Beyond the Group-Level Analysis: a Flexible Mixed Effect Negative Binomial Regression Model for Detecting Unusual Increases in MRI Lesion Counts in Multiple Sclerosis Patient

• Activity Number: 28
• Type: Contributed
• Date/Time: Sunday, August 3, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Yumi Kondo*+ and Yinshan Zhao and Albert John Petkau
• Companies: University of British Columbia and University of British Columbia and University of British Columbia
• Keywords: longitudinal count data ; nonparametric Bayesian ; Dirichlet process ; negative binomial ; random effects model ; safety monitoring in clinical trial

Abstract

An increase of contrast enhancing lesions (CELs) on repeated magnetic resonance imaging has been used as an indicator for potential adverse events in multiple sclerosis (MS) clinical trials. This presentation discusses enhancements for previously proposed method (Zhao et al., (2013)) for identifying unusual increases in CEL activity for individual patients. The procedure signals such patients by estimating the probability of observing CEL counts as large as those observed on a patient's recent scans conditional on the patient's CEL counts on previous scans. This probability index is computed based on a mixed effect negative binomial regression model. As the index values can vary depending on the choice of distribution for the patient-specific random effects (RE), we relax this restrictive RE assumption and model the REs with an infinite mixture of betas, using the Dirichlet process, which effectively allows any form of distribution. As our inference is in the Bayesian framework, we also discuss how to develop an informative prior based on previous clinical trials. This enhanced method is illustrated with CEL data from ten previous MS clinical trials.