p¼0.013), followed by “Disorientation to Time” (11.3, 1.61-78.57, p¼0.015), and “Interest/Initiative” (9.0, 1.42-57.12, p¼0.020). Conclusions: A 2-point change in MMSE correlated with changes in individualized SG-D symptoms in people with VaD or AD/ VaD. Outcome measures, such as the SG-D, that are inherently clin- ically meaningful to patients and their caregivers, can help translate from clinical trials into everyday dementia practice. P3-023 IN PURSUIT OF A SENSITIVE EEG FUNCTIONAL CONNECTIVITY OUTCOME MEASURE FOR CLINICAL TRIALS IN ALZHEIMER’S DISEASE Casper T. Briels 1 , Cornelis J. Stam 2 , Philip Scheltens 1 , Suzanne Bruins 3 , Inge Lues 3 , Alida A. Gouw 2 , 1 Alzheimer Center Amsterdam, Department of Neurology, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, Netherlands; 2 Department of Clinical Neurophysiology, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, Netherlands; 3 Probiodrug AG, Halle, Germany. Contact e-mail: c.briels@vumc.nl Background: Normal cognitive function relies on efficient commu- nication between brain regions. In treatment trials in Alzheimer’s Disease (AD), an improvement of these impaired functional inter- actions could provide biological support for the potential efficacy of the drug. In the phase 2a SAPHIR-trial in early AD with PQ912, a glutaminylcyclase inhibitor, electroencephalography (EEG) analysis showed a significant improvement of global relative theta power in the intervention group compared to placebo. Howev- er, functional connectivity (FC) measured by the global phase lag index (PLI) did not improve. PLI may not be sensitive enough for the detection of small effects. Therefore, we compared the PLI with the amplitude envelope correlation with leakage correction (AEC-c), a presumably more sensitive FC measure to capture po- tential treatment effects. Methods: Data was used from the multi- centre SAPHIR-trial. Patients with mild cognitive impairment to mild dementia both due to AD underwent 12 weeks of placebo or PQ912 treatment. Eyes-closed task free EEG was measured at base- line and follow-up (PQ912 n¼47, placebo n¼56). FC was measured using AEC-c and PLI in the delta (0.5-4Hz), theta (4- 8Hz), alpha (8-13Hz) and beta (13-30Hz) frequency band. Change in FC was compared between treatment groups by t-test and AN- COVA using two models of covariates. Pearson’s correlation coef- ficients were calculated between the change in FC measures and global relative theta power. Results: A significant difference in change of global AEC-c in the alpha frequency band was found be- tween treatment groups (mean change: placebo¼-0.004760.0189 versus PQ912¼0.008560.0252, t-test: p¼0.004, Cohen’s d¼0.58) indicating an increased global FC after treatment. The ef- fect remained significant when corrected for sex, country, ApoE ε4 carriage, age, baseline value (model 1; p¼0.006) and in addition change in relative alpha power (model 2; p¼0.004). There was no correlation between change in global relative theta power and global AEC-c in the alpha band (r¼0.152, p¼0.124). No significant differences between treatment groups were found using the PLI. Conclusions: Functional connectivity, measured with AEC-c in the alpha frequency band, significantly improved after PQ912 treat- ment in early AD, independent of baseline functional connectivity and relative power. AEC-c may be a robust and sensitive FC mea- sure for detecting treatment effects. P3-024 DIETARY FIBRE AND PROTEIN INTAKE AND ASSOCIATION WITH APOLIPOPROTEIN 3 4 ALLELE Warnakulasuriya Fernando 1 , Stephanie R. Rainey-Smith 2 , Samantha L. Gardener 3 , Ralph N. Martins 2,4 and Australian Imaging Biomarkers and Lifestyle Study of Ageing (AIBL), 1 Edith Cowan University, Perth, Western Australia, Australia; 2 Edith Cowan University, Joondalup, Western Australia, Australia; 3 Edith Cowan University, Perth, Australia; 4 Macquarie University, North Ryde, NSW, Australia. Contact e-mail: w.fernando@ecu.edu.au Background: This cross sectional study examined whether intake of fibre and protein, are associated with apolipoprotein E allele, age and gender using Australian Imaging, Biomarkers and Lifestyle study of ageing [AIBL] cohort, which has not being studied previ- ously. Methods: The AIBL study is a longitudinal study of 1112 volunteers including healthy controls (HC). Of the baseline, 692 HCs completed the Cancer Council of Victoria Food Frequency Questionnaire. APOE genotype was determined using polymerase chain reaction amplification and restriction enzyme digest tech- niques. Results: The average age was 69.84 years and mean daily protein and fibre intake was 84.13 and 22.96 g/day respectively. The percentage of carriers of the APOE 3 4 allele and male was 26% and 41.6% respectively. When APOE was used as a categori- cal variable, there were 2.962 and 2.590 higher odds of being in the lowest and middle intake tertile compared to the highest intake pro- tein tertile (p < 0.05), if APOE 3 4 was positive. Additionally, there were 2.340 times more likely to consume the lowest amount of pro- tein, than males by females, if APOE ε4 were present. However, ef- fect of the APOE ε4 allele was lower with increasing age of female on the intake of protein. Conclusions: Our results suggest that APOE ε4 carriers are more likely to have low to medium protein intakes and this is more significant with females than males. There were no associations between fibre intake and APOE and gender. The low intake of protein in females aged  72 years is likely due to genetic or environmental factors other than the APOE ε4. These data suggest that especially women, may benefit by consuming higher protein to prevent diseases that may be associ- ated with the presence of APOE ε4 allele. Although our previous evidence suggested that individuals that undertake higher levels of protein are less likely to be amyloid positive [1], our current work indicated that age, APOE and gender may influence on the intake of protein. Reference: Fernando W, et al (2018) Associations of Dietary Protein and Fiber Intake with Brain and Blood Amyloid- beta. J Alzheimers Dis 61, 1589-1598. P3-025 USE OFA MODIFIED HIDDEN PATHWAY MAZE TEST IN PATIENTS WITH ALZHEIMER’S DISEASE DEMENTIA Chris J. Edgar 1 , Tim Tasker 2 , Adrian Schembri 3 , Babli Millais 2 , Bharat Ruparelia 2 , Odile Dewit 2 , Stephen Jones 2 , Paul Maruff 3 , Pradeep J. Nathan 2 , 1 Cogstate Ltd., London, United Kingdom; 2 Sosei Heptares, Cambridge, United Kingdom; 3 Cogstate Ltd., Melbourne, Australia. Contact e-mail: cedgar@cogstate.com Background: Maze tests have a long history as measures of visuo- spatial learning and executive function. The importance of impair- ments in both visuospatial learning and executive dysfunction has been increasingly recognized in early AD. In contrast to presented Poster Presentations: Tuesday, July 16, 2019 P934