nAChRs as measured by [ 18 F]F-A-85380 and PET after the 8-week galan- tamine treatment regime. P2-372 DECLINE IN COGNITIVE PERFORMANCE AND METABOLISM IN A MEDIAL BRAIN NETWORK DURING AGING IN HEALTHY CONTROLS Jose V. Pardo 1,2 , Joel T. Lee 1 , Sohail A. Sheikh 2 , Christa Surerus-Johnson 1 , Kristin R. Munch 2 , John V. Carlis 2 , Scott M. Lewis 1,2 , Michael A. Kuskowski 1,2 , Maurice W. Dysken 1,2 , 1 Minneapolis VAMC, Minneapolis, MN, USA; 2 University of Minnesota, Minneapolis, MN, USA. Contact e-mail: jvpardo@umn.edu Background: The brain metabolic and cognitive changes occurring in the healthy elderly require further study (e.g., A. Martin et al., 1991; J. Moeller et al., 1996; S. Schultz et al., 1999). Defining these alterations is necessary to distinguish healthy individuals from those with early AD/MCI. In particular, it is unknown whether any declines in cognition relate to specific foci of age-related decline in brain metabolism. Objective(s): To identify brain metabolic changes that correlate with aging and with measures of cognitive performance. Methods: A cohort of healthy subjects from young adulthood to the elderly were recruited for a normative database. The subjects had laboratory and physical examination; cognitive screening with the Minnesota Cognitive Acuity Screen (MNCAS; Knopman et al., 2000); clinical MRI; APOE genotyping; and FDG PET during the eyes closed, resting state. PET data were processed with 3D SSP (S. Minoshima et al., 1995) and in-house software. PET data were not otherwise corrected for structural atrophy so that observed changes may reflect a combination of decreased metabolism or structural atrophy. Results: Several regions of the brain showed significant negative correlations of metabolism with age. The major areas included doral medial prefrontal and anterior cingulate; sub- genual cingulate/basal forebrain; and dorsomedial thalamus (Fig. 1; color scale displays correlations between -1 and 0). There were also less exten- sive regions in various lateral hemisphere areas. Decline in cognitive measures correlated with foci in medial but not lateral regions. Conclu- sions: Medial brain regions known to be interconnected anatomically and to participate in executive function show age-related metabolic declines that correlate with decreases in several cognitive measures. These areas provide a background over which MCI/AD occur; they are typically re- moved by age regression in PET studies of MCI/AD. These findings bear upon the biology of normal aging and offer biological substrates to target age-related cognitive decline. P2-373 3D MAPPING OF GRAY MATTER ATROPHY IN SEMANTIC DEMENTIA AND FRONTAL VARIANT FRONTOTEMPORAL DEMENTIA Liana G. Apostolova 1 , David G. Clark 2 , Charleen Zoumalan 1 , Calen A. Steiner 1 , Aaron McMurtray 3 , Rebecca A. Dutton 1 , Kiralee M. Hayashi 1 , Arthur W. Toga 1 , Jeffrey L. Cummings 1 , Mario F. Mendez 3 , Paul M. Thompson 1 , 1 UCLA, Los Angeles, CA, USA; 2 UAB, Birmingham, AL, USA; 3 GLAVAMC, Los Angeles, CA, USA. Contact e-mail: lapostolova@mednet.ucla.edu Background: SD patients typically present with anomia, loss of semantic knowledge and impoverished circumlocutory speech. FTD patients typi- cally present with decline in social conduct, lack of empathy, disinhibition and poor judgment. Thus far imaging analyses comparing each of these disorders to normal controls have shown atrophy of the left temporal lobe especially the temporal pole and inferior temporal gyrus (ITG) in SD and frontal and temporal lobe atrophy with right-sided predilection in FTD. Objective: To compare the gray matter (GM) atrophy pattern of two frontotemporal lobar degeneration syndromes—frontotemporal dementia (FTD) and semantic dementia (SD). We tested the hypotheses that, relative to SD, FTD patients will have greater GM atrophy of the of the frontal association cortices, and, relative to FTD, SD patients will have greater GM atrophy of the left temporal association cortices. Methods: Five SD (mean age 65.6, SD 7.3 years; mean MMSE 20.8, SD 6.1) and 10 FTD (mean age 63.5, SD 10.7 years; mean MMSE 22.6, SD 5.2) subjects underwent magnetic resonance imaging (MRI): SPGR TR 28, TE 6, FOV 220 mm, 256x192, slice thickness 1.5 mm. Following 3D hemispheric reconstruction, 38 sulci per hemisphere were traced. The individual cortical surfaces were parametrized, flattened and warped so that data from corre- sponding gyri could be explicitly matched before averaging across sub- jects. Segmented GM was mapped onto the corresponding parametric hemispheric model in exact spatial correspondence. An average group 3D gray matter density map was created for each group. Statistical maps of the linkage between structural differences and clinical diagnosis were created. Results: SD diagnosis correlated with GM atrophy in the following re- gions: left posterior superior (r=0.3) and inferior (r=0.4) temporal gyri, left temporooccipital (r=0.4), left fusiform (r=0.4) and the precentral and postcentral gyri bilaterally (r=0.4). FTD diagnosis correlated with GM atrophy of bilateral medial and lateral frontal association (r=0.3-0.4) and the left orbitofrontal (r=0.4) cortices, as well as the right middle and inferior temporal gyri (r=0.3). Conclusions: Using innovative 3D cortical mapping techniques our preliminary work shows a predominantly left- sided and posterior GM atrophy pattern in SD and more right-sided and frontal GM atrophy pattern in FTD. P2-374 MR-GUIDED 3D PET MAPPING OF LONGITUDINAL CHANGES IN REGIONAL CEREBRAL METABOLISM OF NORMAL SUBJECTS Liana G. Apostolova, Ivo D. Dinov, Paul M. Thompson, Charleen Zoumalan, Calen A. Steiner, Erin Siu, Rebecca A. Dutton, Kiralee M. Hayashi, Gary W. Small, Jeffrey L. Cummings, Arthur W. Toga, Michael E. Phelps, Daniel HS Silverman, UCLA, Los Angeles, CA, USA. Contact e-mail: lapostolova@mednet.ucla.edu Background: Neurodegenerative pathology accumulates for years before the onset of clinical decline, and is associated with decreasing cortical glucose metabolism detectable with positron emission tomography (PET). Objective: To evaluate longitudinal PET intensity changes in normal elderly. Methods: Nineteen normal subjects (mean age 60.6 (SD=8.0) S354 Poster P2: Monday Posters