Washington University colleague Mark Mintun got the ball rolling on the project before handing the reins to first author Sheline as he transitioned into a new role at Avid Radiopharmaceuticals, Inc., Philadelphia, Pennsylvania. In a previous study, Sheline and Mintun had analyzed cognitively normal elderly with functional magnetic resonance imaging (fMRI), and found that in those with abnormally high brain amyloid, resting-state connectivity was disrupted in the same default-mode regions that fade in AD (Sheline et al., 2010). Others have published similar findings (e.g., see Hedden et al., 2009 and ARF HAI conference report). “That motivated me to ask whether you could see this even in people without brain amyloid,” Sheline told ARF.

A small fMRI study of non-demented seniors had found default-mode signal differences in ApoE4 carriers, compared with non-carriers (Fleisher et al., 2009), suggesting that ApoE4 contributes to AD-like brain abnormalities. However, because that study did not assess brain amyloid load, it was unclear whether the ApoE4 effects occurred before, along with, of after Aβ deposition. Another study strengthened the possibility that ApoE4 has amyloid-independent effects by uncovering distinct patterns of default-mode activity in ApoE4 carriers who were so young that they most likely had little to no amyloid in their brains (Filippini et al., 2009 and ARF related news story). Along similar lines, recent work shows that young ApoE4 carriers had mitochondrial abnormalities in the absence of detectable Aβ pathology, again suggesting that ApoE4 effects may precede Aβ changes (Valla et al., 2010). Cognitively normal E4 carriers also show brain glucose hypometabolism (see Reiman et al., 2005 and Langbaum et al., 2010).

In the present study, Sheline and colleagues analyzed 100 cognitively normal elderly enrolled in ongoing longitudinal studies of memory and aging at Washington University. All were amyloid-negative by Pittsburgh Compound B (PIB) binding. Compared with the ApoE4 non-carrier group, the 38 ApoE4 carriers had clear abnormalities in functional connectivity by resting-state fMRI. The changes appeared in nine predefined brain regions known to have abnormal connectivity to the precuneus in AD patients. Five of these brain areas—left hippocampus, left parahippocampus, dorsal anterior cingulate, dorsal occipital cortex, and middle temporal cortex—remained significant in post-hoc analysis on the 70 of 100 participants who, besides being PIB-negative, also had cerebrospinal fluid Aβ42 levels falling above the “abnormal” cutoff (see Morris et al., 2010). Beyond these pre-specified regions of interest, several additional areas showed fMRI alterations in unbiased, exploratory whole-brain analyses. The whole-brain findings suggest there might be “some ApoE4 signature for how the brain connects a little differently, independent of AD,” Sheline said, noting these data are preliminary and need confirmation in an independent sample.

“I think it’s a really interesting paper,” said Reisa Sperling of Brigham and Women’s Hospital in Boston. “It’s important to have a large enough sample of ApoE4 carriers who have relatively low levels of amyloid to ask these questions. I think this study adds to the evidence that there are ApoE4 effects on function that are somewhat independent of amyloid.” Sperling did mention one caveat—the possibility that ApoE4-positive participants had slightly more brain amyloid than did ApoE4 non-carriers. Though the absolute values for these measurements were small, the p value for comparing amyloid load between the two groups was 0.06, trending toward a significant difference. Instead of only analyzing ApoE4-positive and ApoE4-negative groups in a binary fashion, Sperling suggested, the data could have been solidified by also showing correlation between fMRI connectivity with amyloid burden, even among the ApoE4 non-carriers.

On the whole, though, the data seem to “indicate that reduced functional connectivity may actually be a precursor rather than a consequence of amyloid pathology in the brain,” commented Alexander Drzezga of the Technical University in Munich, Germany. It could also be that ApoE4 carriers have early amyloid pathology (e.g., soluble Aβ oligomers not detected by PIB imaging) that contributes to synaptic dysfunction and/or reduced connectivity, he noted in an e-mail to ARF (see full comment below). “This study has contributed some very important new insights and strongly encourages further work on the causal interaction between different pathologies involved in the disease, particularly in early or asymptomatic stages.” If further evidence like this comes to light, and not only in ApoE4 carriers, then the staggered curves in biomarker staging diagrams for preclinical AD so frequently cited at conferences (see ARF Webinar and Perrin et al., 2009) might have to shift to the right and accommodate a new connectivity curve as an even earlier marker on their left.—Esther Landhuis.

Reference:
Sheline YI, Morris JC, Snyder AZ, Price JL, Yan Z, D’Angelo G, Liu C, Dixit S, Benzinger T, Fagan A, Goate A, Mintun MA. APOE4 Allele Disrupts Resting State fMRI Connectivity in the Absence of Amyloid Plaques or Decreased CSF Abeta42. J. Neurosci. 15 December 2010;30(50):17035-17040. Abstract