An immune response to recombinant human protein therapeutics, including type I interferons (IFNs), has the potential to have a serious negative impact on safety and efficacy. Monitoring of patients for neutralizing antibodies (NAbs) often is advisable. In the case of IFN-beta therapy for multiple sclerosis (MS), we obtained reproducible quantitative titers of NAbs using an improved and well-characterized assay based on a 10-fold reduction of a challenge dose of IFN-beta. However, the observed titer was significantly affected by the preparation of IFN-beta used as the assay challenge. NAb titers obtained using IFN-beta1b averaged 3-5-fold lower than titers of the same sample assayed using either IFN-beta1a or human fibroblast-derived IFN-beta. This was the case whether neutralizing serum was obtained from patients on therapy with IFN-beta1a or IFN-beta1b. The reason for this apparent titer difference is not fully understood but appears to be related to protein folding or other structural properties that differentiate the IFN-beta1b both from commercial IFN-beta1a preparations and from human fibroblast-derived IFN-beta.