A Clinical Trial Enrichment Tool Based on Subgroups Defined by Machine Learning Predictive Models

ABSTRACT Neurodegenerative disordersincluding amyotrophic lateral sclerosisALSFriedreichandapos s ataxiaFAmultiple sclerosisMSDuchenne muscular dystrophyDMDAlzheimer s diseaseADParkinson s diseasePDand Huntington s diseaseHDare characterized by heterogeneous disease progressionEfforts to identify responder subgroups may uncover subgroups that are more homogeneous in disease related features than the full study populationAs a resulta subgroup may exhibit a statistically significant effect sizeHowevercurrent methodologies for subgroup analysis are limited by the relatively small number of prognostic and predictive indicators that can be used to describe subgroupsThese methods are not well suited to describing subgroups with reduced heterogeneity in disease progressionor in identifying indicators for multifactorial diseasesWe have developed and submitted a patent application for a novel subgroup analysis method based on grouping participants with similar predicted disease progression profiles and analyzing nearest neighbor subgroups within a clinical trialWe call this method Detectable Effect Cluster DECanalysisIn our phaseand phaseSBIR grantswe used ALS as a model disease to develop our API product that uses machine learning disease models to improve trial arm randomization and provide covariates for statistical analysisIn the ongoing phasegrant we are expanding our disease offerings to include ADPD and HDBuilding on a set of ALS disease progression models that we have previously developed and validatedwe seek in this grant application to develop a novel prototype machine learning based subgroup analysis application that we plan on adding to our product offeringsDuring this proposed phasegrantwe will address research level questions regarding the nature of the subgroups defined using DEC analysis including how to define confidence intervals of our DEC clustersand estimated bounds for using prediction thresholds as selection criteria for a confirmatory clinical trialFinallywe will apply DEC analysis to three publicly available clinical trial data sets in an attempt to identify subgroups with significant treatment effectsAimWe will apply methods used in image analysis for identifying statistically significant subgroups to address themultiplicity issue inherent in DEC AnalysisAimWe will use statistical methods to model the confidence intervals of a power analysis in which DEC clusterbased selection criteria would be used for a confirmatory trialAimWe will isolate records from PRO ACT that include whether a patient was treated with riluzole and two otherpublicly available recent ALS datasets to test the application of DEC AnalysisOrigent s current suite of products will answer drug development needs of a full portfolio of neurodegenerative diseasesUltimatelywe see a series of machine learning applications aimed at solving drug development issues for multiple disease areasincluding orphan diseasesThese models and applications will vastly increase the speed and efficiency of drug developmentresulting in fastercheapermore efficient drug trials that yield numerous new medications to ease human pain and suffering NARRATIVE This work will develop Detectable Effect ClusterDECanalysisa novel machine learning based method of subgroup analysisDEC analysis shows great promise in identifying patient subgroups with statistically significant drug effects within largermore heterogeneousfailed therapeutic clinical trialsDEC analysis has the potential to rescue a drug that otherwise would have been discarded as a drug that does not provide therapeutic benefitwhenin factthe opposite is true