Detection of Unusually Large Increases in MRI Lesion Counts in Multiple Sclerosis Patients

• Activity Number: 470
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Yinshan Zhao*+ and John Petkau and David Li and Andrew Riddehough and Anthony Traboulsee
• Companies: University of British Columbia and University of British Columbia and University of British Columbia and University of British Columbia and University of British Columbia
• Keywords: First-order Markov dependence ; longitudinal count data ; negative binomial ; safety monitoring in clinical trials ; semi-parametric estimation ; random effects models

Abstract

An increase of contrast enhancing lesions on repeated magnetic resonance imaging has been used as an indicator for potential adverse events in multiple sclerosis clinical trials. However, it has not been clearly identified what should be considered an 'unexpected increase' of lesion activity for an individual patient. We consider as an index the likelihood of observing lesion counts as large as those observed on the recent scans of a patient conditional on the patient's lesion counts on previous scans. To estimate this index, we develop models with patient specific-random effects. Given the random effect, we assume that the repeated lesion counts from the same patient follow a negative binomial distribution and may be correlated over time. We propose to fit the model using data collected from the trial under review and update the estimation whenever new data become available. We consider two different estimation procedures: maximum likelihood for a parametrized model and a semi-parametric method for a model with an unspecified distribution for the random effects. The performance of our methods are examined using simulations and illustrated using data from a clinical trial.