items and factor analysis-based subscales. Factors were derived using an un- structured covariance matrix of change-from-Baseline item values (placebo and memantine groups combined), a varimax rotation, loading of at least 0.30 for each item, and eigenvalues of at least 1 for each factor. Between- group comparisons (a ¼ 0.05; OC, LOCF) were performed on the intent- to-treat population (placebo: n ¼ 328; memantine ER: n ¼ 333) using an ANCOVA model, with treatment group and study center as factors and base- line value as covariate. In addition, an MMRM analysis that included terms for treatment group, visit, treatment-by-visit interaction, baseline score, baseline-by-treatment interaction, and center was used to compare the groups across the entire trial. No adjustments were made for multiple com- parisons. Results: A significant advantage of memantine ER over placebo was observed for the items of eating (OC only, P ¼ 0.024), clearing the table (OC only, P ¼ 0.010) and finding belongings (OC only, P ¼ 0.029). No other ADL items demonstrated significant between-group differences. Factor analysis yielded 4 subscales: ADLs (eating, walking, toileting, bathing, grooming, dressing), Higher-Level Functions (using telephone, conversing, clearing the table, finding belongings, obtaining beverage, disposing of lit- ter), Praxis (turning faucet on, off, turning light on, off), and Autonomy (watching TV, travelling, being left alone). At study endpoint, the meman- tine ER group significantly outperformed the placebo group on the factor as- sociated with higher-level functions (OC, P ¼ 0.011; LOCF, P ¼ 0.026; MMRM, P ¼ 0.008). Conclusions: In this post-hoc analysis, memantine ER was associated with improvements in ADLs related to higher cognitive processing compared to placebo. P4-245 INTRAVENOUS IMMUNOGLOBULIN IMPROVED COGNITIVE FUNCTIONS IN PATIENTS WITH ALZHEIMER’S DISEASE WITHOUT ANY CHANGES IN PIB RETENTION IN THE BRAIN Masaki Kondo 1 , Takahiko Tokuda 1 , Takashi Kasai 1 , Yoko Oishi 1 , Suzuka Ataka 2 , Hiroyuki Shimada 2 , Takami Miki 2 , Hiroshi Mori 2 , Masanori Nakagawa 1 , 1 Kyoto Prefectural University of Medicine, Kyoto, Japan; 2 Osaka City University Graduate School of Medicine, Osaka, Japan. Background: Immunotherapy is one of the strategies undertaken as a rad- ical treatment for Alzheimer’s disease (AD). Passive immunization with antibodies against amyloid ß peptide 1-42 (Aß42) could inhibit the AD- like neuropathology in APP-transgenic mouse models of AD. There have been some reports that intravenous immunoglobulin (IVIg) could be use- ful for passive immunotherapy for AD because it contains natural poly- clonal anti-Aß antibodies. We carried out an open-label study treating 4 Japanese patients with mild AD with IVIg for 3 months. Methods: All patients were given 0.4 g/kg of IVIg for 3 consecutive days (1 session of IVIg therapy) every month for 3 months (3 sessions of therapy for a total of 3.6 g/kg IVIg). All patients underwent neurological examinations in- cluding cognitive testing and examinations of blood and cerebrospinal fluid (CSF) were carried out to determine baseline levels of functioning. Tests were conducted during each IVIg administration and 1-month after each treatment, and were followed up every couple of months. PIB-PET studies were carried out at baseline and 1 month after the treatment in all patients. Results: Infusions were generally well-tolerated. However, one patient de- veloped generalized seizure after the first session of IVIg therapy. For this patient, IVIg therapy was discontinued. He did not show any signs of en- cephalitis in CSF or MRI examinations. The remaining 3 patients who re- ceived 3 sessions of IVIg therapy showed improvement in scores of cognitive tests after 3 sessions of IVIg compared with scores at baseline. The mean (SD) improvement in MMSE score was 4.3 (2.9). When the pa- tients were evaluated more than 6 months after treatment, two of the four patients showed progressive deterioration in cognitive functions, but the other 2 patients whose MMSE score had improved to the normal range with IVIg remained stable in cognitive functions. No significant change in the PIB retention was observed in any patient after the IVIg treatment compared with the baseline. Conclusions: In this small study, we showed that IVIg therapy has the potential to improve cognitive functions without any changes in PIB retention in their brains. Furthermore, the improving effect of IVIg on cognitive functions continued for more than 6 months in two patients who had been good responders of IVIg, suggesting the more sustained improvement by IVIg could be achieved in patients with the milder AD. P4-246 THE SOUVENIR II STUDY: RATIONALE AND STUDY DESIGN Bruno Vellas 1 , Patrick Kamphuis 2 , John Harrison 3 , Peter de Deyn 4 , Christine von Arnim 5 , Rafael Blesa 6 , Elio Scarpini 7 , Anke Bongers 2 , Rico Wieggers 2 , Philip Scheltens 8 , 1 Gerontopole, INSERM U1027, University of Toulouse III, Toulouse, France; 2 Nutricia Advanced Medical Nutrition, Danone Research, Centre for Specialised Nutrition, Wageningen, Netherlands; 3 CogState Ltd, Warminster, United Kingdom; 4 ZNA Middelheim, University of Antwerp, Antwerpen, Belgium; 5 University Hospital Ulm, Ulm, Germany; 6 Hospital de la Santa Creu i Sant Pau, Barcelona, Spain; 7 Ospedale Maggiore Policlinico, University of Milan, Milan, Italy; 8 Alzheimer Centre, VU University Medical Centre, Amsterdam, Netherlands. Background: Increasing evidence suggests a therapeutic potential for nutri- tion in Alzheimer’s Disease (AD). The multi-nutrient drink SouvenaidÒ, containing a specific combination of nutrients (FortasynTM Connect), is formulated to improve synapse formation and is intended for the dietary management of AD1. The recent proof of concept ‘Souvenir I’ study in 225 drug naive mild AD patients demonstrated that Souvenaid improved memory performance, both in the overall intention to treat population (base- line MMSE 20-26) and in the prespecified subgroup of patients with very mild AD (baseline MMSE 24-26)(Scheltens et al., 2010). Furthermore, the Souvenir I study showed that the use of Souvenaid was safe and well tol- erated throughout 24 weeks of supplementation. To confirm and extend the results of this study, three clinical studies started in 2009. Of these, the Sou- venir II study (NTR1975) was developed with the primary aim to assess the effect of Souvenaid on memory performance in drug naive patients with (very) mild AD. 1 Souvenaid and Fortasyn are registered trademarks of N.V. Nutricia. Methods: The Souvenir II study is a 24-week, randomized, controlled, double-blind, parallel-group, multicentre multi-country study in 259 mild AD patients (MMSE ¼ 20). Primary outcome measure is the memory domain score resulting from aneuropsychological test battery (NTB) (based on the previously validated NTB (Harrison et al., 2007)). The NTB executive function domain score, the totalcomposite score and the individual item scores are secondary study parameters. The Souvenir II NTB consists of the Rey Au- ditory Verbal Learning Test (immediate and delayed recall and recognition); WechslerMemory Scale (WMS) verbal paired associates (immediate and de- layed recall); WMSDigit Span; Trail Making Tests part A and B; Category Flu- ency; Controlled Word Association Test; the ADAS-cog orientation task and the Letter DigitSubstitution Test. Other study outcome measures include safety and tolerance, the Disability Assessment for Dementia scale, blood bio- chemistry andelectroencephalography (EEG). EEG isused to explore effects on brain electrical activity and neuronal network organization. Results: First results are expected to be available end 2011. Conclusions: Within the Sou- venaid Clinical Trial program, the Souvenir II study investigates the effect of Souvenaid on memory performance (NTB) in drug naive patients with (very) mild AD. P4-247 ENDOVASCULAR LOW-ENERGY LASER RADIATION EFFECT ON DYSCIRCULATORY ANGIOPATHY OFALZHEIMER’S TYPE IN THE TREATMENT OF ALZHEIMER’S DISEASE Ivan Maksimovich 1 , 1 Clinic of Cardiovascular Diseases named after Most Holy John Tobolsky, Moscow, Russia. Background: One of the reasons contributing to AD development is the dyscirculatory angiopathy of Alzheimer’s type (DAAT). Hereupon are pre- sented the results of endovascular treatment method aimed at effecting Poster Presentations P4 S791