prevalence and annualized incidence of LMB according to PiB findings in cohort ranging from cognitively-normal elderly (NC) to Alzheimer’s Dis- ease dementia (AD), over 3 years. Methods: 174 participants were studied at entry from the Australian Imaging, Biomarkers and Lifestyle Study of Ageing (AIBL) with 3T SWI MRI and 11C-PiB PET. 120 (79 NC, 23 with Mild Cognitive Impairment (MCI) and 18 AD) returned for imaging at 18 months or 3 years. Images were inspected for microbleeds and infarc- tion, blind to clinical and PiB findings. White matter hyperintensity (WMH) volume was quantified from FLAIR images using a semi-automated algo- rithm. PiB binding was treated as both a continuous variable (SUVR) and dichotomized as previously published (PiB+/-). All regression analyses were adjusted for age and gender. Results: LMB were present at baseline in 28% of NC, 34% of MCI and 44% of AD patients. The prevalence was 53% in PiB+ NC and 16% in PiB- NC. The mean incidence rate of LMB was significantly different between NC (0.18 LMB/year), MCI (0.28) and AD (1.04 LMB/year). In NC, it was 0.38 for PiB+ and 0.09 for PiB- (P<0.01). Baseline LMB were associated with age (OR 1.1, 95% C.I.1.0- 1.2) and SUVR (OR 4.8, 95% C.I.1.5-14.8) in NC participants, and with SUVR (OR 3.6, 95% C.I.1.2-10.3), male gender (OR 7.9, 95% C.I.2.3- 28.1) and WMH volume (OR 1.1, 95%C.I. 1.0-1.1) in MCI and AD. For NC, LMB incidence rate was associated with SUVR and the presence of baseline lacunes, while for AD and MCI patients, LMB incidence was asso- ciated with WMH volume and of baseline lacunes. Conclusions: Incidence of lobar microbleeds is associated with Ab-burden and markers of small ves- sel disease. O4-02-04 ALZHEIMER’S DISEASE PATIENTS WITH MICROBLEEDS AND ADDITIONAL WHITE MATTER HYPERINTENSITIES: DOES IT MATTER? Marije Benedictus, Jeroen Goos, Maja Binnewijzend, Majon Muller, Frederik Barkhof, Philip Scheltens, Niels Prins, Wiesje Van der Flier, VUMC, Amsterdam, Netherlands. Background: Microbleeds (MBs) are generally considered an expres- sion of cerebral small vessel disease (CSVD), and associations with white matter hyperintensities (WMH) have repeatedly been shown. MBs may also be observed in the absence of WMH, however. We stud- ied how patients with Alzheimer’s disease (AD) with both MBs and WMH differ from patients with exclusively MBs and hypothesized that MBs in isolation might indicate underlying cerebral amyloid angi- opathy (CAA), while MBs in the company of WMH might reflect hy- pertensive vasculopathy. Methods: A total of 371 AD patients with available T2 and FLAIR MRI scans were included. MBs were counted and WMH were rated with the Fazekas scale. Patients were categorized depending on MB presence (MB-:0; MB+:>1) and WMH presence (WMH-:Fazekas 0-1; WMH+:Fazekas 2-3), into four groups: MB- WMH-; MB-WMH+; MB+WMH- and MB+WMH+. One-way analysis of variance with post-hoc LSD and Chi-squared tests were performed to compare groups. Results: Of all AD patients (age 69+9, 55% female), 158 (43%) presented with MBs and/or WMH. Within this group, 47 (30%) patients presented with MB+WMH+, 51 (32%) patients had MB+WMH- and 60 (38%) patients had MB-WMH+. ANOVA’s showed group differences for age, sex, systolic blood pressure, hypertension, MTA and lacunes (Table 1). APOE-ε4 status did not differ significantly. Cerebrospinal fluid (CSF) levels of Aß42 differed only for MB-WMH- versus MB+WMH+ (488+169 vs. 420+103, P<0.05). When directly com- paring MB patients with and without WMH, patients with MB+WMH+ were older (74+10 vs. 68+9, P<0.05), presented more often with lacunes (24% vs. 5%, P<0.05) and showed a trend towards presenting more fre- quently with multiple MBs (62% vs. 43%, P ¼ 0.07). Conclusions: In this fairly young sample, a substantial number of AD patients presented with MBs and/or WMH. Furthermore, we found that MBs regularly occur in the absence of WMH. We were not able to show a relationship with characteristics indicating a specific underlying substrate. The group with isolated MBs was younger, presented less often with lacunes and tended towards having fewer MBs than the group with both MBs and WMH, sug- gesting a less advanced disease stage. We are currently rating MB location to further elucidate this matter. Table 1 AD patients categorized according to MB and/or WMH presence MB-WMH- (n¼213) MB-WMH+ (n¼60) MB+WMH- (n¼51) MB+WMH+ (n¼47) Patient characteristics Age 67 6 9 75 6 8 a 68 6 9 b 74 6 10 a, b, c Sex (female) 123 (58%) 40 (67%) 22 (43%) b 19 (40%) a, b MMSE 20 6 5 21 6 5 19 6 5 20 6 5 Camcog 67 6 16 67 6 14 66 6 15 68 6 12 Vascular risk factors BP systolic 144 6 18 154 6 20 a 153 6 22 a 155 6 21 a BP diastolic 82 6 9 84 6 13 86 6 11 a 85 6 12 Hypertension 59 (28%) 32 (54%) 17 (33%) 21 (47%) Hypercholesteremia 49 (23%) 18 (30%) 12 (24%) 16 (36%) Diabetes Mellitus 19 (9%) 7 (12%) 4 (8%) 7 (16%) MRI characteristics GCA score 1.2 6 0.6 1.4 6 0.7 1.2 6 0.7 1.3 6 0.7 MTA score 1.4 6 0.8 1.9 6 0.8 a 1.5 6 0.8 b 1.7 6 0.9 a Lacunes 1 (0.5%) 4 (21%) a 0 b 3 (18%) a, c WMH score 0.7 6 0.5 2.3 6 0.4 a 0.7 6 0.5 b 2.5 6 0.5 a, c Number of MBs 0 0 1 (1-80) 2 (1-120) ApoE ε4 status Noncarrier 62 (35%) 12 (38%) 14 (33%) 7 (24%) Heterozygous carrier 85 (48%) 14 (44%) 16 (37%) 11 (38%) Homozygous carrier 32 (18%) 6 (19%) 13 (30%) 11 (38%) CSF values CSF Aß42 488 6 169 467 6 172 447 6 130 420 6 103 a CSF TAU 678 6 493 644 6 369 760 6 540 561 6 293 Abbreviations: MB: microbleeds; WMH: white matter hyperintensities; MMSE: Mini mental state examination; Camcog: Cambridge examination of cognitive disorder in elderly; BP: blood pressure; GCA: global cortical atrophy; MTA: medial temporal lobe atrophy; CSF: cerebrospinal fluid. Data are represented as number of patients with variable present (%), mean 6 SD and for number of MBs as median (range). MTA score was av- eraged (left and right). WMH score was assessed by using the Fazekas visual rating scale. Availability for incomplete data: MMSE: 367/371; Camcog: 354/371; BP: 283/371; Hypertension: 365/371; Hypercholesteremia and Diabetes Mellitus: 364/371; Lacunes: 370/371; ApoE ε4: 283/371; CSF: 239/371. d: P¼0.07, compared to MB+WMH-. a P<0.05 compared to MB-WMH-. b P<0.05 compared to MB-WMH+. c P<0.05 compared to MB+WMH-. O4-02-05 NEUROPATHOLOGIC BASIS OF WHITE MATTER HYPERINTENSITYACCUMULATION Deniz Erten-Lyons, Randy Woltjer, Hiroko Dodge, Sarah Stanfield, Lindsay Reese, Huong Tran, Joseph Quinn, Kathryn Wild, Barry Oken, Jeffrey Kaye, Lisa Silbert, Oregon Health and Sciences University, Portland, Oregon, United States. Background: The neuropathologic basis of white matter hyperintensity (WMH) volume accumulation observed on brain magnetic resonance imaging (MRI) is not known. Few studies have examined the neuro- pathologic correlates of cross-sectional WMH volumes and longitudinal WMH trajectories. Methods: Sixty-six elderly participants in the Ore- gon Brain Aging Study were included for having an autopsy, >1 Oral Sessions: O4-02: Neuroimaging: Imaging of Vascular Pathology and Alzheimer’s Disease P617