Long-interval T2-weighted subtraction magnetic resonance imaging: a powerful new outcome measure in multiple sclerosis trials

Ann Neurol. 2010 May;67(5):667-75. doi: 10.1002/ana.21958.


Objective: To compare long-interval T2-weighted subtraction (T2w-Sub) imaging with monthly gadolinium-enhanced T1-weighted (Gd-T1w) imaging for (1) detection of active lesions, (2) assessment of treatment efficacy, and (3) statistical power, in a multiple sclerosis (MS), phase 2, clinical trial setting.

Methods: Magnetic resonance imaging (MRI) data over 9 months from 120 patients (61 treatment, 59 placebo) from the oral temsirolimus trial were used. T2w-Sub images were scored for active lesions, independent of the original reading of the monthly Gd-T1w images. Treatment efficacy was evaluated using the nonparametric Mann-Whitney U test, and parametric negative binomial (NB)-regression and power calculations were conducted.

Results: Datasets from 116 patients (58 treatment, 58 placebo) were evaluated. The mean number of T2w-Sub lesions in the treatment group was 3.0 (+/-4.6) versus 5.9 (+/-8.8) for placebo; the mean cumulative number of new Gd-T1w lesions in the treatment group was 5.5(+/-9.1) versus 9.1(+/-17.2) for placebo. T2w-Sub imaging showed increased power to assess treatment efficacy compared with Gd-T1w imaging, when evaluated by Mann-Whitney U test (p = 0.017 vs p = 0.177), or NB-regression without (p = 0.011 vs p = 0.092) or with baseline adjustment (p < 0.001 vs p = 0.002). Depending on the magnitude of the simulated treatment effect, sample size calculations showed reductions of 22 to 34% in the number of patients (translating into reductions of 81-83% in the number of MRI scans) needed to detect a significant treatment effect in favor of T2w-Sub imaging.

Interpretation: Compared with monthly Gd-T1w imaging, long-interval T2w-Sub MRI exhibited increased power to assess treatment efficacy, and could greatly increase the cost-effectiveness of phase 2 MS trials by limiting the number of patients, contrast injections, and MRI scans needed.

Publication types

  • Clinical Trial, Phase II
  • Multicenter Study
  • Randomized Controlled Trial
  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Brain / pathology
  • Dose-Response Relationship, Drug
  • Double-Blind Method
  • Female
  • Gadolinium
  • Humans
  • Image Processing, Computer-Assisted / methods
  • Longitudinal Studies
  • Magnetic Resonance Imaging / methods
  • Male
  • Middle Aged
  • Multiple Sclerosis / drug therapy*
  • Multiple Sclerosis / pathology*
  • Outcome Assessment, Health Care
  • Protein Kinase Inhibitors / administration & dosage*
  • Sirolimus / administration & dosage
  • Sirolimus / analogs & derivatives*
  • Statistics, Nonparametric


  • Protein Kinase Inhibitors
  • temsirolimus
  • Gadolinium
  • Sirolimus