Table 3

Effects of treatment on MRI end points (integrated analysis, intent-to-treat population, double-blind period)

sc IFN β-1a tiw n=171sc IFN β-1a qw n=175Delayed treatment
n=171
T2 lesions, mean (SE)
 Month 3621.16 (1.44)24.61 (1.62)25.79 (1.73)
  Rate ratio (95% CI) vs DT0.82 (0.69; 0.98)0.95 (0.80; 1.14)
Nominal p=0.031Nominal p=0.606
 Month 6021.46 (1.45)24.95 (1.62)26.19 (1.74)
  Rate ratio (95% CI) vs DT0.82 (0.69; 0.98)0.95 (0.80; 1.14)
Nominal p=0.029Nominal p=0.591
New T2 lesions, mean (SE)
 Month 360.47 (0.06)0.61 (0.07)1.14 (0.13)
  Rate ratio (95% CI) vs DT0.41 (0.30; 0.56)0.53 (0.39; 0.73)
Nominal p<0.001Nominal p<0.001
 Month 600.55 (0.07)0.66 (0.08)1.11 (0.13)
  Rate ratio (95% CI) vs DT0.49 (0.36; 0.68)0.59 (0.43; 0.81)
Nominal p<0.001Nominal p<0.001
New Gd+ lesions, mean (SE)
 Month 360.14 (0.03)0.29 (0.05)0.74 (0.13)
  Rate ratio (95% CI) vs DT0.18 (0.11; 0.30)0.39 (0.24; 0.63)
Nominal p<0.001Nominal p<0.001
 Month 600.16 (0.03)0.28 (0.05)0.69 (0.13)
  Rate ratio (95% CI) vs DT0.23 (0.14; 0.38)0.41 (0.25; 0.67)
Nominal p<0.001Nominal p<0.001
New T1 hypointense lesions, mean (SE)
 Month 360.33 (0.05)0.47 (0.06)0.73 (0.09)
  Rate ratio (95% CI) vs DT0.45 (0.32; 0.65)0.64 (0.45; 0.91)
Nominal p<0.001Nominal p=0.012
 Month 600.35 (0.05)0.49 (0.06)0.70 (0.09)
  Rate ratio (95% CI) vs DT0.50 (0.35; 0.70)0.70 (0.50; 0.97)
Nominal p<0.001Nominal p=0.035
  • Data are per patient per scan unless stated otherwise. The negative binomial regression model is corrected for overdispersion with treatment and randomisation stratification factors as covariates and number of scans as an offset variable. Month 60 statistical analyses were exploratory; therefore, all p values for this time point were nominal (p values <0.05 are reported).

  • DT, delayed treatment; Gd+, gadolinium-enhancing; IFN, interferon; qw, once weekly; sc, subcutaneous; tiw, three times weekly.