between aging, health, cognition and lifestyle (Schaie, 2005). The sample included 111 participants, Mage 1⁄4 67 (age range 52 84), imaged on 3 occasions over 4 years. In our sample, there were 32 APOE e4 carriers and 79 APOE e4 noncarriers. Magnetization prepared rapid gradient echo (MPRAGE) imaging was performed on a Philips 3.0 TAchieva scanner. Cortical reconstruction and volumetric segmentation was performedwith the longitudinal pipeline of the FreeSurfer image analysis suite version 5.1.0 (http:// surfer.nmr.mgh.harvard.edu/). Cortical thickness values for each of the 68 parcels defined by the Desikan parcellation (Desikan et al., 2006) were extracted by subject and timepoint. We fit a linear mixed effects model for each parcel that included fixed effects of intercept, age, and APOE e4 carrier status, and the interaction of age andAPOE e4 carrier status to predict slope in cortical thickness of each parcel.Results: The mean estimated cortical thickness at age 60 (intercept) was thinner for APOE e4 carriers than APOE e4 non-carriers in the left inferior parietal parcel and left and right frontal pole. Slope differences (age X e4 carrier status interaction) were found in: left temporal pole and superior frontal regions, and right transverse temporal and caudal anterior cingulate (Figure 1). In all regions except the right transverse temporal, APOE e4 carriers had a steeper rate of decline than non-carriers.Conclusions: The APOE 4 allelemodulatesmean thickness and rates of change both in areas associated with normal aging and in areas associated with progression to AD.