Outcomes | Natalizumab (n=54) | Fingolimod (n=54) | HR (95% CI)* | P value† |

MRI outcomes: T1 Gd+ lesions | ||||

Cumulative probability of developing new T1 Gd+ lesions over study, % | ||||

≥1 | 40.68 | 57.99 | 0.60 (0.31–1.16) | 0.126 |

≥2 | 11.54 | 48.48 | 0.25 (0.09–0.68) | 0.007 |

≥3 | 10.02 | 41.38 | 0.24 (0.08–0.77) | 0.016 |

Number of patients with new T1 Gd+ lesions from baseline to 24 weeks, n/N (%) | 16/47 (34.0)‡ | 24/45 (53.3)‡ | NA | 0.062 |

Change from baseline in T1 Gd+ lesion vol to 24 weeks, mean (SD) | 0.5 (31.2)§ | 1.8 (19.7)§ | NA | 0.532 |

MRI outcomes: T2 lesions | ||||

Number of patients with new/newly enlarging T2 lesions at 24 weeks, n/N (%) | 6/15 (40.0) | 10/16 (62.5) | NA | 0.21 |

Number of new/newly enlarging T2 lesions at 24 weeks per patient, mean (SD) | 1.3 (2.5)§ | 1.9 (2.2)§ | NA | 0.263 |

Change from baseline in T2 lesion volume to 24 weeks, mean (SD) | 0.1 (4.4)§ | 3.3 (5.0)§ | NA | 0.053 |

Relapse outcomes | ||||

Cumulative probability of relapse over study, %¶ | 1.9 | 22.3 | 0.08 (0.01–0.64)** | 0.017 |

ARR on study (95% CI) | 0.02 (0.00–0.13) | 0.20 (0.11–0.37) | 0.09 (0.01–0.72)†† | 0.023^{‡‡} |

*All HRs and rate ratios compare natalizumab to fingolimod.

†P value based on a Cox model adjusted for the baseline number of Gd+ lesions, age, baseline EDSS score and years since the first symptom (for the cumulative probability of new T1 Gd+ lesions during follow-up), from a χ

^{2}test between the two treatment groups (for the number of patients with new lesions) or based on a Wilcoxon rank-sum test between the two treatment groups (for the number of new/newly enlarging T2 lesions and changes in lesion volume).‡Includes patients with new T1 Gd+ lesions at any time point after baseline. Not all patients received treatment through 24 weeks.

§Natalizumab, n=15; fingolimod, n=16. Includes only patients who had MRI data through 24 weeks.

¶Cumulative probabilities at 36 weeks are reported, as no relapse events were observed after 36 weeks.

**Based on Cox model adjusted for the number of relapses in the year before baseline, age, baseline EDSS score and years since the first symptom.

††Value indicated is a rate ratio based on a negative binomial model of ARR with treatment as effect, adjusted for the number of relapses in the year before baseline, years since the first symptom, baseline EDSS score and baseline age.

‡‡P value based on a negative binomial model of ARR with treatment as effect, adjusted for the number of relapses in the year before baseline, years since the first symptom, baseline EDSS score and baseline age.

ARR, annualised relapse rate; EDSS, Expanded Disability Status Scale; Gd+, gadolinium enhancing; MS, multiple sclerosis; NA, not applicable.