BACKGROUND: In patients with relapsing-remitting multiple sclerosis (RRMS), subcutaneous (sc) interferon (IFN)ß-1a and IFNß-1b have been shown to reduce relapse rates. A formulation of IFNß-1a has been produced without fetal bovine serum and without human serum albumin as an excipient (not currently approved for use in the US). The objectives of this study were to evaluate tolerability, injection-site redness, subject-reported satisfaction with therapy, and clinical safety and efficacy of the serum-free formulation of IFNß-1a versus IFNß-1b in IFNß-treatment-naïve patients with RRMS. The objectives of the extension phase were to evaluate long-term safety and tolerability of IFNß-1a. METHODS: This randomized, parallel-group, open-label study was conducted at 27 clinical sites in the US. Eligible patients aged 18-60 years were randomized to receive either IFNß-1a, titrated to 44 µg sc three times weekly (tiw) (n = 65), or IFNß-1b, titrated to 250 µg sc every other day (n = 64) over 12 weeks. Following this, all patients received IFNß-1a 44 µg tiw for 82-112 weeks. Primary endpoint was mean change in patient-reported pain, as assessed by visual analog scale (VAS) diary pain score (from 0 mm [no pain] to 100 mm [worst possible pain]) at the injection site, from pre-injection to 30 min post-injection over the first 21 full-dose injections. Secondary assessments included proportion of patients pain-free as recorded by VAS diary and the Short-Form McGill Pain questionnaire VAS. RESULTS: A total of 129 patients were included in the intent-to-treat analysis. Mean (standard deviation) change in VAS diary pain score was not significantly different between groups, although numerically lower with IFNß-1a versus IFNß-1b from pre-injection to immediately post-injection (1.46 [2.93] vs. 4.63 [10.57] mm), 10 min post-injection (0.70 [1.89] vs. 1.89 [5.75] mm), and 30 min post-injection (0.67 [2.32] vs. 1.14 [4.94] mm). Proportion of patients pain-free at all time periods post-injection was also not significantly different between groups. Adverse events were consistent with the known safety profiles of these treatments. CONCLUSIONS: In IFNß-treatment-naïve patients with RRMS, both the serum-free formulation of IFNß-1a and IFNß-1b treatments were generally accompanied by low-level injection-site pain and were well tolerated. TRIAL REGISTRATION: ClinicalTrials.gov NCT00428584.