Use of Quantile Regression to Characterize Solanezumab Effects across Percentiles of Disease Progression in EXPEDITION Alzheimer's Disease Trials (P7.102)

OBJECTIVE: Use quantile regression to characterize solanezumab effects in Alzheimer’s disease (AD) patients based on degree of disease progression. BACKGROUND: Twenty-seven percent of patients with clinically defined mild-to-moderate AD enrolled in 2 solanezumab trials had baseline assessments of amyloid status. Of those, 17[percnt] were negative and did not exhibit progressive decline typical of AD. Because comparison of solanezumab- and placebo-treatment effects on clinical progression in amyloid positive versus negative patients in the full cohort was precluded, we developed a novel statistical strategy to examine treatment effects in patients who did and did not show typical disease progression. DESIGN/METHODS: Quantile regression was used to examine solanezumab- and placebo-treatment differences in a pooled dataset. Patients demonstrating clinical decline typical of AD were represented in upper clinical-change percentiles; patients with a clinical course atypical of AD were represented in lower percentiles. This approach modeled change from baseline on 2 primary measures: Alzheimer9s Disease Assessment Scale-Cognitive Subscale (ADAS-Cog14) and Alzheimer’s Disease Cooperative Study-Activities of Daily Living Inventory Instrumental domain (ADCS iADL). RESULTS: ADAS-Cog14 at Week 80: In patients with mild AD, solanezumab treatment effect was greater in the upper percentiles (AD-typical). Compared with placebo, solanezumab slowed disease progression by 1.27, 1.98, 2.55, and 3.07 points in the 30th, 40th, 60th, and 80th percentiles. In lower percentiles (AD-atypical), solanezumab showed less effect. ADCS-iADL at Week 80: In patients with mild AD, compared with placebo, solanezumab slowed disease progression by 0.47, 0.99, 1.79, and 1.69 points in the 30th, 40th, 60th, and 80th percentiles. In patients with moderate AD, solanezumab did not show effects across most percentiles on either measure. Conclusions: Results support design of current/future solanezumab trials to be limited to mild AD patients with evidence of amyloid pathology. Further, quantile regression is useful for retrospectively analyzing data when amyloid status was not uniformly determined. Disclosure: Dr. Chen has received personal compensation for activities with Eli Lilly & Company. Dr. Ma has received personal compensation for activities with Eli Lilly & Company as an employee. Dr. Sundell has received personal compensation for activities with Eli Lilly & Company as an employee. Dr. Alaka has received personal compensation for activities with Eli Lilly and Company as an employee. Dr. Alaka holds stock and/or stock options in Eli Lilly and Company. Dr. Schuh has received personal compensation for activities with Eli Lilly & Company as an employee. Dr. Raskin has received personal compensation for activities with Eli Lilly & Company as an employee. Dr. Dean has received personal compensation for activities with Eli Lilly & Company. Dr. Dean holds stock and/or stock options in Eli Lilly & Company.