P2‐195: Cerebral microbleeds in a phase 2 clinical trial of mild‐to‐moderate Alzheimer's disease with the gamma secretase inhibitor BMS‐708163

Clinical trials of Aß lowering treatment in AD have directed attention to the prevalence/incidence of cerebral microbleeds (CMBs). The clinical significance of CMBs in AD and their possible association with amyloid lowering treatment needs investigation. This study determined the prevalence, incidence and risk factors for CMBs in a phase 2 dose finding and safety study of BMS-708163, a gamma secretase inhibitor designed for selective inhibition of amyloid beta (Aß) synthesis. Post-hoc analysis of a phase 2, six month placebo controlled multiple dose arm study of BMS-708163 in mild-to-moderate AD (MMSE 16-26, Hachinski score &< 4). This analysis included n = 175 subjects having centrally read MRI scans at baseline and week 24. The MRI protocol included a sagittal millimetric 3D T1 sequence, and axial T2*, T2 FLAIR and T2 Spin Echo sequences with 5 mm contiguous slices and no gap. CMBs were defined as focal T2* hypointensities of <10 mm. Analytic methods included descriptive statistics, with Fisher's Exact test, t-tests and Analysis of Covariance for comparison tests. At baseline, 26% of subjects had CMBs, 19% having one CMB and 7% having > 2 (range up to 122). Risk factors included a history of stroke and Apo E genotype. The incident rate of CMBs was 14% without significant differences (p>0.4) amongst the treatment arms (10% placebo, 13-19% across BMS-708163 treatment groups). The risk of incident CMBs was associated with the presence of baseline lesions. Of subjects with incident CMBs, 50% had 1, while 17% had >3 new lesions. There were no significant differences on clinical outcomes between the incident CMB + and CMB - groups nor within the aggregated BMS-708163 treatment group evaluated by incident CMB status (p>0.14 at Week 24). The prevalence rate and risk factors of CMBs in this trial (history of stroke, ApoE genotype) are consistent with other clinic studies, while the incident rate is higher. Treatment with BMS-708163 did not appear to be associated with an increased risk of incident CMBs, or with adverse clinical effects. However, this study sample is limited in size. More data are needed to further evaluate these potential relationships.