Research paper
Analysis of neutralizing antibodies to therapeutic interferon-beta in multiple sclerosis patients: A comparison of three methods in a large Australasian cohort

https://doi.org/10.1016/j.jim.2005.11.011Get rights and content

Abstract

Persistent high-titre neutralizing antibodies (NAB) to therapeutic interferon-β (IFNβ) in multiple sclerosis patients reduce therapeutic efficacy. Difficulties in standardization of cell-based bioactivity assays have hindered interlaboratory comparison of NAB titres and the determination of a clinically relevant definition of seropositivity. We determined NAB status in Australasian multiple sclerosis patients receiving IFNβ using both the antiviral cytopathic effect (CPE) assay (n = 227) and the more specific ELISA for the type I interferon-inducible MxA protein (n = 350). While the log10 titres determined in the two assays were highly correlated (p < 0.0001; r = 0.967) with similar distributions, the MxA assay was more sensitive, detecting lower concentrations of NAB than the CPE assay. The range of titres determined in the CPE assay was 10 to > 7290; and 9 to 53,700 in the MxA assay, with ranked titre distribution highlighting the arbitrary nature of currently accepted definitions of NAB seropositivity. Bioactivity of injected IFNβ  was significantly reduced in NAB-positive patients (p = 0.006; NAB MxA titres = 184 to 5340) compared to NAB-negative patients as assessed ex vivo using real-time RT-PCR analysis of MxA gene induction. The range of MxA mRNA levels in healthy controls was remarkably consistent with previously published results, regardless of the assay standardization method [Gilli, F., Sala, A., Marnetto, F., Lindberg, R.L., Leppert, D. and Bertolotto, A. (2003) Comparison of IFNbeta bioavailability evaluations by MxA mRNA using two independent quantification methods. Abstract, ECTRIMS Meeting, Milan, Italy; Pachner, A., Narayan, K., Price, N., Hurd, M. and Dail, D. (2003a) MxA Gene Expression Analysis as an Interferon-beta Bioactivity Measurement in Patients with Multiple Sclerosis and the Identification of Antibody-Mediated Decreased Bioactivity. Mol. Diagn. 7, 17–25]. Assessment of IFNβ response ex vivo accounts for both circulating factors and the cellular response to IFNβ, and the data support the development of the MxA gene induction assay for the routine screening of patients receiving IFNβ.

Introduction

Interferon-beta (IFNβ) is a first-line treatment in relapsing–remitting multiple sclerosis (MS), significantly reducing relapse rates, lesion load and disability (PRISMS Study Group, 2001, Panitch et al., 2002, Frank et al., 2004). Neutralizing antibodies to IFNβ (NAB) may develop within months of commencing therapy; they are a subset of anti-IFNβ antibodies that prevent binding of IFNβ to its receptor and the subsequent biological effects (Deisenhammer et al., 1999, Bertolotto et al., 2003, Pachner et al., 2003b, Gilli et al., 2004) resulting in a loss of clinical efficacy of IFNβ (PRISMS Study Group, 2001, Malucchi et al., 2004, Petkau et al., 2004).

NAB are measured by quantifying the inhibition of IFNβ bioactivity by patient sera in responsive cell lines such as the lung carcinoma-derived epithelial line A549 or the HeLa-derived Wistar Institute Susan Hayflick (WISH) cells. The cytopathic effect (CPE) assay for NAB recommended by the World Health Organization (WHO) measures inhibition of the antiviral activity of IFNβ by patient serum (World Health Organisation, 1985). Antiviral assays require the use of a potentially biohazardous reagent and are subject to high inter-assay variability (Nestaas et al., 1996). Due to the indirect nature of the CPE assay, controls for viral and antiviral activity of the sample must be performed in conjunction with the test for IFNβ-neutralizing activity.

The MxA assay for NAB measures serum inhibition of the type I interferon-specific induction of the MxA protein by IFNβ (Files et al., 1998, Pungor et al., 1998, Pawlowski et al., 2003). This is a nested procedure in which MxA induction in cells cultured in the presence of IFNβ and patient serum is quantified by an enzyme-linked immunosorbent assay (ELISA) of cell lysates. The MxA assay avoids the variability and safety concerns associated with antiviral assays. The assay does not require a control for endogenous viral activity, and is not confounded by antiviral factors (other than type I interferon) in the sera. However, the variability inherent in cell-based assays remains (Deisenhammer et al., 2004, Vartanian et al., 2004).

A third approach to determining the bioactivity of IFNβ is to measure the induction of IFNβ-responsive genes in the patient's cells following IFNβ injection. MxA mRNA and protein induction in peripheral blood leukocytes by IFNβ is significantly impaired or abolished in NAB-positive individuals (Deisenhammer et al., 1999, Bertolotto et al., 2003, Pachner et al., 2003a). All relapsing–remitting MS patients in a recent study showed MxA induction at the initiation of IFNβ therapy, suggesting that lack of a cellular response to IFNβ occurs as a result of continued therapy rather than as an inherent non-responsiveness to IFNβ (Gilli et al., 2005). Regular monitoring of MxA mRNA levels may prove a useful biomarker of the ability to respond to IFNβ in vivo.

In this study, we have compared the CPE and MxA in vitro assays and the MxA mRNA ex vivo assay for analysis of NAB and their effect on IFNβ bioactivity in multiple sclerosis patients in Australia and New Zealand (Australasia).

Section snippets

Samples and reagents

 For NAB in vitro assays, serum samples were obtained from MS patients at least 24 h after the last IFNβ injection. The number of individuals tested by CPE and MxA assays was 227 and 350 respectively. Control sera from 20 volunteer donors were obtained from the Australian Red Cross Blood Service. Whole blood for MxA mRNA IFNβ in vivo bioactivity testing was obtained from 13 MS patients 12–24 h after the last IFNβ injection, and from 20 IFNβ-naïve control volunteers. Reference human anti-human

CPE assay results

Of 227 individuals tested for NAB by the CPE assay, 35 (15%) were NAB-positive with a range of titres from 20 to > 7290. The log10 titre of the NIH antiserum was 2.92 ± 0.11 against IFNβ-1a (n = 7 determinations) which is within the stated range for titre against human fibroblast IFNβ (3.24 ± 0.34; Research reference reagent note No. 45, NIAID, NIH, Maryland, USA).

Intra- and inter-assay variation was monitored using a NAB-positive rabbit antiserum, included in every assay. Only assays with an

Discussion

It is now widely accepted that NAB reduce the in vivo bioactivity and therapeutic effect of IFNβ. While the CPE assay is the WHO standard for NAB measurement, the MxA assay is now used by some investigators (Vartanian et al., 2004). We compared these in vitro techniques for the quantification of NAB in large cohorts of patients receiving IFNβ therapy. While results from the two assays were highly correlated and generated similar ranked titre distributions, the MxA assay showed higher sensitivity

Acknowledgements

We would like to thank Biogen-Idec for providing IFNβ, rabbit polyclonal anti-IFNβ antibodies and monoclonal anti-MxA antibodies, and Dr. Vijay Jethwa, Dr. Susan Goelz and Miki Pawlowksi (Biogen-Idec, USA) for their invaluable technical assistance. This project was supported by an education grant from Biogen-Idec.

References (33)

  • J.A. Frank et al.

    Interferon-beta-1b slows progression of atrophy in RRMS: three-year follow-up in NAb− and NAb+ patients

    Neurology

    (2004)
  • F. Gilli et al.

    Comparison of IFNbeta bioavailability evaluations by MxA mRNA using two independent quantification methods

  • F. Gilli et al.

    Neutralizing antibodies against IFN-beta in multiple sclerosis: antagonization of IFN-beta mediated suppression of MMPs

    Brain

    (2004)
  • C. Gneiss et al.

    Interferon-beta: the neutralizing antibody (NAb) titre predicts reversion to NAb negativity

    Mult. Scler.

    (2004)
  • S.E. Grossberg et al.

    The neutralization of interferons by antibody. II. Neutralizing antibody unitage and its relationship to bioassay sensitivity: the tenfold reduction unit

    J. Interferon Cytokine Res.

    (2001)
  • A. Kracke et al.

    Mx proteins in blood leukocytes for monitoring interferon beta-1b therapy in patients with MS

    Neurology

    (2000)
  • Cited by (0)

    View full text