Combining different therapies may improve disease control in patients with relapsing-remitting multiple sclerosis (RRMS). This study assessed the efficacy and safety of minocycline added to subcutaneous (sc) interferon (IFN) β-1a therapy. This was a double-blind, randomized, placebo-controlled multicentre study. Within 3 months (±1 month) of starting sc IFN β-1a 44 μg three times weekly, patients with RRMS were randomized to minocycline 100 mg twice daily or placebo, added to sc IFN β-1a, for 96 weeks. The primary efficacy endpoint was the time to first qualifying relapse. Secondary efficacy endpoints were the annualized relapse rate for qualifying relapses, the number of new/enlarging T2-weighted lesions and change in brain volume [magnetic resonance imaging (MRI) was performed only in a few selected centres]. In addition, a number of tertiary efficacy endpoints were assessed. One hundred and forty-nine patients received minocycline and 155 received placebo; MRI data were available for 23 and 27 patients, respectively. The time to first qualifying relapse did not differ significantly for minocycline versus placebo (hazard ratio 0.85; 95% confidence interval 0.53, 1.35; log-rank = 0.50; P = 0.48). There were no statistically significant differences between the two groups on other efficacy endpoints, although some numerical trends in favour of minocycline were observed. No unexpected adverse events were reported, but more patients discontinued because of adverse events with minocycline versus placebo. Minocycline showed no statistically significant beneficial effect when added to sc IFN β-1a therapy.