Background Interferon beta is used to modify the course of relapsing multiple sclerosis. Despite interferon beta therapy, many patients have relapses. Natalizumab, an α 4 integrin antagonist, appeared to be safe and effective alone and when added to interferon beta-1a in preliminary studies. Methods We randomly assigned 1171 patients who, despite interferon beta-1a therapy, had had at least one relapse during the 12-month period before randomization to receive continued interferon beta-1a in combination with 300 mg of natalizumab (589 patients) or placebo (582 patients) intravenously every 4 weeks for up to 116 weeks. The primary end points were the rate of clinical relapse at 1 year and the cumulative probability of disability progression sustained for 12 weeks, as measured by the Expanded Disability Status Scale, at 2 years. Results Combination therapy resulted in a 24 percent reduction in the relative risk of sustained disability progression (hazard ratio, 0.76; 95 percent confidence interval, 0.61 to 0.96; P = 0.02). Kaplan–Meier estimates of the cumulative probability of progression at two years were 23 percent with combination therapy and 29 percent with interferon beta-1a alone. Combination therapy was associated with a lower annualized rate of relapse over a two-year period than was interferon beta-1a alone (0.34 vs. 0.75, P