Article Text

Apolipoprotein E: non-cognitive symptoms and cognitive decline in late onset Alzheimer’s disease
  1. Newcastle upon Tyne, UK
  1. Dr C Ballard, The Neurochemical Pathology Unit, Newcastle General Hospital, Westgate Road, Newcastle-upon-Tyne NE4 6BE, UK.

    Statistics from

    Holmes et al 1 presented interesting data suggesting that the presence of an apolipoprotein E (ApoE) ε4 allele did not effect the rate of progression of cognitive impairment. They also examined various non-cognitive symptoms, finding a positive association between depression and ApoE ε2 genotype.

    We followed up 51 patients with DSMIIIR dementia, representative of those with dementia in contact with psychiatric services in the United Kingdom for one year. Standardised clinical diagnosis were made according to the NINCDS ADRDA, DSMIIIR, Hachinski scale, and the criteria of McKeith et al for senile dementia of Lewy body type. Cognitive function was assessed at baseline and one year of follow up with the cognitive section of the Cambridge assessment for mental disorders in the elderly (CAMCOG). Psychiatric symptoms were also evaluated. Depression was examined using the Cornell depression scale and psychotic symptoms using the Burns symptom checklist. These were diagnosed according to RDC criteria and the criteria of Burnset al. Psychiatric symptoms were evaluated at monthly intervals for one year. A detailed description of the methodology is given elsewhere.2 ApoE genotyping was performed on genomic DNA extracted from whole blood samples according to established procedures.3

    The mean age of the patients was 79; 38 (75%) were women and they had mean CAMCOG scores of 48. Thirty six patients had possible or probable Alzheimer’s disease, 10 had a diagnosis of vascular dementia, and three received a diagnosis of senile dementia of Lewy body type. Two patients had DSMIIIR dementia which could not be further classified. Five patients were homozygous for ε4 and 23 were heterozygous.

    Patients with one or more ε4 alleles had a mean deterioration of 12.8 points on the CAMCOG score over one year compared with a mean of 11.4 in patients without any alleles, a non-significant difference (Mann-Whitney U test z=0.67, P=0.50), supporting the findings of Holmes et al. Patients with one or more ApoE ε4 alleles were, however, significantly less likely to develop RDC major depression (odds ratio 0.31, 95% confidence interval 0.09-0.99) and were significantly more likely to experience at least one psychotic symptom over the year of follow up (Fisher’s exact test P=0.006). The last association was more impressive as all 28 patients with one or more ε4 alleles experienced psychotic symptoms. Too few patients had the apolipoprotein ε2 genotype to permit meaningful statistical analysis.

    This is clearly just a small pilot study, but it does offer some provisional evidence that ε4 positive people may have a different non-cognitive phenotype. This is supported by other recent work,4 also reporting that patients with ε4 genotype were significantly more likely to experience psychosis. Although the multiple comparisons made in the study of Holmes et al 1 necessitate caution, it would have been valuable to report the actual P values, particularly as other work does suggest a possible association between ε4 genotype and psychotic symptoms.


    Holmes et al reply:

    In reply to the comments by Ballard et al we have extended our original study1-1 by ApoE genotyping an additional 68 Alzheimer’s disease cases from the Camberwell Dementia Case Register1-2 bringing the total to 232 cases fulfilling NINCDS-ADRDA criteria (154 probable and 78 possible, mean age 82.4 (SEM 0.4) years, 178 (77%) women). In all other respects the patient selection and assessment were identical to that previously reported and gave rise to frequencies of 0.03, 0.66, and 0.30 for the ApoE ε2, ε3, and ε4 alleles respectively. In this extended study in accord with our previous findings the only significant association was between the presence of the ApoE ε2 allele and the Cornell score (Mann-Whitney U test, P = 0.03) and with the presence of depressive symptomatology (Mann-Whitney U test, P = 0.01) after Bonferroni correction. However, we welcome the opportunity to present the full data set before Bonferroni correction (table) as it concurs with the findings of Ballard et al and other studies1-3 which suggest an association of the ApoE ε4 allele with psychotic phenomena as well as suggesting a possible association with wandering behaviours and eating disturbance. Unlike Ballard et al we did not, however, find a negative (or positive) association between depressive symptomatology and the presence of the ApoE ε4 allele suggesting that it is the presence of the ApoE ε2 allele rather than the absence of the ApoE ε4 allele that is associated with depressive symptomatology in Alzheimer’s disease.

    Association between the presence of non-cognitive symptoms and the presence of the ApoE ε2 and ε4 alleles by Mann Whitney U test (n=232)


    1. 1-1.
    2. 1-2.
    3. 1-3.
    View Abstract

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.