The precision of T1 hypointense lesion volume quantification in multiple sclerosis treatment trials: a multicenter study.

Abstract

The volume of hypointense lesions on T1 weighted brain MRI represents an increasingly used MR endpoint in phase III MS treatment trials. In this study we evaluated the reproducibility of hypointense T1 lesion volume quantification in a cohort of Multiple Sclerosis (MS) patients. The gadolinium enhanced T1 weighted brain MR images of 33 MS patients from three European centers were used in this study. These images were acquired as part of a phase III trial of interferon beta-1b in secondary progressive MS. The MRI machine manufacturers and imaging parameters varied according to the MRI acquisition center. Three experienced observers used a semi-automated local thresholding technique to quantify the hypointense T1 lesion volume on two occasions, separated by a delay. The intra and inter observer coefficients of variation were 3.7% and 4.9% respectively, with similar values derived for images obtained at all three sites. There was a generally high level of agreement between the lesion volumes obtained by the three raters. However, a modest but significant measurement drift was identified between the first and second sessions for one of the three raters, highlighting the very real possibility of measurement drift even for experienced observers. Our results support the increasing role for T1 hypointense lesion volume as an outcome measure in multicenter phase III MS treatment trials. Multiple Sclerosis (2000) 6 237 - 240