The role of MRI in clinical trials of multiple sclerosis: comparison of image processing techniques

Magnetic resonance imaging (MRI) has received considerable attention in recent years over its potential for providing indices of multiple sclerosis activity and progression in clinical trials of new pharmaceuticals. The perceived advantages of MRI-derived measurements include greater objectivity, sensitivity, and reproducibility when compared with clinical rating scales. Clinical scales are also somewhat biased toward lesions affecting locomotion. However, the myriad permutations of MRI acquisition parameters, analysis methodologies, and disease indices demand careful consideration when employing MRI. Moreover, the use of MRI in research into the basic mechanisms of a disease may have different requirements than its use in a clinical trial setting. Consequently, a conference was held, sponsored by the US and Canadian multiple sclerosis societies, to review the present status of various MRI processing strategies and their potential role in clinical trials. Thirteen laboratories from North America and Europe as well as regulatory agencies and statistical consultants made formal presentations followed by extended discussion. This report presents the conclusions reached and recommendations for further action that emerged from the meeting.