The ALS Stratification Prize-Using the Power of Big Data and Crowdsourcing for Catalyzing Breakthroughs in Amyotrophic Lateral Sclerosis (ALS) (P5.102)

摘要: Objectives: To develop novel tools to stratify ALS patients Background: Amyotrophic lateral sclerosis is a neurodegenerative disease with significant clinical heterogeneity. The heterogeneity of the patient population presents substantial barrier understanding mechanisms and planning interpretation trials, leading large, expensive, potentially unbalanced trials. Recently, new data sources become available, allowing development quantitative address heterogeneity. Methods: The DREAM Stratification Prize4Life Challenge crowdsourcing initiative that aims problem regards important targets such as ALSFRS progression survival. In challenge, which ran on summer 2015, we asked participants derive meaningful subgroups patients, related targets, along features characterize them. challenge used from Pool Resource Open-Access Clinical Trial (PRO-ACT) database, well National registries Italy Ireland. Results: The drew in ~300 70+ final submissions 31 teams worldwide based statistical machine learning approaches. Three significantly outperformed baseline approaches were declared top performers. best performing algorithms are able either patients potential interest. addition, solutions substantially reduce number needed for an trial, improve selection. also identify first time several prognostic blood test differentiated according prognosis interest. Conclusions: These results demonstrate value large datasets challenges developing better natural history, factors variables improving development. Disclosure: Dr. Zach has nothing disclose. Kueffner Atassi received personal compensation activities Biogen Idec consultant. Chio Idec, Cytokinetics, Italfamaco. Cudkowicz GlaxoSmithKline, Teva Neuroscience, Cytokinetics. Hardiman royalty payment Royal College Surgeons Ireland. Stolovitzky