The objective of this study is to characterize the timing and extent of radiologic MS disease recurrence during the 24-week natalizumab treatment interruption period in RESTORE. RESTORE was a randomized, partially placebo-controlled exploratory study. Natalizumab-treated patients with no gadolinium-enhancing (Gd+) lesions at screening (n = 175) were randomized 1:1:2 to continue natalizumab (n = 45), switch to placebo (n = 42), or switch to other therapies (n = 88) for 24 weeks. MRI assessments were performed every 4 weeks. Predictors of increased numbers of Gd+ lesions during natalizumab treatment interruption were evaluated. The numbers of Gd+ lesions were compared with retrospectively collected pre-natalizumab MRI reports and data from placebo-treated patients from two historical randomized clinical trials. Gd+ lesions were detected in 0 % (0/45) of natalizumab patients, 61 % (25/41) of placebo patients, and 48 % (39/81) of other-therapies patients during the randomized treatment period. Gd+ lesions were detected starting at week 12; most were observed at week 16 or later. Thirteen percent (14/107) of patients had >5 Gd+ lesions on ≥1 (of 6) scans during the randomized treatment period versus 7 % (7/107) of patients pre-natalizumab (based on medical record of a single scan). Younger patients and those with more Gd+ lesions pre-natalizumab were more likely to have increased MRI activity. Distribution of total and persistent Gd+ lesions in RESTORE patients was similar to placebo-treated historical control patients. In most patients, recurring radiological disease activity during natalizumab interruption did not exceed pre-natalizumab levels or levels seen in historical control patients.