MRI Lesion Burden in Multiple Sclerosis

Challenge

A multiple sclerosis trial's primary endpoint rested on quantitative MRI outcomes—new and enlarging T2 lesions, gadolinium-enhancing lesions, and brain volume change—each highly sensitive to acquisition consistency. Scanner software upgrades mid-trial threatened to introduce systematic measurement shifts, and spinal cord imaging quality varied considerably across the 52-site network. The sponsor required quantitative results comparable across the entire 36-month study duration.

Our Solution

Intrinsic Imaging harmonized acquisition protocols across all sites and enforced them through prospective quality control, rejecting non-conforming examinations before they reached analysis. Scanner software or hardware changes triggered standardized requalification with phantom and volunteer scanning to quantify any measurement shift. Central analysis applied validated lesion segmentation with expert neuroradiologist review of every examination, and brain volume measurements were computed under a controlled, version-locked analysis pipeline.

Results

• Scanner changes qualified without endpoint discontinuity
• Lesion counting reproducibility above 95%
• Quantitative MRI dataset supported the Phase III efficacy claim