities presented with a nonmemory profile. CONCLUSION: These results illustrate that specific types of EEG abnormalities are associated with different cognitive profiles in AD, providing biological support in terms of brain functioning for variability in cognitive impairment.

Differential effects of cognitive reserve and brain reserve on cognition in Alzheimer disease.

OBJECTIVE: To examine cross-sectional effects of cognitive reserve (CR) and brain reserve (BR) on cognition across the spectrum of Alzheimer disease (AD). METHODS: We included 663 AD biomarker-positive participants with dementia (probable AD, n = 462) or in the predementia stages (preclinical/prodromal AD, n = 201). Education was used as a proxy of CR and intracranial volume as a proxy of BR. Cognition was assessed across 5 domains (memory, attention, language, visuospatial, and executive functions). We performed multiple linear regression models to examine effects of CR and BR on cognitive domain Z scores, adjusted for cerebral atrophy. Furthermore, we assessed differences in effects according to disease stage and across degrees of total reserve using a 4-level variable (high CR/high BR, high CR/low BR, low CR/high BR, and low CR/low BR). RESULTS: We found positive, independent effects of both CR and BR across multiple cognitive domains. Stratification for disease stage showed that effects of CR on attention and executive functioning were greater in predementia than in dementia (ß = 0.39 vs ß = 0.21 [Welch t = 2.40, p < 0.01] and ß = 0.46 vs ß = 0.26 [t = 2.83, p < 0.01]). Furthermore, we found a linear trend for better cognitive performance in all domains in the high CR/high BR group, followed by high CR/low BR, low CR/high BR, and then low CR/low BR (p for trend <0.05). CONCLUSIONS: CR and BR both independently mitigate cognitive symptoms in AD. The positive effect of CR is most strongly expressed in the predementia stages and the additive effects of high CR and BR are most beneficial.

Thinner cortex in patients with subjective cognitive decline is associated with steeper decline of memory.

We aimed to investigate associations between regional cortical thickness and rate of decline over time in 4 cognitive domains in patients with subjective cognitive decline (SCD). We included 233 SCD patients with the total number of 654 neuropsychological assessments (median = 3, range = 2-8) and available baseline magnetic resonance imaging from the Amsterdam Dementia Cohort (125 males, age: 63 ± 9, Mini-Mental State Examination score: 28 ± 2). We assessed longitudinal cognitive functioning at baseline and follow-up in 4 cognitive domains (composite Z-scores): memory, attention, executive function, and language. Thickness (millimeter) was estimated using FreeSurfer for frontal, temporal, parietal, cingulate, and occipital cortices. We used linear mixed models to estimate effects of cortical thickness on cognitive performance (dependent variables). There were no associations between cortical thickness and baseline cognition, but a faster subsequent rate of memory loss was associated with thinner cortex of the frontal [ß (SE) = 0.20 (0.07)], temporal [ß (SE) = 0.18 (0.07)], and occipital [ß (SE) = 0.22 (0.09)] cortices (all p < 0.05FDR). These findings illustrate that early cortical changes, particularly in the temporal cortex, herald incipient cognitive decline related to neurodegenerative diseases, most prominently Alzheimer's disease.

Hypometabolism of the posterior cingulate cortex is not restricted to Alzheimer's disease.

When differential diagnosis of dementia includes both Alzheimer's disease (AD) and the behavioural variant of frontotemporal dementia (bvFTD), distribution of cerebral glucose metabolism as measured using [18F]­2­fluoro­2­deoxy­d­glucose positron emission tomography ([18F]FDG-PET) may be helpful. One important clue for differentiation is the presence of hypometabolism in the posterior cingulate cortex (PCC), usually associated with AD. PCC hypometabolism however, could also be present in bvFTD. Therefore, the specificity of PCC hypometabolism was examined. Based on visual reading PCC hypometabolism was present in 69-73/81 probable AD patients, in 10-16/33 probable bvFTD patients, and in 0-1/22 cognitive normal (CN) subjects. Findings were validated using a PCC to reference tissue [18F]FDG standard uptake value ratio (SUVr) cut-off, which was derived from the receiver operating characteristic (ROC) separating probable AD from CN, resulting in 9-14/33 bvFTD patients having PCC hypometabolism, depending on the reference tissue used. In conclusion, PCC hypometabolism is not restricted to AD.

Data-Driven Differential Diagnosis of Dementia Using Multiclass Disease State Index Classifier.

Clinical decision support systems (CDSSs) hold potential for the differential diagnosis of neurodegenerative diseases. We developed a novel CDSS, the PredictND tool, designed for differential diagnosis of different types of dementia. It combines information obtained from multiple diagnostic tests such as neuropsychological tests, MRI and cerebrospinal fluid samples. Here we evaluated how the classifier used in it performs in differentiating between controls with subjective cognitive decline, dementia due to Alzheimer's disease, vascular dementia, frontotemporal lobar degeneration and dementia with Lewy bodies. We used the multiclass Disease State Index classifier, which is the classifier used by the PredictND tool, to differentiate between controls and patients with the four different types of dementia. The multiclass Disease State Index classifier is an extension of a previously developed two-class Disease State Index classifier. As the two-class Disease State Index classifier, the multiclass Disease State Index classifier also offers a visualization of its decision making process, which makes it especially suitable for medical decision support where interpretability of the results is highly important. A subset of the Amsterdam Dementia cohort, consisting of 504 patients (age 65 ± 8 years, 44% females) with data from neuropsychological tests, cerebrospinal fluid samples and both automatic and visual MRI quantifications, was used for the evaluation. The Disease State Index classifier was highly accurate in separating the five classes from each other (balanced accuracy 82.3%). Accuracy was highest for vascular dementia and lowest for dementia with Lewy bodies. For the 50% of patients for which the classifier was most confident on the classification the balanced accuracy was 93.6%. Data-driven CDSSs can be of aid in differential diagnosis in clinical practice. The decision support system tested in this study was highly accurate in separating the different dementias and controls from each other. In addition to the predicted class, it also provides a confidence measure for the classification.

Screening for Mild Cognitive Impairment and Dementia with Automated, Anonymous Online and Telephone Cognitive Self-Tests.

BACKGROUND: Many older people worry about cognitive decline. Early cognitive screening in an anonymous and easily accessible manner may reassure older people who are unnecessarily worried about normal cognitive aging while it may also expedite help seeking in case of suspicious cognitive decline. OBJECTIVE: To develop and validate online and telephone-based automated self-tests of cognitive function. METHODS: We examined the feasibility and validity of the self-tests in a prospective study of 117 participants of whom 34 had subjective cognitive decline (SCD), 30 had mild cognitive impairment (MCI), and 53 had dementia. The ability of these self-tests to accurately distinguish MCI and dementia from SCD was examined with ROC curves. Convergent validity was examined by calculating rank correlations between the self-tests and neuropsychological tests. RESULTS: Both the online and telephone cognitive self-tests were feasible, because the majority of participants (86% and 80%, respectively) were able to complete them. The online self-test had adequate diagnostic accuracy in the screening for MCI and dementia versus SCD with an Area under the Curve (AUC) of 0.86 (95% CI: 0.78-0.93). The AUC of the MMSE was 0.82 (95% CI: 0.74-0.89). By contrast, the telephone self-test had lower diagnostic accuracy (AUC = 0.75, 95% CI: 0.64-0.86). Both self-tests had good convergent validity as demonstrated by moderate to strong rank correlations with neuropsychological tests. CONCLUSION: We demonstrated good diagnostic accuracy and convergent validity for the online self-test of cognitive function. It is therefore a promising tool in the screening for MCI and dementia.

Rare Genetic Variant in SORL1 May Increase Penetrance of Alzheimer's Disease in a Family with Several Generations of APOE-ɛ4 Homozygosity.

BACKGROUND: The major genetic risk factor for late onset Alzheimer's disease (AD) is the APOE-ɛ4 allele. However, APOE-ɛ4 homozygosity is not fully penetrant, suggesting co-occurrence of additional genetic variants. OBJECTIVE: To identify genetic factors that, next to APOE-ɛ4 homozygosity, contribute to the development of AD. METHODS: We identified a family with nine AD patients spanning four generations, with an inheritance pattern suggestive of autosomal dominant AD, with no variants in PSEN1, PSEN2, or APP. We collected DNA from four affected and seven unaffected family members and performed exome sequencing on DNA from three affected and one unaffected family members. RESULTS: All affected family members were homozygous for the APOE-ɛ4 allele. Statistical analysis revealed that AD onset in this family was significantly earlier than could be expected based on APOE genotype and gender. Next to APOE-ɛ4 homozygosity, we found that all four affected family members carried a rare variant in the VPS10 domain of the SORL1 gene, associated with AßPP processing and AD risk. Furthermore, three of four affected family members carried a rare variant in the TSHZ3 gene, also associated with AßPP processing. Affected family members presented between 61 and 74 years, with variable presence of microbleeds/cerebral amyloid angiopathy and electroencephalographic abnormalities. CONCLUSION: We hypothesize that next to APOE-ɛ4 homozygosity, impaired SORL1 protein function, and possibly impaired TSHZ3 function, further disturbed Aß processing. The convergence of these genetic factors over several generations might clarify the increased AD penetrance and the autosomal dominant-like inheritance pattern of AD as observed in this family.

Widespread disruption of functional brain organization in early-onset Alzheimer's disease.

Early-onset Alzheimer's disease (AD) patients present a different clinical profile than late-onset AD patients. This can be partially explained by cortical atrophy, although brain organization might provide more insight. The aim of this study was to examine functional connectivity in early-onset and late-onset AD patients. Resting-state fMRI scans of 20 early-onset (<65 years old), 28 late-onset (≥65 years old) AD patients and 15 "young" (<65 years old) and 31 "old" (≥65 years old) age-matched controls were available. Resting-state network-masks were used to create subject-specific maps. Group differences were examined using a non-parametric permutation test, accounting for gray-matter. Performance on five cognitive domains were used in a correlation analysis with functional connectivity in AD patients. Functional connectivity was not different in any of the RSNs when comparing the two control groups (young vs. old controls), which implies that there is no general effect of aging on functional connectivity. Functional connectivity in early-onset AD was lower in all networks compared to age-matched controls, where late-onset AD showed lower functional connectivity in the default-mode network. Functional connectivity was lower in early-onset compared to late-onset AD in auditory-, sensory-motor, dorsal-visual systems and the default mode network. Across patients, an association of functional connectivity of the default mode network was found with visuoconstruction. Functional connectivity of the right dorsal visual system was associated with attention across patients. In late-onset AD patients alone, higher functional connectivity of the sensory-motor system was associated with poorer memory performance. Functional brain organization was more widely disrupted in early-onset AD when compared to late-onset AD. This could possibly explain different clinical profiles, although more research into the relationship of functional connectivity and cognitive performance is needed.

Preclinical AD predicts decline in memory and executive functions in subjective complaints.

OBJECTIVE: We assessed whether preclinical Alzheimer disease (AD) based on CSF biomarkers at baseline predicts decline in cognitive functioning as measured by repeated neuropsychological tests for 4 cognitive domains in patients with subjective complaints. METHODS: We included 132 patients with subjective complaints from our memory clinic-based Amsterdam Dementia Cohort, who underwent lumbar puncture and had repeated (range 2-7) neuropsychological evaluations. Follow-up was 2 ± 1 years. CSF biomarkers amyloid-ß (Aß42), total tau (Tau), and hyperphosphorylated tau-181 were used to define National Institute on Aging-Alzheimer's Association (NIA-AA) preclinical AD stages. Predictive value of preclinical AD stages as defined by CSF biomarkers, individual biomarkers, and Aß42/tau ratio was assessed using linear mixed models. Outcome measures were compound z scores for memory, attention, executive functioning, language, and global cognition. Analyses were adjusted for age, sex, and education. RESULTS: Patients were 61 ± 8 years old; 56 (42%) were women. Average baseline Mini-Mental State Examination score was 28.3 ± 1.5. Patients who fulfilled criteria for preclinical AD (stage 1: n = 11 + stage 2: n = 10) showed decline over time in memory (ß ± SE -0.41 ± 0.14, p < 0.01), executive functions (-0.21 ± 0.08, p < 0.01), and global cognition (-0.29 ± 0.10, p < 0.01). There were no differences in cognitive decline between NIA-AA preclinical AD stages 1 and 2. In patients with normal CSF biomarkers, we observed memory improvement (0.19 ± 0.07, p < 0.01) and stable performance in all other domains. CONCLUSIONS: CSF evidence of preclinical AD in patients with subjective complaints predicted cognitive decline over time, encompassing more than memory alone. Executive functioning and global cognitive functioning also deteriorated. On the other hand, 2-year prognosis for patients without evidence of AD pathophysiology was good.

Early onset Alzheimer's disease is associated with a distinct neuropsychological profile.

Alzheimer's disease (AD) in younger patients is associated with a higher prevalence of atypical symptoms. We examined neuropsychological performance according to age-at-onset. We assessed cognition in 172 patients with AD (81 early and 91 late onset) in five cognitive domains (memory, language, visuo-spatial functioning, executive functioning, attention). Dementia severity was assessed using the Mini-Mental State Examination (MMSE) and global cognitive decline using Cambridge Cognitive Examination (CAMCOG). Analyses of variance were performed with age-at-onset as between-subjects factor, and gender and education as covariates. Analysis was repeated after stratification for dementia severity (based on median MMSE). In early onset AD, age (mean ± SD) was 60 ± 4 years; 44 (54%) were female. In late onset AD, age was 72 ± 5 years; 47 (52%) were female. Dementia severity and global cognitive decline did not differ between groups (early onset: MMSE: 20 ± 5, CAMCOG: 69 ± 15, late onset: MMSE: 21 ± 5, CAMCOG: 70 ± 15; p > 0.05). Early onset patients performed worse than late onset patients on visuo-spatial functioning (p < 0.01), executive functioning (p < 0.001), and attention (p < 0.01). Late onset patients performed worse on memory, although not significantly (p = 0.11). Stratification for dementia severity showed that in mildly demented early onset patients, memory function was remarkably preserved compared to late onset patients (p < 0.01). In moderate AD, differences in memory function disappeared, but early onset patients performed worse on visuo-spatial functioning (p < 0.01), executive functioning (p < 0.001), and attention (p < 0.01) than late onset patients. Adjustment for APOE left results unchanged. In conclusion, early onset AD presents with a different cognitive profile and the disease course seems different. Relative sparing of memory function in early stages stresses the need to adequately test other cognitive domains.

Reduction of nocturnal slow-wave activity affects daytime vigilance lapses and memory encoding but not reaction time or implicit learning.

Total sleep deprivation in healthy subjects has a profound effect on the performance on tasks measuring sustained attention or vigilance. We here report how a selective disruption of deep sleep only, that is, selective slow-wave activity (SWA) reduction, affects the performance of healthy well-sleeping subjects on several tasks: a "simple" and a "complex" vigilance task, a declarative learning task, and an implicit learning task despite unchanged duration of sleep. We used automated electroencephalogram (EEG) dependent acoustic feedback aimed at selective interference with-and reduction of-SWA. In a within-subject repeated measures crossover design, performance on the tasks was assessed in 13 elderly adults without sleep complaints after either SWA-reduction or after normal sleep. The number of vigilance lapses increased as a result of SWA reduction, irrespective of the type of vigilance task. Recognition on the declarative memory task was also affected by SWA reduction, associated with a decreased activation of the right hippocampus on encoding (measured with fMRI) suggesting a weaker memory trace. SWA reduction, however, did not affect reaction time on either of the vigilance tasks or implicit memory task performance. These findings suggest a specific role of slow oscillations in the subsequent daytime ability to maintain sustained attention and to encode novel declarative information but not to maintain response speed or to build implicit memories. Of particular interest is that selective SWA reduction can mimic some of the effects of total sleep deprivation, while not affecting sleep duration.

Cognitive impairment in Alzheimer's disease is modified by APOE genotype.

AIM: We examined whether impairment in specific cognitive domains in Alzheimer's disease (AD) differed according to APOE genotype and age at onset. METHODS: Cognitive functions of 229 consecutive AD patients were assessed using Visual Association Test (VAT), Memory Impairment Screen+ (MIS+), VAT object naming, fluency test and Trail Making Test (TMT). Dementia severity was assessed using MMSE. ANOVAs were performed with APOE genotype and age at onset as independent variables and sex, education and MMSE as covariates. RESULTS: 28% of patients were APOE epsilon4-negative, 58% heterozygous and 14% homozygous. A significant association between APOE genotype and VAT and MIS+ was found when correcting for sex and education. An interaction effect between APOE genotype and age at onset on VAT and VAT object naming was found, with young carriers performing worse than young noncarriers. By contrast, when additionally correcting for MMSE, a significant association between APOE genotype and VAT object naming, TMT-A and TMT-B was found, with noncarriers performing worse than carriers. CONCLUSION: Memory was more impaired among APOE epsilon4 carriers than among noncarriers. By contrast, naming, executive functions and mental speed were more impaired among APOE epsilon4 noncarriers. This suggests that the APOE genotype modifies the clinical phenotype in terms of cognitive impairment in AD.