symptomatic fAD patients and 40 controls, we acquired T1w-MRI and DWI at two time points, approximately one year apart. Each DWI (2x64 direc- tions, 7 b0) was used to create a tensor image which was rigidly aligned to its corresponding T1w. Each follow-up pair of images was non-linearly aligned to the corresponding baseline pair using the proposed multi-modal registration technique. The resulting longitudinal maps were normalized to an inter-subject average for comparison. Voxel-wise statistical analysis was performed using permutation test (n¼5000) with age and gender as covari- ate. For comparison, we performed the same experiment using only the T1w scans. Results: Figure 1-bottom shows areas of significant difference in rate of volume change. It can be noticed that using the proposed multi-modal method, significant areas in the white matter are removed. We argue that vol- ume change differences in the white matter obtained from T1w MRI only are a consequence of the registration regularisation and are thus erroneous findings. Conclusions: We present a morphometric analysis based on multi-modal image registration. This technique could provide data driven findings that are less susceptible to error introduced by algorithmic choices of the registration. All software used to perform this analysis has been made open-source to enable others to conduct similar experiments. P4-134 WHITE MATTER HYPERINTENSITIES PREDICT MILD COGNITIVE IMPAIRMENT AND DEMENTIA IN PATIENTS WITH SUBJECTIVE COGNITIVE COMPLAINTS Marije Renske Benedictus 1 , Argonde Corien Van Harten 2 , Philip Scheltens 3 , Frederik Barkhof 4 , Niels Prins 5 , Wiesje M. Van der Flier 4 , 1 VUMC Alzheimer Centre Amsterdam, Amsterdam, Netherlands; 2 VUMC Alzheimer Center Amsterdam, Amsterdam, Netherlands; 3 VUMC, Amsterdam, Netherlands; 4 VU MC, Amsterdam, Netherlands; 5 VUMC Alzheimer Center, Amsterdam, Netherlands. Contact e-mail: m.benedictus@vumc.nl Background: Cerebral small vessel disease (SVD) is one of the leading causes of cognitive decline. We assessed the prevalence and predictive value of SVD with regard to clinical decline in a well-characterized set of patients with subjective cognitive complaints. Methods: We included 334 patients (age 62 + 10yrs, 157[47%] females) with subjective cognitive complaints (criteria for mild cognitive impairment [MCI] not fulfilled) from the mem- ory clinic based Amsterdam Dementia Cohort. Baseline MRI was used to determine severity of white matter hyperintensities (WMH; Fazekas scale 0-3;no/mild/moderate/severe), number of lacunes (0/1/2/>3) and number of microbleeds (0/1/2/>3). Additionally, the presence of medial temporal lobe atrophy (MTA) was determined. We performed univariate Cox propor- tional-hazard analysis to investigate the predictive value of WMH, lacunes and microbleeds on progression to MCI and all-cause dementia. We repeated the analysis with restricting clinical progression to MCI and de- mentia due to Alzheimer’s disease (AD), and in addition, to MCI and non-AD dementia. Results: After a mean follow-up of 3 + 2 years 53 (16%) patients progressed clinically to MCI (38) or dementia (AD:8;FTD:3;VAD:3; unspecified:1). Progressors were older (p<0.01) and had a longer duration of follow-up (p<0.05). Hundred seventy-six (53%) presented with some degree of WMH, but only 10 (3%) had severe WMH. Microbleeds were present in 49 (15%) and lacunes in 27 (8%) pa- tients. Compared to no WMH, mild WMH (HR:2.8;95% CI 1.2-6.3) and se- vere WMH (HR:4.2;95%CI 1.2-13.8) predicted clinical progression in age and sex adjusted models (Table 1; model 1). Adjustment for MTA did not change these result (model 2). Microbleeds and lacunes were no predictors. When clinical progression was restricted to MCI or non-AD dementia, re- sults remained largely unaltered. When clinical progression was restricted to MCI or dementia due to AD, effect estimates remained comparable, but significance was lost. Conclusions: Half of the patients with subjective cognitive complaints have SVD on MRI. In these patients, WMH increase the risk of clinical progression to MCI and dementia. P4-135 INFLUENCE OFAMYLOID-BETA ACCUMULATION ON COGNITIVE IMPAIRMENT OF IDIOPATHIC NORMAL PRESSURE HYDROCEPHALUS Masahiko Bundo 1 , Takashi Kato 2 , Takashi Sakurai 1 , Akinori Nakamura 1 , Kengo Ito 1 , 1 National Center for Geriatrics and Gerontology, Obu, Japan; 2 National Center for Geriatrics and Gerontology, Obu, Japan. Contact e-mail: ibundou@ncgg.go.jp Background: It is known that idiopathic normal pressure hydrocephalus (iNPH) can comorbid with Alzheimer disease (AD). Previous studies, prob- ing amyloid (Ab) deposition by cortical biopsy during ventriculo-peritoneal shunt, have shown that Ab deposition was found in 40w45 % of iNPH pa- tients. However, cognitive changes associated with AD comorbidity has been not well elucidated. Therefore, the objective of this study was to inves- tigate the influence of Ab deposition on cognitive functions in iNPH patients using 11 C-PiB PET. Methods: We enrolled 42 iNPH and 235 AD patients. All iNPH patients underwent 11 C-PiB PET scans and both halves of them (21/42) were Ab(-) and Ab(+), respectively. We compared the scores of Mini-Mental State Examination (MMSE), Alzheimer Disease Assessment Scale (ADAS), Frontal Assessment Battery (FAB), Digit Span, Raven’s Colored Progressive Matrices, and logical memory (immediate and de- layed) of Wechsler Memory Scale Revised among these groups. Results: Results show that MMSE, ADAS, and logical memory scores were almost the same between Ab(+) iNPH and AD, however, Ab(-) iNPH showed significantly better performance than other groups. On the other hand, both Ab(-) and Ab(+) iNPH showed significantly lower scores of FAB than those in AD, suggesting frontal lobe dysfunction in iNPH irrespective of the amyloid deposition. Conclusions: Results of this study suggest that Ab deposition can affect several cognitive functions in iNPH patients. Co- morbidity of the AD should be taken into consideration when we evaluate the cognitive function of iNPH patients. P4-136 DOES HIPPOCAMPAL VOLUME PREDICT POSITIVE AMYLOID STATUS ONFLORBETAPIR- PET IN HEALTHY CONTROLS AND PRODROMAL STAGES OF ALZHEIMER’S DISEASE? Paula T. Trzepacz 1 , Yu Peng 1 , Peter Castelluccio 2 , Helen M. Hochstetler 1 , Abhinay Joshi 3 , Grazia Dell’Agnello 4 , Elisabeth K. Degenhardt 1 , Katherine J. Selzler 1 , Michael M. Witte 1 , Michael D. Devous 5 , Adam J. Schwarz 1 , 1 Eli Lilly and Company, Indianapolis, Indiana, United States; 2 Bucher & Christian Consulting, Inc, Philadelphia, Pennsylvania, United States; 3 Avid Radiopharmaceuticals, Inc., Philadelphia, Pennsylvania, United States; 4 Eli Lilly Italia SpA, Sesto Fiorentino, Italy; 5 Avid Radiopharmaceuticals, Inc., Philadelphia, Pennsylvania, United States. Contact e-mail: trzepacz_paula_t@lilly.com Background: Reduced hippocampal volume (HV) on MRI is associated with Alzheimer’s disease (AD). We evaluated HV prediction of amyloid Table 1 Cox proportional hazard models for clinical progression to MCI or all-cause dementia Model 1 Model 2 No WMH reference reference Mild WMH 2.8 (1.2; 6.3) 2.8 (1.2; 6.3) Moderate WMH 1.9 (0.7; 5.6) 2.0 (0.7; 5.6) Severe WMH 4.2 (1.2; 13.8) 4.1 (1.2; 14.0) No microbleeds reference reference 1 microbleed 0.4 (0.1; 1.4) 0.4 (0.1; 1.4) 2 microbleeds 0.6 (0.1; 4.7) 0.7 (0.1; 4.7) 3 microbleeds 2.1 (0.7; 5.7) 2.0 (0.7; 5.8) No lacunes reference reference 1 lacune 0.7 (0.2; 2.7) 0.7 (0.2; 2.8) 2 lacunes 0.6 (0.1; 4.1) 0.5 (0.1; 4.0) 3 lacunes 1.7 (0.4; 7.1) 2.0 (0.4; 6.9) Abbreviations: WMH: white matter hyperintensties, categories are based on the Fazekas scale (0-3). Univariate models with data represented as HR (95%CI), with the first category (no WMH/ microbleeds/lacunes) as a reference category. Model 1: Univariate, adjusted for age and gender; Model 2: Univariate, adjusted for age, gender and medial temporal lobe atrophy. Poster Presentations: P4 P836