The association between variation in the apolipoprotein E gene (APOE) and Alzheimer’s disease (AD) is well established . Of the three alleles (e2, e3, and e4), the APOE e4 allele is associated with increased risk of AD and a younger age at disease onset, whereas APOE e2 appears be protective. Given the phenotypic and pathological overlap between the two diseases, APOE has also been investigated as a candidate gene for susceptibility to PD-D; however, there has been a lack of consensus among studies to date.
Our own meta-analysis of 17 studies  suggested a modest association between increased APOE e4 frequency and PD-D (n = 501) compared with PD without dementia (n = 1145; odds ratio 1.74; 95% CI 1.4-2.2). However, the reliability of this association is questionable for a number of reasons, including small study sizes, substantial variation in the criteria used to define dementia, significant heterogeneity of odds ratios between studies, and evidence of publication bias. In a large cohort of 2412 PD patients, APOE e4 and APOE e2 carrier status was not associated with either PD risk or MMSE score measured at an average of 7 years from disease onset .
Longitudinal studies of APOE have likewise produced inconsistent results. In a cohort of 212 patients with established PD and an average disease duration of 7 years, APOE e4 carriers had a more rapid global cognitive decline than non-carriers as measured by their total score on the Mat- tis Dementia Rating Scale . However, the CamPaIGN study found no effect of APOE genotype on either the overall rate of cognitive decline or the risk of dementia at any time point [28, 57]. Likewise, no significant associations between APOE e4 and time to the onset of dementia were observed in a second study of 64 patients with prevalent PD followed longitudinally over a 12-year period . It is possible that these inconsistencies relate to variations in pathological specificity, with concurrent Alzheimer’s-type pathology accounting for some of the observed associations, particularly in older patients with longer disease durations. However, post-mortem studies have linked the APOE e4 allele to cortical Lewy body disease in the absence of significant Alzheimer’s type pathology, with a stronger association for DLB than PD-D , leading to the suggestion that in the presence of an a-synucleinopathy, APOE e4 might accelerate the onset of dementia such that it precedes parkinsonism, via a mechanism independent of amyloid processing . Hence exclusion of DLB cases may explain the lack of association between APOE e4 and Lewy body disease in some studies, particularly in incident PD cohorts.