28 May 2010. Clinicians have long puzzled over the heterogeneity of Alzheimer disease symptoms—for example, some patients display the classic disorientation and forgetfulness, whereas others struggle more with planning and organization. A PNAS paper published online on May 17 suggests that ApoE genotype could, in part, explain such differences. Analyzing ApoE4 carriers and non-carriers from the (Alzheimer’s Disease Neuroimaging Initiative (ADNI)), cognitive neurologists David Wolk, University of Pennsylvania, Philadelphia, and Bradford Dickerson, Massachusetts General Hospital, Boston, report a strong link between performance in specific cognitive domains and neuroanatomical changes in the regions that support those functions. E4 carriers had poor memory retention with greater medial temporal lobe atrophy, whereas non-carriers did worse on working memory tasks and had increased frontoparietal atrophy. The findings suggest that ApoE not only influences AD risk, but may also affect how the disease manifests itself.
Prior studies have hinted that ApoE allelic variants could underlie phenotype differences in AD. ApoE4 carriers seem to have increased plaque deposition (Polvikoski et al., 1995) and intensified medial temporal lobe atrophy (e.g., Geroldi et al., 1999; Agosta et al., 2009 and ARF related news story) compared to E4 non-carriers. However, data on the latter have also been controversial, with some studies finding no ApoE-related brain volume differences (Jack et al., 1998; Drzezga et al., 2009). Similarly, research has at times suggested greater episodic memory impairment in E4 carriers (e.g., Marra et al., 2004; van der Vlies et al., 2007), but not always (Hashimoto et al., 2001; Pievani et al., 2009). It is conceivable that different misdiagnosis rates of E4 carriers and non-carriers could have skewed the results of past studies. Furthermore, most studies focused on either cognition or neuroanatomical changes in a single brain area—rarely both. “This study brought together a lot of those different elements, hopefully in a more rigorous and comprehensive way,” Wolk told ARF.
He and Dickerson analyzed 91 mild AD patients (67 carrying at least one ApoE4 allele and 24 non-carriers) from ADNI—a large, well-characterized population with rich neuroimaging and psychometric test data. Many in ADNI have had cerebrospinal fluid evaluations, too. The authors used the latter information to ease misdiagnosis concerns in the current study—by restricting participants to those with AD-typical tau and Aβ1-42 profiles.
ApoE4 carriers and non-carriers showed striking differences in immediate and delayed recall in the Auditory Verbal Learning Test, which involved memorizing a list of 15 words. During the first learning trial, in which participants heard the list and repeated as many words as they could recall, the non-carriers actually did worse. “This was more of a working memory trial, like remembering a phone number,” Wolk said. Over time, though, some of the learned information gets transferred to long-term memory, and by the fifth trial, the E4 carriers and non-carriers were doing about the same, suggesting comparable overall learning. “At that point, there’s some sort of combination in terms of how much working memory and long-term memory contribute to performance,” Wolk said. However, when asked to recall the words 30 minutes later, the E4 carriers did much worse than the non-carriers.
Patterns of brain atrophy in the two groups correlated well with their memory deficits. E4 carriers, who did comparatively worse on delayed recall, had smaller hippocampi and more cortical thinning in the medial temporal lobe, which is required for memory retention. Non-carriers, who had more trouble with immediate recall, had greater atrophy in parietal and frontal regions believed to mediate short-term memory storage. The cognitive deficits in each group correlated with atrophy in the brain regions that subserve those functions, Wolk said. “There was nice linkage between cognitive and neuroanatomical phenotypes.”
Furthermore, the study suggests E4’s effect on cognition may be independent of Aβ, said Yadong Huang, Gladstone Institute of Neurological Disease, San Francisco. Plaques accumulate more in the cortex than in the hippocampus, and E4 carriers have greater Aβ deposition than do non-carriers. This suggests that E4 carriers have more cortical plaques and thus should have greater cortical dysfunction. However, that was not the case in the current study. E4 carriers were more impaired in hippocampal cognitive domains, suggesting that ApoE4 primarily affected cognitive functions subserved by less plaque-laden brain areas.
The findings could also have implications for early diagnosis and monitoring of AD. “The disease might hit certain areas more prominently in carriers versus non-carriers,” Wolk said, noting that some cognitive and neuroanatomical measures could thus be more sensitive in one group or the other. Wolk and colleagues are extending their analysis to milder patients and have preliminary data suggesting that younger people may already have ApoE-linked cognitive deficiencies. Recent studies have found abnormal cerebral blood flow (Thambisetty et al., 2010 and ARF related news story) and default network activity (Filippini et al.., 2009 and ARF related news story) in young E4 carriers.
Huang said the current data fit with the notion that AD might more aptly be called “Alzheimer diseases.” He predicts the disease will be divided into subclasses. “One of them will be E4-related AD, the other E4-unrelated AD,” said Huang, citing the E4-non-carrier-selective benefits found in post-hoc analyses of GlaxoSmithKline’s rosiglitazone (see ARF conference story) and Elan’s passive immunotherapy (see ARF conference story) trial data.—Esther Landhuis.
Wolk DA, Dickerson BC, and the Alzheimer’s Disease Neuroimaging Initiative. Apolipoprotein E genotype has dissociable effects on memory and attentional-executive network function in Alzheimer’s disease. PNAS Early Edition. 17 May 2010. Abstract