Can we get rid of neutralizing antibodies against interferon‐β?

M. Buttmann,K. Toyka
DOI: https://doi.org/10.1111/j.1468-1331.2008.02333.x
2009-01-01
European Journal of Neurology
Abstract:Recombinant protein therapeutics such as interferon (IFN)-b may induce a humoral immune response just as any other protein seen by the immune system as nonself. A subgroup of these antibodies not only binds to IFN-b (BAbs) but may interfere with its biological actions by blocking functionally relevant epitopes – and these are called neutralizing antibodies (NAbs) [1–3]. IFN-b elicits its biological effects by binding to a single known heterodimeric cell surface receptor, the type I IFN receptor [4]. It seems obvious that blockade of this interaction in principle may inhibit the immunomodulatory effects of therapeutically applied recombinant IFN-b in patients with multiple sclerosis (MS). In several independent controlled studies, it was shown that NAbs, particularly if persistently present at high titres (>100–200 NU/ml), are associated with an increased relapse rate and MRI disease activity while evidence for a positive correlation with EDSS progression still remains weak or absent [5–9]. If one were to prove that the modest therapeutic effects of IFN-b on disease progression are abrogated by the presence of NAbs which occur only in a subgroup of patients, this would require randomized clinical trials designed and powered to show this kind of effect. Despite an ongoing discussion on how to standardize NAb determination and in what extent NAb determination should be implemented into clinical routine, there is a widely held consensus between experts that persistently high NAb titres most likely inhibit the therapeutic effect of IFN-b in MS patients [10–13]. Therefore, it seems appropriate to investigate ways to reduce the immune response to IFN-b and at least partially restore its bioactivity in this subgroup of patients, which would then allow us to continue with IFN-b treatment. In this issue of the European Journal of Neurology, the Danish Multiple Sclerosis Study Group reports on an attempt to restore IFN-b bioactivity in 73 MS patients who were persistently NAb-positive and did not show in vivo bioactivity of IFN-b. This group was split into half and 38 patients who were stopped of IFN-b treatment received 500 mg methylprednisolone (MP) on three consecutive days orally monthly over 6 months. The remainder of 35 patients formed the control group, who were switched from IFN-b to glatiramer acetate or stopped to take immunomodulatory treatments and did not receive glucocorticosteroids. This approach of intermittent high-dose steroid pulse therapy was based on a previous report suggesting a >50% reduced prevalence of NAbs during the first 15 months of IFN-b1b treatment when 1 g of intravenous MP per month was added to IFN-b1b from the start, i.e. prophylactically [14]. It is of note that in this report the rate of patients developing high NAb titres (>1:100) was not significantly influenced by MP treatment despite a favourable trend, although any such treatment would be more likely to down-regulate a primary immune response than an ongoing antibody response. In view of this, it is not quite unexpected that the investigated Danish patients with high-titer NAbs did not show a significant effect of MP treatment on NAb levels either; only the in vivo myxovirus resistance protein A (MxA) response to IFN-b was positively influenced by MP treatment in a small proportion of patients (21% vs. 11% regaining an in vivo response, not reaching significance). MxA is a type I IFNinducible protein often used to measure in vivo bioactivity of IFN-b although no strong evidence was provided to date that MxA induction correlates with the clinical treatment response to IFN-b [15]. The authors were aware that methodological restrictions such as limited group size, non-randomization and open-label treatment may have affected the outcome of this trial. Are there more potent strategies to effectively restore IFN-b bioactivity? Firstly, switching to another IFN-b formulation is not a promising approach as NAbs show considerable cross reactivity between commercially available IFN-b formulations [16,17]. Secondly, strategies to remove antibodies from the circulation by plasma exchange or to neutralize them by intravenous immunoglobulins containing anti-idiotypic antibodies are invasive and expensive. More importantly, they may have only a transient effect as plasma cells and memory B cells are not removed by these treatments and a re-challenge with recombinant IFN-b would again induce NAb generation. Even endogenously produced IFN-b may enhance NAb formation through a mechanism known as epitope spreading [18]. A more promising third strategy may be immunosuppressive co-treatment along with IFN-b therapy. Published clinical trials on combination therapy of IFN-b with immunosuppressive drugs such as methotrexate, azathioprine or cyclophosphamide were not specifically designed to study effects on NAb levels. Therefore, the number of NAbpositive patients in these trials was too low to allow even preliminary conclusions [19–21]. Theoretically, one of the most potent currently available approaches may be B cell depletion by the anti-CD20 monoclonal antibody rituximab which is currently under evaluation for MS therapy [22,23]. A recent trial investigated 375 mg/m rituximab per week over 4 weeks as an addon therapy in 13 patients already receiving treatment
What problem does this paper attempt to address?