Journal of Interferon & Cytokine Research The Official Publication of the International Society for Interferon and Cytokine Research

Transcription

1 Journal of Interferon & Cytokine Research The Official Publication of the International Society for Interferon and Cytokine Research Editors-in-Chief Ganes C. Sen Department of Molecular Genetics NE20 Cleveland Clinic Foundation 9500 Euclid Avenue Cleveland, OH Tel.: (216) Fax: (216) Editorial Board Samuel Baron Galveston, TX Filippo Belardelli Rome, Italy Cornelia Bergmann Cleveland, OH Jay H. Hoofnagle Bethesda, MD David A. Hume Queensland, Australia Wolfgang Jelkmann Lübeck, Germany Ilkka Julkunen Helsinki, Finland Anthony Meager Hertfordshire, UK Karen L. Mossman Hamilton, Ontario, Canada Hisayuki Nomiyama Kumamoto, Japan Kikuo Onozaki Nagoya, Japan Thomas A. Hamilton Department of Immunology NE40 Cleveland Clinic Foundation 9500 Euclid Avenue Cleveland, OH Tel.: (216) Fax: (216) Senior Consulting Editor Philip I. Marcus (Storrs, CT) Associate Editors Ernest C. Borden (Cleveland, OH) Ferdinando Dianzani (Rome, Italy) James Finke (Cleveland, OH) Michael Gale Jr. (Seattle, WA) Lionel B. Ivashkiv (New York, NY) Dhan Kalvakolanu (Baltimore, MD) Wendy C. Brown Pullman, WA Michele Caraglia Naples, Italy Divaker Choubey Cincinnati, OH Suhayl Dhib-Jalbut New Brunswick, NJ Adrian M. Di Bisceglie St. Louis, MO Raymond P. Donnelly Bethesda, MD Mariano Esteban Madrid, Spain D. Mark Estes Galveston, TX Eleanor N. Fish Toronto, Canada David R. Fitzpatrick Queensland, Australia John M. Kirkwood Pittsburgh, PA Georg Kochs Freiburg, Germany Sergei V. Kotenko Newark, NJ Susan E. Krown New York, NY Santo Landolfo Torino, Italy Thomas E. Lane Irvine, CA Andrew Larner Richmond, VA Xiaoxia Li Cleveland, OH Jean Lindenmann (Honorary) Zurich, Switzerland Joost J. Oppenheim Frederick, MD Sidney Pestka Piscataway, NJ Lawrence M. Pfeffer Memphis, TN Richard Pine New York, NY Stephen Polyak Seattle, WA Stefan Rose-John Kiel, Germany Menachem Rubinstein Rehovot, Israel Charles E. Samuel Santa Barbara, CA Robert H. Silverman Cleveland, OH John E. Sims Seattle, WA Nicholas Lukacs (Ann Arbor, MI) Keiko Ozato (Bethesda, MD) W. Robert Fleischmann Minneapolis, MN Daniel J. Lindner Cleveland, OH Peter Staeheli Freiburg, Germany Leonidas Platanias (Chicago, IL) Gerald Sonnenfeld (Clemson, SC) Michael G. Tovey (Villejuif, France) Graham R. Foster London, UK Takashi Fujita Kyoto, Japan Sergio A. Lira New York, NY Xiaojing Ma New York, NY Deborah Vestal Toledo, OH Jan Vilcek New York, NY Howard A. Young (Frederick, MD) Otto Haller Freiburg, Germany Fabienne Mackay Victoria, Australia Carl F. Ware La Jolla, CA Albert Zlotnik (Irvine, CA) Kathryn C. Zoon (Bethesda, MD) Rune Hartmann Århus, Denmark Thomas R. Malek Miami, FL Bryan R.G. Williams Victoria, Australia

2 Interdisciplinary Coverage of the Unique Aspects and Challenges of Adolescent and Young Adult (AYA) Oncology About the Journal Journal of Adolescent and Young Adult Oncology (JAYAO) is a bimonthly peer-reviewed journal covering the unique aspects and challenges of adolescent and young adult (AYA) oncology. Editor-in-Chief Leonard Sender, MD Frequency Quarterly ISSN Online ISSN X JAYAO Includes Original articles Reviews Perspectives Clinical protocols Roundtable discussions Adolescents and young adults ages are a distinct patient population within oncology. This novel journal focuses on the unique biological, clinical, psychosocial, and survivorship issues in this age group. JAYAO is dedicated to improving adolescent and young adult cancer care management and outcomes through the promotion of interdisciplinary research, education, communication, and collaboration between health care professionals. Key Benefits Editor-in-Chief Leonard Sender, MD, a leader in the AYA field, founded a number of nonprofits in conjunction with key groups including: i2y I m Too Young for This!: support group for AYA patients SeventyK.org: an AYA Bill of Rights Explores long-term health consequences of chemotherapy and other radioimmunotherapies Enables hospitals and clinics to develop improved AYA-friendly facilities and services You may also be interested in these related journals: J Interferon and Cytokine Research Cancer Biotherapy Hybridoma The Official Journal of The Society for Adolescent and Young Adult Oncology (SAYAO)

3 Journal of Interferon & Cytokine Research VOLUME 30 NUMBER 10 OCTOBER 2010 IMMUNOTHERAPY OF MULTIPLE SCLEROSIS Special Issue Editor: Michael G. Tovey OVERVIEW Immunotherapy of Multiple Sclerosis 713 M.G. Tovey REVIEWS Current and Future Role of Interferon Beta in the Therapy of Multiple Sclerosis 715 R.A. Farrell and G. Giovannoni Single-Nucleotide Polymorphisms in Response to Interferon-Beta Therapy 727 in Multiple Sclerosis K. Vandenbroeck and M. Comabella Role of Differential Expression of Interferon Receptor Isoforms on the Response 733 of Multiple Sclerosis Patients to Therapy with Interferon Beta F. Gilli Differential Gene Expression and Translational Approaches to Identify Biomarkers 743 of Interferon Beta Activity in Multiple Sclerosis E. Croze Regulatory Effects of Interferon-β on Osteopontin and Interleukin-17 Expression 751 in Multiple Sclerosis J. Hong and G.J. Hutton Critical Review: Assessment of Interferon-β Immunogenicity in Multiple Sclerosis 759 K. Bendtzen On the Role of Aggregates in the Immunogenicity of Recombinant Human Interferon Beta 767 in Patients with Multiple Sclerosis M.M.C. van Beers, W. Jiskoot, and H. Schellekens (Continued)

4 PEGylated Interferon Beta-1a: Meeting an Unmet Medical Need in the Treatment 777 of Relapsing Multiple Sclerosis D.P. Baker, R.B. Pepinsky, M. Brickelmaier, R.S. Gronke, X. Hu, K. Olivier, M. Lerner, L. Miller, M. Crossman, I. Nestorov, M. Subramanyam, S. Hitchman, G. Glick, S. Richman, S. Liu, Y. Zhu, M.A. Panzara, and G. Davar Natalizumab Therapy of Multiple Sclerosis 787 M. Hutchinson DISSERTATION SUMMARIES 791 Instructions for authors can be found on our website: Cover art: Role of IFNβ in control of Infl ammatory Th1 and Th17 cells in multiple sclerosis. IFNβ can modify peripheral immune responses by altering both the migration and differentiation of T cells. These effects may occur through direct action of IFNβ on Th1 and Th17 cells or indirectly through the action of IFNβ on Dendritic cells (DCs). Reprinted with permission from: Hong J, Hutton GJ Regulatory Effects of IFN-β on Osteopontin and IL-17 Expression in MS. J Interferon Cytokine Res 30(10):

5 JOURNAL OF INTERFERON & CYTOKINE RESEARCH Volume 30, Number 10, 2010 ª Mary Ann Liebert, Inc. DOI: /jir OVERVIEW Immunotherapy of Multiple Sclerosis Michael G. Tovey This special issue of the journal is devoted to a series of short critical reviews on the immunotherapy of multiple sclerosis (MS). Each author has strived to provide a focused analysis of a particular aspect of the field rather than an exhaustive review of the literature. The impetus for this volume arose from the interest elicited by the Workshop on the Immunotherapy of MS that took place on the occasion of the Tri-Society Conference on Cytokines in Lisbon Portugal in October The special issue also addresses some aspects of the immunotherapy of MS that were not covered during the Workshop due to time constraints. MS is a chronic autoimmune disease of the central nervous system (CNS) of unknown etiology characterized by a perivascular mononuclear inflammatory infiltrate, plaque-like demyelination, and axonal injury. MS also exhibits the characteristics of an autoimmune disease, including the presence of auto-reactive T-cells that are thought to target specific epitopes in the CNS. The current standard of care for the treatment of relapsing-remitting MS (RRMS), the most common form of the disease, is the use of one or the other disease-modifying therapies: interferon beta (IFNb) and glatiramer acetate (Copaxone Ò ). Human IFNb-1b (Betaseron Ò /Betaferon Ò ) and subsequently IFNb-1a (Avonex Ò and Rebif Ò ) were the first disease-modifying therapies licensed for the treatment of RMMS. Avonex and Rebif are both glycosylated forms of native human IFNb-1a produced in Chinese hamster ovary cells. Betaseron is a nonglycosylated protein produced in Escherichia coli that has a serine substitution for the unpaired cystine at position 17 of the native protein. Glatiramer acetate is a random polymer of the amino acids glutamic acid, lysine, alanine, and tyrosine that is thought to mimic myelin basic protein a putative autoantigen in MS. Various mechanisms have been invoked to explain the action of glatiramer acetate in MS, including generation of suppressor T-cells, effects on antigen-presenting cells, and expansion of regulatory T-cells. Rachel Farrell and Gavin Giovannoni describe the current and evolving role of IFNb in the therapy of RRMS relative to other treatment options currently available or in development. Treatment with IFNb shortly after diagnosis reduces relapse rate and brain lesions detectable by magnetic resonance imaging, and may slow progression toward disability. Approximately one-third of patients do not respond, however, to IFNb therapy, and approximately two-thirds of those who do respond, experience clinical relapses within 2 years of initiation of therapy. There is also considerable variability in disease progression and individual responses to therapy, and a delay between initiation of treatment and the detection of beneficial effects. It is important, therefore, to identify patients incapable of responding to IFNb therapy either before treatment or at least in the early stages of treatment, so that they can be offered alternative therapy early in the course of the disease. Thus, there is need to identify genetic variants or biomarkers that either separately or in combination can identify individuals capable of responding to IFNb therapy. Four of the articles in the special issue address different approaches currently employed to attain this important goal. Koen Vandenbroeck and Manuel Comabella describe the use of pharamacogenetics, and in particular identification of polymorphic variants associated with response to IFNb therapy in patients with RRMS, with the goal of identifying before the onset of therapy, patients most likely to benefit from IFNb treatment. The review focuses on both the study of candidate genes and whole-genome single-nucleotide polymorphism scans. IFNb exerts its biological action by binding to a highaffinity hetrodimetic cell surface receptor comprising 2 transmembrane polypepties, IFNAR1 and IFNAR2. Ligand binding results in activation of the Janus kinases, Jak1 and Tyk2, and phosphorylation and activaton of the latent cytoplasmic signal transducers and activators of transcription (STAT1) and STAT2, which form a transcription complex together with IFN regulatory factor-9. Translocation of this complex to the nucleus results in the activation of a specific set of IFN-sensitive genes that encode the effector molecules responsible for mediating the biological activities of type I IFNs. As IFNs exert their biological action by transcriptional regulation of a specific set of IFN-responsive genes, differences in clinical response would be expected to be associated with different patterns of IFN-induced gene expression. Ed Croze discusses the advantage of expression profiling of multiple IFN responsive genes grouped according to gene ontology categories, and how monitoring expression of a preselected group of IFN responsive genes can provide increased statistical power relative to a single biomarker. Selection of appropriate genes is based on the analysis of samples from normal individuals, untreated MS patients, as well as responders, and nonresponders to IFNb therapy. Croze suggests that many of the difficulties associated with defining responders or nonresponders, compensating for patient demographics, disease state, and clinical end points, and so on, Laboratory of Viral Oncology, Institut Andre Lwoff, Villejuif, France. 713

6 714 TOVEY can be overcome by the study of cohorts of patients from large clinical studies. Selection of genes may also be hypothesis driven, based on genes involved in processes thought to be involved the clinical response to IFNb treatment. Croze also outlines the advantages of a systems biology approach based on literature reports, genomic databases, and observations made directly in the disease setting and validated using accepted treatment outcomes. Variable and complex regulation of IFN receptor expression on target cells may also influence an individual s ability to respond to treatment with IFNb. Francesca Gilli describes the results of studies that show that high levels of functional transmembrane IFNAR2 expression before treatment is associated with a greater biological response to IFNb treatment and that IFNAR2 expression is decreased upon prolonged exposure to IFNb in patients who respond to therapy. In contrast, production of soluble IFNAR2 can sequester IFNb, thereby inhibiting IFNb activity. Despite a detailed understanding of the IFNb signaling pathway and the transcriptional regulation of IFN responsive genes, the biological effects and molecular mechanisms underlying the therapeutic effects of IFN therapy in MS remain poorly understood. The anti-inflammatory activity of IFNb, and in particular regulation of the expression of Th1 and Th2 cytokines and inhibition of the migration of inflammatory T-cells into the CNS are thought to contribute to the therapeutic action of IFNb in MS. Jian Hong and George Hutton discuss recent findings that show that IFNb also downregulates expression of the pro-inflammatory cytokine osteopontin and the differentiation of IL-17 secreting Th17, cells that are thought to play an important role in driving inflammation in MS. Patients who develop antibodies to IFNb and, in particular, those who develop persistent high levels of neutralizing antibodies appear to have a worse clinical outcome. Klaus Bendtzen reviews the importance of the use of appropriate assays to detect anti-ifnb antibodies to obtain data of clinical relevance that can guide the clinician in the choice of treatment options. Miranda van Beers, Huub Schellekens, and Wim Jiskoot describe studies designed to understand the factors that can lead to a break in immune tolerance and the production of antibodies to IFNb. Immunogenicity is determined by the interplay between product-related factors such as differences in the amino acid sequence or glycosylation pattern between the recombinant and native molecules, or the extent of aggregation, and non-product-related factors such as the route and frequency of administration, concomitant medication, disease state, and so on. The authors discuss the factors that can influence the formation of aggregates, such as differences in formulation, with reference to the 3 currently marketed recombinant human IFNb products: Avonex, Betaseron/ Betaferon, and Rebif. The authors also discuss the use of transgenic mice, immune tolerant for human IFNb, to study the role played by aggregates in the immunogenicity of these products. Such studies are critical for the development of lessimmunogenic and better-tolerated treatment options. Darren Baker describes the use of pegylation to improve the pharmacokinetics of native IFNb and thus provide a treatment option requiring a reduced frequency of administration. The relative advantages of different approaches for pegylation of IFNs are discussed, and the clinical development of a PEGylated form IFNb-1a, in which the alpha amino group of the N-terminal amino acid has been specifically targeted for modification, is described. The characteristics of PEGylated IFNb-1a, including biological activity, the spectrum and severity of side effects, and immunogenicity, are also outlined. The humanized monoclonal antibody natalizumab (Tysabri Ò ), which targets a4b1 integins, thereby blocking attachment of leukocytes to the cerebral endothelium and reducing inflammation at the blood brain barrier, was approved for the treatment of relapses in patients with RRMS in Michael Hutchinson discusses the use of natalizumab for the treatment RRMS including patients treated with IFNb with breakthrough disease. Although natalizumab is highly effective for the treatment of active RRMS in some patients, therapy has to be closely monitored due to the risk of developing progressive multifocal leukoencephalopathy in a minority of patients. Rachel Farrell and Gavin Giovannoni analyze the available data on the use of several monoclonal antibodies, including Alemtuzumab (anti-cd52), Rituximab (anti-cd20), and Daclizuman (anti-cd25), which show considerable promise for the treatment of MS, although with the risk of rare but serious adverse events. Although a number of oral therapies, including Fumaderm Ò, Cladribine, Fingolimod, Teriflunomide, and Laquinimod, also show promise for the treatment of MS, it is probable that current therapeutic options, including IFNb, will continue to provide benefit to patients with RRMS for some time to come either alone or in combination with emerging therapies. Improved knowledge of the mode of action of IFNb and other biopharaamaceuticals such as natalizumab will also contribute to improving therapy as will the development of improved formulations and PEGylated forms of IFNb. Author Disclosure Statement Author has no ethical or financial conflicts of interest to disclose. Address correspondence to: Prof. Michael G. Tovey Laboratory of Viral Oncology Institute André Lwoff Villejuif France tovey@vjf.cnrs.fr Received 15 July 2010/Accepted 15 July 2010

7 JOURNAL OF INTERFERON & CYTOKINE RESEARCH Volume 30, Number 10, 2010 ª Mary Ann Liebert, Inc. DOI: /jir REVIEWS Current and Future Role of Interferon Beta in the Therapy of Multiple Sclerosis Rachel A. Farrell, 1 and Gavin Giovannoni 2 Interferon beta was the first specific disease-modifying therapy licensed for multiple sclerosis (MS) and in its many forms remains the most commonly prescribed agent worldwide. It, however, has a modest effect in reducing relapse rates, magnetic resonance imaging activity, and disability, and many patients are unable to tolerate it because of the associated side effects or mode of administration. With the licensing of glatiramer acetate, natalizumab and mitoxantrone as disease-modifying therapies for MS alternative options are available to people with MS. Many exciting new therapies are also in the pipeline, namely, the monoclonal antibodies alemtuzumab, rituximab, and daclizumab and the promising oral agents BG00012, cladribine, fingolimod, laquinimod, and teriflunomide. In this article we review the immunopathology of MS and the proposed mechanisms of action of currently available and anticipated treatments. We also review the efficacy of each drug, use of combination therapy strategies, and the potential role of the interferon beta preparations in the future. Introduction Descriptions of people with symptoms suggestive of multiple sclerosis (MS) date back to the 14th century; however, it was Charcot (1868) who defined MS as the disease entity we know today. Considered a neurological curiosity in the mid-19th century, by the turn of the 20th century, MS was recognized as one of the most common causes of admission to neurology hospital wards. A century later, MS is the most common nontraumatic cause of neurological disability in young adults. In the 1950s, however, as treatments for other neurological diseases were developed, they were applied to MS, creating for the first time the notion that treatment and a potential cure were possible. Treatments used included anticoagulation (Putnam and others 1947; Cox and others 1949), adrenocorticotrophic hormone ( Johnson and others 1950), nitrogen mustards (Simarro-Puig and others 1951), electroconvulsive therapy (Savitsky and others 1951), pyridostigmine, and diet (fat soluble vitamins and vitamin B12) to name but a few. However, it was in 1968 the first randomized, double-blind, placebo-controlled treatment trial for MS was performed: investigating the use of adrenocorticotrophic hormone over placebo in subjects with MS (Rose and others 1968, 1969, 1970). During the 1970s treatment of MS was largely based on using nonspecific immunosuppressive drugs such as azathioprine and methotrexate and continued use of corticosteroids with suboptimal results. For treatments of MS today there are articles identified in PubMed when searched using the term Multiple Sclerosis and treatment. Immunopathology MS When considering effective treatment of a disease entity, understanding the etiology and pathology is essential. The etiology of MS remains unknown but it is considered to be provoked in genetically susceptible individuals by a complex interplay of genes and the environment. Whether this is related to a single or numerous sequential exposures is unknown and potential agents include infections (viruses), sunlight/vitamin D, and smoking. The hallmarks of MS pathology are sharply defined lesions in the central nervous system (CNS), which are characterized by inflammation, demyelination, relative axonal preservation, and gliosis (Hohlfeld and Wekerle 2004). These lesions (plaques) may develop in any part of the CNS, though they have a tendency to accumulate near the periventricular and outer surfaces of the brain and spinal cord and are centered on one or several medium sized vessels (Lassmann and Wekerle 2006). Whereas lesions in the white matter (WM) are pathognomonic of MS pathology, changes are also seen in the nonlesional WM, including astrocytic proliferation, perivascular inflammation, blood brain barrier (BBB) leakage, a certain degree of sclerosis of blood vessels, and occasional demyelination (Allen and McKeown 1979; Miller and others 2003; Albert and others 2007). 1 Institute of Neurology, University College London, London, United Kingdom. 2 Blizard Institute of Cell and Molecular Science, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London, United Kingdom. 715

8 716 FARRELL AND GIOVANNONI Apart from MS plaques in the WM, studies carried out as early as 1890 suggested that plaques can also be found in the gray matter (Brownell and Hughes 1962; Dawson 1916), including the cerebral and cerebellar cortex, the basal ganglia, and the spinal cord (Kutzelnigg and others 2005, 2007; Gilmore and others 2006). Despite the relative preservation of axons and neuronal cell bodies in MS lesions, damage to both does occur, and this damage probably determines the long-term clinical manifestations of the disease (Trapp and others 1998; Peterson and others 2001). Although demyelinated lesions are the most characteristic feature of MS pathology, changes also occur in nonlesional brain tissue, the so-called normal-appearing WM (Zeis and others 2008) and normal-appearing gray matter (Albert and others 2007). Current concepts of MS lesion formation are often based on findings derived from experimental autoimmune encephalomyelitis (EAE), an animal model that resembles certain features of MS (Mix and others 2008). Against this backdrop MS has been described as a disease that is primarily mediated by autoreactive T-cells (CD4þ), which target-specific epitopes in the CNS. It has been postulated that by crossing the BBB and coming into contact with target antigens, a cell-mediated inflammatory reaction is initiated. Putative autoantigens include myelin basic protein (MBP), proteolipid protein (PLP), myelin oligodendroctye glycoprotein (MOG), myelin-associated glycoprotein (MAG), and ab crystallin (van Sechel and others 1999; Pender and Greer 2007). Once activated, T-cells produce an array of proinflammatory cytokines, which stimulate other T-cells, B-cells, natural killer (NK) cells, macrophages, and microglia, which in turn augment and perpetuate the inflammatory process. These cytokines promote increased permeability of the BBB, alteration of adhesion molecule expression, production of antibodies, and recruitment of other cells of immune function into the CNS. The presence of macrophages and T-cells (both CD4þ and CD8þ) has been shown in the brain parenchyma of MS patients (Sospedra and Martin 2005), and clonotypic CD8þ T-cells have also been described in the cerebrospinal fluid (CSF) ( Jacobsen and others 2002). The environment in MS lesions is complex with roles for both Th1 and Th2 CD4þ cells, cytotoxic CD8þ cells, macrophages, NK cells, and microglia. Secretion of both pro- and anti-inflammatory molecules, proteases, nitric oxide derivatives, reactive oxygen species, cytokines interleukin-6 (IL-6), tumor necrosis factor alpha (TNF-a), IL-4, and IL-10 may occur in the same lesion and may be essential in CNS repair (Hohlfeld and others 2007). The discovery of Th17 cells (CD4þ subset) has further advanced understanding of T-cell regulation. These cells, stimulated by IL-23, secrete IL-17 along with IL-6 and TNF-a, and EAE evidence suggests that they play an important role in inflammation and lesion formation (Bowman and others 2006). While demyelination is the most obvious feature of the MS lesion, neurodegeneration has also been described the first report >140 years ago (Charcot 1868). Damage to axons may be mediated directly by cytotoxic T-cells, macrophages, antibodies, loss of trophic support by oligodendrocytes, and oxidative stress (Sospedra and Martin 2005; Sayre and others 2008). Axonal transection in MS lesions is well described in pathological studies, the frequency of which appears to be related to the degree of inflammation in lesions (Trapp and others 1998). Loss of trophic support for axons and exposure to toxic metabolites leads to degeneration of axons. Redistribution of Na þ channels, mitochondrial failure, and Ca 2þ - mediated toxicity may all contribute to axonal degeneration in MS (Waxman 2006a, 2006b). Remyelination, a reparative process mediated by oligodendrocytes, depends on numerous factors, including switching the balance of the inflammatory process in favor of production of immunomodulatory cytokines and growth factors such as IL-4, IL-10, brainderived neurotrophic factor (BDNF), and transforming growth factor-b (TGF-b). Factors such as those mentioned above play a role in remyelination, which may, at least in part, underpin remission of symptoms and clinical signs between relapses. Research involving stem cell transplantation suggested that there may be a need for a certain degree of inflammation necessary to facilitate remyelination (Foote and Blakemore 2005). Understanding the mechanisms underlying remyelination is key to developing treatments that will have an effect not only on the inflammatory but also on the neurodegenerative component of MS, and repair (Hemmer and others 2002; Franklin and Ffrench-Constant 2008; Aktas and others 2010). Treatment of MS Potential targets for treatment include immune dysfunction (T-cells/B-cells), permeability of BBB (adhesion molecules), components of the inflammatory cascade (cytokines), putative autoantigens, demyelination, axonal loss (neuroprotection Naþchannel blockade, and Ca 2þ -mediated toxicity), and remyelination and regenerative processes (growth factors). As the initial causative step remains unknown, agents with a broad spectrum of activity are most likely to be effective. There are currently 7 medications licensed for use in MS, including interferon beta (IFN-b)-1a (Avonex, Biogen-Idec; Rebif, Merck-Serono) and 1b (Betaseron/Betaferon, Bayer- Schering; Extavia, Novartis), glatiramer acetate (GA) (Copaxone, Teva), natalizumab (Tysabri, Biogen-Idec/Elan), and mitoxantrone (Novantrone, Wyeth/Serono). First-line disease-modifying treatments (ie, the IFN-bs and GA) have now been available in the United States and Europe for up to 17 years. These drugs are partially effective in reducing the number of relapses by about one-third, reducing the number of lesions on magnetic resonance imaging (MRI) by *70% and may delay disease progression. Current opinion favors starting treatment early in the course of the disease, as neurodegeneration (eg, brain atrophy) can be detected from the very first manifestations of the disease, and at least a proportion of these degenerative changes may be secondary to inflammation (Frischer and others 2009). Hence, starting treatment early could potentially reduce the occurrence of relapses and development of disability ( Jacobs and others 2000; Comi and others 2001a; Kappos and others 2007; Clerico and others 2008). The complexity of decision making with respect to treatment choices in MS, and the number of available treatments continue to increase. The effects of the currently used agents are modest, but they have a very good safety profile that seems to have stood the test of time (IFN-b and GA). Reducing relapse rate probably prevents the acquisition of permanent disability caused by incomplete recovery from relapses, but this does not necessarily have an effect on the insidious progression that occurs independently of attacks in the secondary progressive phase of the disease. IFN-b and GA have both been shown to reduce relapse rates and MRI measures of disease activity. Although they are well

9 INTERFERON BETA IN MULTIPLE SCLEROSIS THERAPY 717 tolerated, their effectiveness is only partial, and none has been shown to be effective in primary progressive MS (PPMS) (Leary and others 2003; Montalban 2004; Wolinsky 2004). To define the role of IFN-b in treating patients with MS, it is important to be aware of the alternative therapies for MS, their efficacy, and safety profiles. Two drugs that are licensed for use in MS that are more efficacious than the first-line therapies are natalizumab and mitoxantrone, but their use is limited by unfavorable adverse effect profiles. Current and future use of IFN-b in people with MS will largely depend on the availability of alternative agents that are effective and have improved tolerability and favorable side effect profiles. We will review the available agents and those that are the most likely drugs of the future to discuss the role of the IFNs in future treatment strategies of MS. IFN-b in MS Type I IFNs were originally developed as therapeutic agents because of their anti-viral activity (Borden and others 2007). Although this may partially explain their effectiveness in MS, they may have numerous other immunomodulatory activities, including altering the Th1/Th2 balance (Hussien and others 2001), antagonizing proinflammatory cytokines (IFN-g, IL-12, and TNF-a), downregulating major histocompatibility complex class II expression, and affecting antigen presentation (Yong and others 1998), antiproliferative effects on T-cell expansion, differentiation, and increased T-cell apoptosis (Sharief and others 2001; Yong 2002). There is also evidence that type I IFNs inhibit transmigration of immune cells across the BBB (Leppert and others 1996). Two types of recombinant IFN-b products have been synthesized: IFN-b-1a is produced in Chinese hamster ovary cells and is genetically identical to the human form of IFN-b; in contrast, IFN-b-1b is produced in Escherichia coli, contains one amino-acid substitution that differs from the human form, and is nonglycosylated. In 1993 the first trial investigating the use of IFN-b in MS was performed using IFN-b- 1b, since then several separate trials have been conducted for each agent and more recently several head-to-head trials have been published. On average, all IFN-b products have been shown to reduce the annualized relapse rate by approximately one-third (MS Study Group 1993; Jacobs and others 1996; PRISMS 1998). MRI indices are also favorable with a 50% 70% reduction in disease activity using conventional MRI markers of disease activity (Gd-enhancing T1 lesions and new or enlarging T2 lesions). Efficacy regarding disability measures has been variable, with some trials showing an effect, others being inconclusive, and others that did not include disability as an outcome measure ( Jacobs and others 1996; PRISMS 1998; PRISMS Study Group and the University of British Columbia MS/MRI Analysis Group 2001). However, many of these pivotal studies were underpowered to show true effect on disability. Follow-up studies evaluating early use of IFN-b in subjects with a clinically isolated syndrome have shown delayed time to first relapse and conversion to clinically definite MS, and may also have an impact on disability progression ( Jacobs and others 2000; O Connor 2003; Kappos and others 2007; Koch and others 2007). Current prescribing guidelines differ between countries, but in the majority IFN-b is licensed for treatment of relapsing-remitting MS (RRMS) and in subjects with a clinically isolated syndrome who have evidence on MRI to suggest that they are at high risk of conversion to MS. The IFN-bs have generally been well tolerated, and side effects included flu-like symptoms, injection site reactions, myalgia, and in a few cases depression. In general, depression has not been a major side effect in latter trials of IFN-b. Various strategies have been proposed to manage these side effects (Munschauer and Kinkel 1997). Further adverse effects include abnormal liver function tests, anemia, leukopenia, and thrombocytopenia, all of which can be monitored with regular laboratory tests. One of the most significant problems with IFN-b treatment is the production of drug-specific antibodies. Up to 45% (mean *25%) of patients develop neutralizing antibodies (NAbs) to IFN-b products. NAbs may be detected as early as 3 months after commencement of therapy, but usually appear between 6 and 18 months. Binding antibodies occur earlier. Of the 2 IFN-b-1a products the subcutaneous IFN-b-1a (Rebif Ò ) is more immunogenic than intramuscular IFN-b-1a (Avonex Ò ) and antibodies are more frequently induced by IFN-b-1b than 1a (Cook and others 2001; Panitch and others 2002; Bertolotto and others 2003; Malucchi and others 2004). The development of NAbs is a significant factor contributing to treatment failure, and the reduction in relapse rate in subjects who remain NAb negative may be as high as 50% (Interferon Beta Study Group 1996). It has been shown in numerous trials that patients who become antibody positive have higher relapse rates, lesion activity on MRI, and also higher rate of disease progression (Francis and others 2005; Kappos and others 2005). There has been much controversy with respect to the significance of these antibodies in patients with MS treated with IFN-b and how to manage them (Hartung and Munschauer 2004; Farrell and Giovannoni 2007). Currently, separate European and North American guidelines govern the incorporation of NAb testing into clinical practice. The AAN reported in 2007 their recommendations focusing on the effect of NAbs on clinical and radiologic outcomes (Goodin and others 2007). The committee concurred that there is probably a reduction in efficacy of treatment because of NAbs and there is likely to be greater antibody production in response to IFN-b-1b than to IFN-b-1a, and that intramuscular (IM) IFN-b-1a is clearly less immunogenic than other IFN therapies. Despite the consistent finding of NAb levels >1:200 being associated with a reduction of efficacy, the committee was unable to make definite recommendations for changing therapy. In contrast, the EFNS recommended that patients treated with IFN-b are tested for the presence of NAbs at 12 and 24 months of therapy (level A recommendation). In those whom remain NAb negative at months further testing is not routinely required (level B recommendation). They stated that class I evidence shows that the presence of NAbs significantly reduces the effect of IFN-b on relapse rate and active lesions and burden of disease seen with MRI. In patients who are NAbþve measurements should be repeated at intervals of 3 6 months and therapeutic options should be reevaluated (level A recommendation). Therapy with IFN-b should be discontinued in patients with high titers of NAbs at repeated measurements with 3- to 6-month intervals (level A recommendation) (Sorensen and others 2005). New guidelines from a European consortium are in print currently, which will further support the recommendation that even in patients doing well clinically, NAb and/or MxA

10 718 FARRELL AND GIOVANNONI bioactivity assessments should be monitored and used in clinical management. Particularly, in cases of sustained high titer NAb positivity and/or lack of MxA bioactivity, a switch to a non-ifn-b therapy should be considered. Glatiramer Acetate GA (Copaxone Ò ) is a mixture of synthetic polypeptides that quantitatively resembles MBP, a putative autoantigen in MS. GA was found to suppress MBP-induced EAE (Teitelbaum and others 2004). Its mode of action is unclear but it has been shown to compete with major histocompatibility complex binding of MBP and limit activation of MBP reactive T-cells. A broad T-cell response is seen after GA injection, and it is therefore likely that GA also mimics antigens other than MBP. GA leads to a shift in the T-cell population toward a Th2 cytokine response profile (Miller and others 1998; Kim and others 2004). It has been proposed that these Th2 cells migrate through the BBB into the brain parenchyma, where they are activated. As a result, they produce immunomodulatory cytokines that counteract the proinflammatory Th1 response, an effect called bystander suppression. GA reactive T-cells have also been shown to secrete brain-derived neurotrophic factor (BDNF), which has anti-inflammatory and neuroprotective functions (Stadelmann and others 2002; Ziemssen and others 2002). In animal models of MS it has been shown that GA reactive T- cells cross into the CNS and produce these factors (Aharoni and others 2003). Most of the described effects have been demonstrated in in vitro studies, and a number of questions with regard to GA s mechanism of action remain as yet unanswered. In the pivotal trial 20 mg of GA subcutaneous (SC) daily was shown to reduce relapse rates by approximately one-third and also had a favorable effect on MRI outcomes, Gadolinium enhancing lesions, and measures of brain atrophy; however, a significant effect was not seen on T2 lesions ( Johnson and others 1995; Ge and others 2000). Subsequent studies showed significant reduction of T2 lesions (Comi and others 2001b) and reduction of T1 black holes (Filippi and others 2001). Regarding adverse events injection site reactions are common with GA and lipodystrophy has been described (Drago and others 1999). The most common side effect (*15%), however, is an idiosyncratic self-limiting reaction of chest tightness, flushing, dyspnea, and palpitations, which can be quite frightening for the individual but generally resolves quickly and spontaneously. Natalizumab Natalizumab (Tysabri Ò, Biogen-Idec/Elan) is a humanized monoclonal antibody (mab) that targets a4-integrin, acting as a selective adhesion molecule inhibitor. The glycoprotein a 4 b 1 integrin is also known as very late antigen 4 (VLA4). It is expressed on the surface of lymphocytes and monocytes and plays an important role in cell adhesion and trafficking across the blood brain and other endothelial barriers. VLA4 also acts as a regulator of activation of the immune system in areas of inflammation (Frenette and Wagner 1996a, 1996b). Natalizumab is given as a monthly infusion at a standard dose of 300 mg. It is not clear exactly how natalizumab exerts its clinical effect but is thought to be due to the significantly reduced migration of leukocytes into the CNS parenchyma. Two phase III trials of natalizumab have shown significant reduction in relapse rate and disease progression: the first trial (AFFIRM) involved treatment naive patients randomized to placebo (n ¼ 315) or 300 mg natalizumab (n ¼ 627) every 28 days for up to 28 months. The second trial (SEN- TINEL) involved patients who were established on IFN-b 1a 30 mg IM once weekly (Avonex) and had experienced one or more relapses in the previous year. Patients continued Avonex therapy and were randomized to either 300 mg natalizumab (n ¼ 589) or placebo (n ¼ 582) infusions every 28 days. In the AFFIRM study patients treated with natalizumab had a 66% reduction in relapse rate compared with placebo, and 96% of subjects receiving natalizumab had no new Gd-enhancing lesions as compared with 68% on placebo (Polman and others 2006; Miller and others 2007). A reduction in new and enlarging T2 lesions was also found. In the SENTINEL study patients on both Avonex and natalizumab had a 54% reduction in annualized relapse rate as compared to placebo. With regard to MRI outcomes 96% of the combined treatment group had no enhancing lesions as compared with 76% of the Avonex only group. Similarly, fewer patients developed new or enlarging T2 lesions, on combination therapy (Rudick and others 2006). However, at the end of February 2005 the use of natalizumab was temporarily suspended after 2 patients receiving natalizumab in combination with IFN-b-1a (Avonex) developed progressive multifocal leukoencephalopathy (PML), one of whom subsequently died. A third subject with Crohn s disease whose death had been attributed to CNS malignancy has also been confirmed as having PML. Since this time natalizumab was relaunched in June 2006 and several patients have subsequently developed PML (43 patients as per March 2010) and several have been naive to prior immunomodulatory treatments. Emerging evidence suggests that subjects treated for >2 years are at highest risk (Clifford and others 2010). Although many patients are now treated with natalizumab worldwide, vigilance is required with regard to its safety profile. Other side effects include anaphylactic reactions (*0.8%) and other hypersensitivity reactions, increased risk of infection, headache, depression, arthralgia, and rash. Approximately 10% of patients receiving natalizumab developed antibodies that were neutralizing in vitro, with only 6% persisting beyond 12 months of treatment (Calabresi and others 2007). Therapeutic effectiveness was reduced in those who remain persistently antibody positive, and relapse rates returned to that of those on placebo in both trials. Infusion reactions were also found to be more frequent and severe in those remaining antibody positive. The safety profile and immunogenicity of natalizumab as another biological agent to treat MS are important factors to consider in its use. Mitoxantrone Mitoxantrone (Novantrone Ò Serono) was originally designed as a chemotherapeutic agent and is most commonly used in treating breast cancer. It is an anthracenedione that inhibits topoisomerase-2, preventing the successful unwinding of DNA. It was shown to be effective in EAE (Lublin and others 1987) and in both active RRMS and secondary progressive MS (SPMS) (Edan and others 1997; Hartung and others 2002; Le Page and others 2008). It is currently licensed for use in patients with aggressive relapsing MS who have

11 INTERFERON BETA IN MULTIPLE SCLEROSIS THERAPY 719 failed first-line therapy. The mechanism of action of mitoxantrone in MS may be mediated on a number of levels. It is immunosuppressive, inhibiting the proliferation of T-cells, B- cells, and monocytes and reducing secretion of proinflammatory cytokines (Fidler and others 1986); however, it is nonspecific in its immunosuppressive action. It is typically administered as IV pulses of 12 mg/m 2 every 3 months for 2 years (Hartung protocol) or 20 mg monthly for 6 months (Edan protocol). It has several important side effects; most notably, it is cardiotoxic and warrants monitoring of cardiac function (Ghalie and others 2002a). In view of its cardiotoxicity, the maximum recommended lifetime dose is 140 mg/m 2. There have been several reports of therapyrelated acute myelogenous leukemia in patients with MS who have received mitoxantrone (Brassat and others 2002; Ghalie and others 2002b). Typically, therapy-related leukemia, with topoisomerease II inhibitors such as mitoxantrone, develops within 2 4 years after chemotherapy has been started, and the prognosis is generally poor among the breast cancer patients; those with MS appear to fare better. Initially, it was reported to occur in 0.07% patients receiving mitoxantrone but Serono have released data describing 2 cases in 802 MS patients (0.25%) (Edan 2005), and more recently follow-up studies have described the risk as 1% 2% (Le Page and others 2008). The postmarketing data in breast cancer have shown a higher risk in patients who have had a combination of chemotherapeutic agents. It has therefore been advised that strict adherence to blood count recommendations should be followed, including complete blood count, white cell differential, and platelets before each infusion. (Prescribing information NOVANTRONE, Merck-Serono). There is also a risk of premature ovarian failure or infertility, which may be permanent. Transient amenorrhea occurs in *12% of patients and persistent amenorrhea in *10% of patients. The risk of persistent amenorrhea is higher in woman older than 35 years (14%) and lower in women <35 years of age (6.5%) (Edan 2003). Monoclonal Antibodies Natalizumab was the first licensed mab for the treatment of MS, but several others are in the pipeline and have completed promising phase II trials. Alemtuzumab Campath 1H (Alemtuzumab Genzyme) is currently in phase III trials. It is a humanized IgG1 mab that is licensed for use in B-cell chronic lymphocytic leukemia and by binding to CD52 rapidly depletes any CD52-bearing cells (T-cells, B-cells, NK-cells, monocytes/macrophages, and some granulocytes), thus wiping out cells involved in autoimmunity and the inflammatory cascade. Indirect actions are thought to include stabilization of the BBB and eventual repopulation of the immune system with initially B-cells followed by T-cells enriched with CD25 high T-cells. It is hypothesized that this induces immune tolerance and thus the dramatic reduction in disease activity after treatment. Treatment is given as infusions over a 5-day period on an annual basis. Initial trials evaluated alemtuzumab in SPMS subjects and it did not show a beneficial effect on disease progression (Coles and others 2006). However, when used to treat subjects with early RRMS the difference was remarkable. Thereafter, a trial was conducted evaluating alemtuzumab in early RRMS (<3 years post onset) as compared to the best available treatment Rebif 44 mg three times weekly (ttw) (CAMS 223 study). Participants receiving alemtuzumab saw a dramatic reduction in relapses, improvement in expanded disability status scale (EDSS), and robust MRI data (Gdþve lesions, T2 lesion load, and atrophy measures) (Coles and others 2008). Regarding tolerability, alemtuzumab is administered as an intravenous immunoglobulin (IVI) over a 3- to 5-day period and frequent infusion reactions comprised of pyrexia, malaise, and rash have been described due to acute cytokine release. This is improved by use of pretreatment corticosteroids. However, 1 patient died due to undetected immune thrombocytopenic purpura (ITP) (5 others were identified), others developed Graves disease (*30%), and Goodpasture s syndrome has also been reported in 3 patients. Patients treated with alemtuzumab also experienced higher rates of infections. Two large phase III trials are underway evaluating alemtuzumab in treatment naive MS patients and in those who have failed first-line therapy. Rituximab Recent focus on the role of B-cells in MS supports the use of rituximab, a mab targeting CD20, as a potential therapeutic option. CD20 is only expressed on mature B-cells and not on plasma cells and rituximab causes a transient depletion of B-cells (*6 months). It also has indirect effects on macrophages and T-cell-mediated immune response and induces repopulation of B-cells from the bone marrow with naive B-cells. Clinical trials to date have demonstrated a rapid reduction in disease activity on MRI that was sustained in a phase I, 72-week open label study (Bar-Or and others 2008) and confirmed in a 48-week double-blinded phase II trial (Hauser and others 2008). Phase III trials are underway assessing follow-on anti-cd20 compounds. In PPMS rituximab showed a trend in reducing time to sustained progression in treated patients, but this failed to reach significance, and there was also a reduction in accumulation of T2 lesions suggestive of biological response (Hawker and others 2009). Regarding subgroup analysis, younger subjects <51 years or those with Gdþve lesions at baseline showed a reduced time to confirmed disease progression and these predictors had an additive effect; those >51 years with Gdþve lesions at baseline, treated with placebo had 3 times the risk of confirmed disease progression than those treated with rituximab (hazard ratio [HR] ¼ 0.33, P ¼ 0.009). In the accompanying editorial (Hartung and Aktas 2009) the difficulties with study design and outcome measures in PPMS are discussed highlighting the issue that outcomes requiring a sustained EDSS progression of in a 2-year period may lead to failure to identify actual treatment effects. Thus, as with natalizumab and alemtuzumab the evidence available to us seems to indicate that rituximab is efficacious in RRMS and early in the disease process. PML has also been a problem in rheumatology patients treated with rituximab, and cases have been reported in patients treated for lymphoma and rheumatoid arthritis also. Daclizumab Daclizumab has been widely used in prevention of transplant rejection and blocks CD25, which forms part of the IL-2 receptor (Tac epitope). In resting T-cells the level of CD25 is

12 720 FARRELL AND GIOVANNONI low but is significantly upregulated in activated T-cells (Granucci and others 2003). The scientific rationale behind the development of daclizumab was to reduce T-cell proliferation by blocking the formation of the IL-2 high affinity receptor. One of the proposed benefits of daclizumab is that rather than lead to a rapid depletion of lymphocytes in vivo, its effect is mediated via expansion of the regulatory CD56 bright NK cell pool that have a role in killing activated T-cells in inflammatory lesions (Bielekova and others 2006). It also inhibits survival of CD25 Foxp3þ T reg cells and this may play a role in some of the associated side events. Regarding efficacy, 3 phase II trials have been completed investigating the role of daclizumab in patients who have failed IFN-b treatment or as an adjunct to IFN-b. In both, daclizumab was successful in stabilizing active disease and was also effective in reducing Gd þ lesions on MRI. The subsequent CHOICE study compared IFN-b alone with IFN-b and daclizumab, and found that the addition of daclizumab significantly reduced the number of new or enhancing lesions on MRI (Wynn and others 2010). Extension studies to further explore both efficacy and safety profile of the drug are recruiting at present. What is apparent with all of the mab therapies is that they are highly effective in reducing the inflammatory component of MS; however, none have yet shown a beneficial effect when commenced in patients already in the progressive phase of MS. All carry the risk of serious (albeit rare) life-threatening adverse events and thus may not be attractive options to all patients. As we have seen in the case of mitoxantrone and natalizumab, time will reveal the true extent of the risk. Oral Agents Current first-line disease-modifying therapies are administered as SC or IM injections and cause frequent side effects, including injection site reactions, flu-like symptoms, and lipodystrophy in the case of GA. The mabs, although more effective, require regular infusions and thus admission to a day unit/hospital; they also carry a risk of life-threatening adverse events. The holy grail has been to produce oral agents that are better tolerated and equally efficacious as the injectibles. There are 5 candidates emerging as the most promising in this race: BG00012 (Fumarate), cladribine, fingolimod, laquinimod, and teriflunomide, and the first oral agent is likely to be licensed in 2010/2011. BG00012 BG00012 or dimethyl fumarate (Biogen Idec) is a commonly used agent in the treatment of psoriasis (Fumaderm Ò ) and has been reformulated for use in MS. One of the advantages of this drug is its long history and known safety profile. Fumarate activates the nuclear factor E2-related factor 2 transcriptional pathway (Nrf2), which controls phase-2 detoxifying enzyme gene expression, and plays an important role in the oxidative stress response and immune homoeostasis (Itoh and others 1997; Chen and others 2006). Activation of the Nrf2 pathway defends against oxidative-stressinduced neuronal death (Calabrese and others 2005; Li and others 2005), protects the BBB (Zhao and others 2007), and protects myelin integrity in the CNS (Hubbs and others 2007). Dimethyl fumarate induces enzymes expression in astroglial and microglial cells. It also inhibits expression of cytokines and adhesion molecules implicated in the inflammatory response in vitro. These effects suggest that BG00012 may have both neuroprotective and anti-inflammatory effects (Wierinckx and others 2005; Schilling and others 2006). Phase II studies have been completed and showed that 240 mg three times daily (tds) oral fumarate reduced Gd þ lesions by 69% as compared with placebo and also reduced new T2 and T1 hypointense lesions over a 24-week period. There was a trend that annualized relapse rate (ARR) reduced, but due to the short duration of the study, this did not reach significance (Kappos and others 2008). Two phase III trials are ongoing, one of which is a head-to-head study with GA. These are due to finish next year. Cladribine Cladribine (2-chlorodeoxyadenosine) is a synthetic purine nucleoside analog prodrug that is taken orally for 10 or 20 days per year. Cladribine accumulates and is incorporated into the DNA of lymphocytes as a result of a high ratio of deoxycytidine kinase to 5 0 nucleotidase activity and selectively induces apoptosis in lymphocytes (Carson and others 1983). With regard to the proposed mechanism of action in MS, it is proposed that by causing a sustained reduction in lymphocytes (CD4þ T-cells and CD8þ T-cells) (Beutler and others 1996) in both active and nonactive phase but only a transient reduction in other immune cells such as neutrophils and monocytes (Rice and others 2000), it is known to cross the BBB (Liliemark 1997) and has been shown to reduce levels of proinflammatory chemokines in the CSF (Bartosik- Psujek and others 2004). Recently, a phase III study was completed comparing the efficacy of 2 doses of cladribine (3.5 versus 5.25 mg/kg) and placebo and patients were followed for 96 weeks (Giovannoni and others 2010). Relapse rate was the primary endpoint and was reduced by 58% and 55%, respectively, in the treatment arms. The proportion remaining relapse free was higher in the treatment arms and there was reduction in progression as measured by EDSS (33% and 31%, respectively). Secondary outcomes included MRI parameters, Gdþ lesions, and T2 lesions with a combined 73% reduction in MRI activity. The drug was well tolerated causing lymphopenia frequently and severe neutropenia in 3 subjects with slightly higher rates of mild to moderate infection and 20 developing VZV infection (3 primary infections). There were 3 subjects who developed malignancy after treatment with cladribine; however, the malignancies were heterogynous and spread across organ systems. Until more safety data emerge, the link with cladribine treatment is not yet clear. Two further phase III trials are ongoing to investigate the efficacy of the dose of cladribine. Fingolimod Fingolimod (FTY720) is a structural analog of sphingosine- 1-phosphate that plays an important role in lymphocyte migration from lymph nodes into the periphery. Fingolimod also readily crosses the BBB and interacts with S1P receptors in the CNS (Brinkmann 2009). Fingolimod has also been shown to prevent demyelination and to promote remyelination in animal models ( Jung and others 2007). Fingolimod downregulates expression of inflammatory genes and vascular adhesion molecules, reducing matrix metalloproteinase gene (MMP-9), and increases tissue inhibitor of metallo-

13 INTERFERON BETA IN MULTIPLE SCLEROSIS THERAPY 721 proteinase, tissue inhibitor of metalloporteinases (TIMP-1), resulting in an environment that favors preservation of BBB integrity (Foster and others 2009). Late-stage rescue therapy with fingolimod reversed BBB leakiness and reduced demyelination, with clinical improvement in neurologic function. In a phase III study 2 doses (5 or 1.25 mg) of oral FTY720 demonstrated superiority over IFN-b-1a IM at 12 months with regard to relapse rate (52% and 38% reduction with FTY720 versus IFN-b-1a IM) (P < ). Both doses reduced MRI inflammatory activity compared with IFN-b-1a IM (Cohen and others 2010). Fingolimod was generally well tolerated with 87% of fingolimod-treated participants completing the study on treatment. Overall safety profile of the fingolimod 0.5 mg dose appeared better than 1.25 mg dose, including lower rates of serious infections and bradycardia. The FREE- DOMS study phase III compared high and low dose fingolimod with placebo and showed reduced relapse rate, disability progression, and MRI lesion load (Kappos and others 2010). Mild side effects occurred in >90% of subjects and serious events in *10%. In the fingolimod group bradycardia and atrio-ventricular conduction block was reported more frequently than in the placebo arm, and these cardiovascular effects appear to be dose dependent. Other adverse events included macular edema, infections, hypertension, a reduction in respiratory function (FEV1), and possibly an increase in the incidence of secondary malignancies. Laquinimod Laquinimod is an oral immunomodulatory agent derived from linomide and is better tolerated. Linomide was poorly tolerated in trials of MS with frequent side effects, including fatigue, malaise, flu-like symptoms, and several reports of serositis and neuropathy (Andersen and others 1996; Karussis and others 1996; Wolinsky and others 2000). Laquinimod has been shown to be effective in animal models (Brunmark and others 2002) and is thought to be effective in treated MS by inducing the release of transforming growth factor and shifts the immune response toward a Th2-type profile rather than suppressing the immune system. Reduced leukocyte infiltration in the CNS induced a deviation of MBP-specific cells from a Th1 to Th2/Th3 pattern (Yang and others 2004). Two phase I trials with laquinimod demonstrated that the drug was well tolerated by healthy volunteers and patients with MS. To date there have been 2 positive studies showing safety and efficacy in patients with MS. The first tested 2 different doses of oral laquinimod (0.1 and 0.3 mg/day) versus placebo in 180 with RRMS (Polman and others 2005). Mean cumulative number of active MRI lesions was the primary outcome measure treating subjects with 0.3 mg laquinimod or placebo and showed a 44% reduction. Clinical outcome parameters (relapse rate, disability) were not different between the groups. A further phase IIb study evaluated the effect of 2 doses of laquinimod (0.3 and 0.6 mg) compared with placebo, and primary outcome was MRImonitored disease activity over 36 weeks. Compared with placebo, 0.6 mg laquinimod per day showed a 40.4% reduction of the baseline adjusted mean cumulative number of Gdþ lesions per scan, whereas treatment with 0.3 mg per day showed no significant effects (Comi and others 2008). Laquinimod (0.6 mg) is currently being further evaluated in 2 phase III trials in a head-to-head study with IFN-b-1a IM (Avonex) and against placebo to evaluate effect on relapse rate, disability, MRI, and safety. Teriflunomide Teriflunomide is the active metabolite of leflunomide (Arava), which is an approved treatment for rheumatoid arthritis, and acts by reversibly inhibiting the mitochondrial enzyme dihydro-orotate dehydrogenase, which plays a crucial role in pyrimidine synthesis (Xu and others 1997). Teriflunomide also has an anti-inflammatory effect by interfering with T- and B-cell proliferation (Korn and others 2004). In animal models (EAE) Leflunomide has been found to suppress disease activity by inhibition of TNF-a and IL-2 (Korn and others 2001; Smolen and others 2004). Phase II results have recently been published in patients with RRMS and SPMS. Two teriflunomide doses (7 and 14 mg once-daily) were compared to placebo for a period of 36 weeks. The primary endpoint of the study was the number of active MS lesions and new lesions on MRI and was significantly reduced in both treatment arms. EDSS progression was delayed in the high-dose arm and a trend toward reduction in relapses was observed (O Connor and others 2006). There are several further phase II and III trials currently recruiting to further evaluate the efficacy of teriflunomide alone or as an add-on agent to IFN-b or GA. Regarding side effects, generally, the tablets were well tolerated and although all patients during the trial reported side effects, the difference between groups did not reach significance and were thus considered unrelated to the drug. There were similar numbers of serious events in the placebo and treatment groups and no deaths were reported. Based on reproductive studies in animals, female patients are advised to avoid pregnancy when taking teriflunomide as it is known to be teratogenic; data regarding fathering a child are unclear and caution is advised. Combination Strategies As illustrated, there are 7 licensed therapies and many more in the pipeline that treat MS by modulating the immune system in a variety of ways, primarily acting on T-cells, B- cells, gene expression, cytokine release, or immune cell proliferation. Effective treatments exist for RRMS; however, none have shown efficacy in SPMS or PPMS. Reduction of disability progression has only been shown within the RRMS cohorts, particularly in early treatment trials. It will emerge in due course whether these individuals will enter the SPMS phase or if the disease has truly been halted. Combination treatment strategies are widely used throughout medicine and are being explored in MS also. The rationale is to combine drugs that treat MS by targeting different aspects of the immunopathological cascade. Some drugs effectively target the inflammatory component of the disease and others may facilitate neuroprotection and repair. Agents that are being evaluated in combination with IFN-b include nonspecific immunosuppression, mycophenolate mofetil, tacrolimus, methotrexate, and high dose IV methylprednisolone. Potentially neuroprotective agents include statins (atorvastatin and simvastatin), doxycycline, lamotrigine, topiramate, riluzole, epigallocatechin- 3-gallate (active ingredient in green tea), minocycline, and mesenchymal stem cells. Several approaches may be employed with combination therapies: concomitant treatment (CombiRx trial), induction/maintenance (eg, mitoxantrone followed by

14 722 FARRELL AND GIOVANNONI GA) (Ramtahal and others 2006), or as add on treatment for example in the SENTINEL study when natalizumab was added on to IFN-b-1a (Avonex) (Rudick and others 2006). Combination Studies with Corticosteroids To date, there have been many pilot studies evaluating combination strategies, so we will discuss them broadly rather than individually. Three large combination studies have been published investigating the role of immunosuppression in combination with IFN-b and had negative results. The Avonex Combination Trial had 4 arms treated with additional methylprednisolone methotrexate for 12 months and used new or enlarging T2 lesions as the primary endpoint (Cohen and others 2009); other measures included the T2 lesion volume, Gd-enhancing lesions, and Multiple Sclerosis Functional Composite (MSFC), none of which reached significance. The Avonex Steroids Azathioprine trial compared 3 treatment arms evaluating the use of add on oral methylprednisolone with and without azathioprine in patients who had failed first-line treatments. Participants were followed for 3 years with a 1-year extension phase. An effect was found with a reduction in ARR in the triple therapy group compared to the dual-placebo arm (0.73 versus 1.05 P < 0.05) and in relation to T2 lesion volume at 2 years, but this effect was not sustained at 5 years (Havrdova and others 2009). The MECOMBIN study evaluated combination IFN-b with methylprednisolone (orally for 3 days every month) in treatment naive patients with EDSS <4 and used disability progression as the primary outcome. This study did not show a positive outcome in relation to the primary outcome; however, an effect on ARR, T2 lesion volume change, and MSFC was noted (Ravnborg and others 2009). However, such a dosing regime with corticosteroids has intolerable side effects for many and in the long term is not an acceptable treatment for most patients. The beneficial effect of steroids on relapse rate was also found in the NORMINS study, which recruited subjects established on IFN-b who had experienced at least one relapse in the preceding year while treated with IFN-b. Sixty-six patients were assigned to IFN-b and oral methylprednisolone and 64 were assigned to IFN-b and placebo. A high proportion of patients withdrew from the study before week 96 (26% on methylprednisolone versus 17% on placebo). The mean yearly relapse rate was 0.22 for methylprednisolone compared with 0.59 for placebo (62% reduction, 95% confidence interval 39% 77%; P < ) (Sorensen and others 2009). Sleep disturbance and neurological and psychiatric symptoms were the most frequent adverse events recorded in the methylprednisolone group. While the addition of 200 mg oral methylprednisolone daily reduced relapses as compared with placebo, the adverse events led to significant discontinuation; thus, despite being a positive study, this is not a treatment protocol that is likely to be widely used in treating people with MS. Conclusions IFN-b was the first specific disease-modifying treatment available for people with MS. It remains the most commonly prescribed first-line agent for relapsing disease and has been shown to have a modest effect on clinical and MRI outcomes. It is unclear what the mechanism of action responsible for the therapeutic effect is; however, possible mechanisms include reduced T-cell proliferation and T-cell activation, apoptosis of autoreactive T-cells, IFN-g antagonism, altered cytokine expression (Th2 favored), altered BBB permeability via integrin and MMP expression, and possibly an antiviral effect. Addition of polyethylene glycol (PEGylation) is a technique that has been successfully employed with IFN-a 2a in the treatment of hepatitis C, to improve the bioavailability of the IFN molecule and improve its efficacy (Lindsay and others 2001). Biogen Idec are currently recruiting subjects to evaluate the efficacy and safety of PEGylated IFN-b-1a (BIIB017) in subjects with relapsing MS. PEGylated IFN-b-1a or placebo is administered at a dose of 125 mg by SC injection every 2 4 weeks (2 active dosing arms) and the endpoints will include annualized relapse rate and MRI measures. If this is found to be effective, it would be a better treatment option than the existing preparations due to the dosing schedule. In addition to PEGylated IFN-b-1a several biosimilar products are available outside the protected markets of the European Union (EU) and North America. Whether or not these products will gain access to these markets will depend on regulatory requirements; at a minimum these agents will have to be shown to have low immunogenicity and some read-out that they are bioequivalent. It is apparent that not all patients respond to IFN-b treatment, and this lack of response may be mediated by NAbs or other unknown mechanisms. For nonresponders there are other options for treatment, including GA, natalizumab, or mitoxantrone, and in the next year or two, oral agents will be available outside of clinical trials. When assessing all of these agents it is difficult to compare individual trials as the patient cohorts are quite different regarding duration of disease and pretreatment relapse rates, exposure to prior disease modifying drugs (DMDs), primary outcome measures, and definitions of disease progression, and head-to-head trials versus IFN-b are ongoing or planned with many of these agents. In early aggressive MS, alemtuzumab outperformed IFN-b-1a 44 mcg TIW (Rebif) in the CAMS 223 trial. However, there appears to be a trade-off between efficacy and safety with many of the more potent agents carrying a higher risk for serious adverse events (PML, AML, cardiotoxicity, and ITP). However, it is recognized in subjects with MS that they are frequently willing to take more risk in the hope of preventing relapses and disease progression. MS patients will tolerate a relatively high risk of death for effective treatments of MS. Quality of life, worsening disease state, and poor ambulation are significantly correlated with MS patients tolerance to risk. Incorporation of these and weaker influences into a risk tolerance nomogram may aid clinicians in guiding patients to treatment which is appropriate to their level of risk tolerance (Kolattukudy and others 2007). There remains an unmet need in treating people MS, in that agents with improved efficacy, tolerability, and safety profiles are required. The anticipated agents, including mabs (alemtuzumab, rituximab, daclizumab, and oral agents), seem to meet 2 of these, but appear to confer a higher risk for the patient (or with some agents the risks remain uncertain). While the increased efficacy and tolerability of these agents suggests that IFN-b will no longer play a leading role in MS treatment, it is unlikely that those already established on treatment with IFN-b who have stable disease will elect to switch immediately to a drug whose adverse effects are potentially life threatening or unknown. It is likely that IFN-b will also play a role in combination strategies as an immunomodulatory agent after induction therapy with a cytotoxic agent or possibly in conjunction with neuroprotective drugs.

15 INTERFERON BETA IN MULTIPLE SCLEROSIS THERAPY 723 Author Disclosure Statement Professor Giovannoni reports having received consulting fees from Bayer-Schering Healthcare, Biogen-Idec, Genzyme, GlaxoSmithKline, Merck-Serono, Novartis, Protein Discovery Laboratories, Teva-Aventis, Vertex Pharmaceuticals and UCB Pharma; lecture fees from Bayer-Schering Healthcare, Biogen-Idec, Pfizer, Teva-Aventis, Vertex Pharmaceuticals; and grant support from Bayer-Schering Healthcare, Biogen-Idec, GW Pharma, Merck-Serono, Merz, Novartis, Teva-Aventis and UCB Pharma. References Aharoni R, Kayhan B, Eilam R, Sela M, Arnon R Glatiramer acetate-specific T cells in the brain express T helper 2/3 cytokines and brain-derived neurotrophic factor in situ. Proc Natl Acad Sci U S A 100: Aktas O, Kieseier B, Hartung HP Neuroprotection, regeneration and immunomodulation: broadening the therapeutic repertoire in multiple sclerosis. Trends Neurosci 33: Albert M, Antel J, Bruck W, Stadelmann C Extensive cortical remyelination in patients with chronic multiple sclerosis. Brain Pathol 17: Allen IV, McKeown SR A histological, histochemical and biochemical study of the macroscopically normal white matter in multiple sclerosis. J Neurol Sci 41: Andersen O, Lycke J, Tollesson PO, and others Linomide reduces the rate of active lesions in relapsing-remitting multiple sclerosis. Neurology 47: Bar-Or A, Calabresi PA, Arnold D, and others Rituximab in relapsing-remitting multiple sclerosis: a 72-week, open-label, phase I trial. Ann Neurol 63: Bartosik-Psujek H, Belniak E, Mitosek-Szewczyk K, Dobosz B, Stelmasiak Z Interleukin-8 and RANTES levels in patients with relapsing-remitting multiple sclerosis (RR-MS) treated with cladribine. Acta Neurol Scand 109: Bertolotto A, Gilli F, Sala A, and others Persistent neutralizing antibodies abolish the interferon beta bioavailability in MS patients. Neurology 60: Beutler E, Sipe J, Romine J, McMillan R, Zyroff J, Koziol J Treatment of multiple sclerosis and other autoimmune diseases with cladribine. Semin Hematol 33: Bielekova B, Catalfamo M, Reichert-Scrivner S, and others Regulatory CD56(bright) natural killer cells mediate immunomodulatory effects of IL-2Ralpha-targeted therapy (daclizumab) in multiple sclerosis. Proc Natl Acad Sci U S A 103: Borden EC, Sen GC, Uze G, and others Interferons at age 50: past, current and future impact on biomedicine. Nat Rev Drug Discov 6: Bowman EP, Chackerian AA, Cua DJ Rationale and safety of anti-interleukin-23 and anti-interleukin-17a therapy. Curr Opin Infect Dis 19: Brassat D, Recher C, Waubant E, and others Therapyrelated acute myeloblastic leukemia after mitoxantrone treatment in a patient with MS. Neurology 59: Brinkmann V FTY720 (fingolimod) in multiple sclerosis: therapeutic effects in the immune and the central nervous system. Br J Pharmacol 158: Brownell B, Hughes JT The distribution of plaques in the cerebrum in multiple sclerosis. J Neurol Neurosurg Psychiatry 25: Brunmark C, Runstrom A, Ohlsson L, and others The new orally active immunoregulator laquinimod (ABR ) effectively inhibits development and relapses of experimental autoimmune encephalomyelitis. J Neuroimmunol 130: Calabrese V, Ravagna A, Colombrita C, and others Acetylcarnitine induces heme oxygenase in rat astrocytes and protects against oxidative stress: involvement of the transcription factor Nrf2. J Neurosci Res 79: Calabresi PA, Giovannoni G, Confavreux C, and others The incidence and significance of anti-natalizumab antibodies: results from AFFIRM and SENTINEL. Neurology 69: Carson DA, Wasson DB, Taetle R, Yu A Specific toxicity of 2-chlorodeoxyadenosine toward resting and proliferating human lymphocytes. Blood 62: Charcot M Histologie de la sclerose en plaque. Gazette Hôpitaux (Paris) 41:554, , 566. Chen XL, Dodd G, Thomas S, and others Activation of Nrf2/ARE pathway protects endothelial cells from oxidant injury and inhibits inflammatory gene expression. Am J Physiol Heart Circ Physiol 290:H1862 H1870. Clerico M, Faggiano F, Palace J, Rice G, Tintore M, Durelli L Recombinant interferon beta or glatiramer acetate for delaying conversion of the first demyelinating event to multiple sclerosis. Cochrane Database Syst Rev (2):CD Clifford DB, De LA, Simpson DM, Arendt G, Giovannoni G, Nath A Natalizumab-associated progressive multifocal leukoencephalopathy in patients with multiple sclerosis: lessons from 28 cases. Lancet Neurol 9: Cohen JA, Barkhof F, Comi G, and others Oral fingolimod or intramuscular interferon for relapsing multiple sclerosis. N Engl J Med 362: Cohen JA, Imrey PB, Calabresi PA, and others Results of the Avonex Combination Trial (ACT) in relapsing-remitting MS. Neurology 72: Coles AJ, Compston DA, Selmaj KW, and others Alemtuzumab vs. interferon beta-1a in early multiple sclerosis. N Engl J Med 359: Coles AJ, Cox A, Le PE, and others The window of therapeutic opportunity in multiple sclerosis: evidence from monoclonal antibody therapy. J Neurol 253: Comi G, Filippi M, Barkhof F, and others. 2001a. Effect of early interferon treatment on conversion to definite multiple sclerosis: a randomised study. Lancet 357: Comi G, Filippi M, Wolinsky JS. 2001b. European/Canadian multicenter, double-blind, randomized, placebo-controlled study of the effects of glatiramer acetate on magnetic resonance imaging measured disease activity and burden in patients with relapsing multiple sclerosis. European/Canadian Glatiramer Acetate Study Group. Ann Neurol 49: Comi G, Pulizzi A, Rovaris M, and others Effect of laquinimod on MRI-monitored disease activity in patients with relapsingremitting multiple sclerosis: a multicentre, randomised, doubleblind, placebo-controlled phase IIb study. Lancet 371: Cook SD, Quinless JR, Jotkowitz A, Beaton P Serum IFN neutralizing antibodies and neopterin levels in a cross-section of MS patients. Neurology 57: Cox LB, Fantl P, Fitzpatrick M The treatment of disseminated sclerosis by prolonged lowering of the blood prothrombin level. Med J Aust 1: Dawson JW The histology of disseminated sclerosis. Trans Roy Soc Edinb 50:517. Drago F, Brusati C, Mancardi G, Murialdo A, Rebora A Localized lipoatrophy after glatiramer acetate injection in patients with remitting-relapsing multiple sclerosis. Arch Dermatol 135: Edan G Immunosuppressive therapy in multiple sclerosis. Neurol Sci 26 (Suppl. 1):S18.

16 724 FARRELL AND GIOVANNONI Edan G, Miller D, Clanet M, and others Therapeutic effect of mitoxantrone combined with methylprednisolone in multiple sclerosis: a randomised multicentre study of active disease using MRI and clinical criteria. J Neurol Neurosurg Psychiatry 62: Edan G, Morrissey SP, Hartung HP Use of mitoxantrone to treat MS. In: Multiple Sclerosis Therapeutics, 2nd ed (Cohen JA, Rudick RA, eds). London (UK): Martin Dunitz. pp Farrell RA, Giovannoni G Measuring and management of anti-interferon beta antibodies in subjects with multiple sclerosis. Mult Scler 13: Fidler JM, DeJoy SQ, Gibbons JJ Jr Selective immunomodulation by the antineoplastic agent mitoxantrone. I. Suppression of B lymphocyte function. J Immunol 137: Filippi M, Rovaris M, Rocca MA, Sormani MP, Wolinsky JS, Comi G Glatiramer acetate reduces the proportion of new MS lesions evolving into black holes. Neurology 57: Foote AK, Blakemore WF Inflammation stimulates remyelination in areas of chronic demyelination. Brain 128: Foster CA, Mechtcheriakova D, Storch MK, Balatoni B, Howard LM, Bornancin F, Wlachos A, Sobanov J, Kinnunen A, Baumruker T FTY720 rescue therapy in the dark agouti rat model of experimental autoimmune encephalomyelitis: expression of central nervous system genes and reversal of blood-brain-barrier damage. Brain Pathol 19: Francis GS, Rice GP, Alsop JC Interferon beta-1a in MS: results following development of neutralizing antibodies in PRISMS. Neurology 65: Franklin RJ, Ffrench-Constant C Remyelination in the CNS: from biology to therapy. Nat Rev Neurosci 9: Frenette PS, Wagner DD. 1996a. Adhesion molecules part 1. N Engl J Med 334: Frenette PS, Wagner DD. 1996b. Adhesion molecules part II: blood vessels and blood cells. N Engl J Med 335: Frischer JM, Bramow S, Dal-Bianco A, and others The relation between inflammation and neurodegeneration in multiple sclerosis brains. Brain 132: Ge Y, Grossman RI, Udupa JK, and others Glatiramer acetate (Copaxone) treatment in relapsing-remitting MS: quantitative MR assessment. Neurology 54: Ghalie RG, Edan G, Laurent M, and others. 2002a. Cardiac adverse effects associated with mitoxantrone (Novantrone) therapy in patients with MS. Neurology 59: Ghalie RG, Mauch E, Edan G, and others. 2002b. A study of therapy-related acute leukaemia after mitoxantrone therapy for multiple sclerosis. Mult Scler 8: Gilmore CP, Bo L, Owens T, Lowe J, Esiri MM, Evangelou N Spinal cord gray matter demyelination in multiple sclerosis-a novel pattern of residual plaque morphology. Brain Pathol 16: Giovannoni G, Comi G, Cook S, and others A placebocontrolled trial of oral cladribine for relapsing multiple sclerosis. N Engl J Med 362: Goodin DS, Frohman EM, Hurwitz B, and others Neutralizing antibodies to interferon beta: assessment of their clinical and radiographic impact: an evidence report: report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology 68: Granucci F, Feau S, Angeli V, Trottein F, Ricciardi-Castagnoli P Early IL-2 production by mouse dendritic cells is the result of microbial-induced priming. J Immunol 170: Hartung HP, Aktas O Bleak prospects for primary progressive multiple sclerosis therapy: downs and downs, but a glimmer of hope. Ann Neurol 66: Hartung HP, Gonsette R, Konig N, and others Mitoxantrone in progressive multiple sclerosis: a placebo-controlled, doubleblind, randomised, multicentre trial. Lancet 360: Hartung HP, Munschauer FE III Assessment and management of neutralizing antibodies in patients with multiple sclerosis. J Neurol 251 (Suppl. 2):II40 II42. Hauser SL, Waubant E, Arnold DL, and others B-cell depletion with rituximab in relapsing-remitting multiple sclerosis. N Engl J Med 358: Havrdova E, Zivadinov R, Krasensky J, and others Randomized study of interferon beta-1a, low-dose azathioprine, and low-dose corticosteroids in multiple sclerosis. Mult Scler 15: Hawker K, O Connor P, Freedman MS, and others Rituximab in patients with primary progressive multiple sclerosis: results of a randomized double-blind placebo-controlled multicenter trial. Ann Neurol 66: Hemmer B, Cepok S, Nessler S, Sommer N Pathogenesis of multiple sclerosis: an update on immunology. Curr Opin Neurol 15: Hohlfeld R, Kerschensteiner M, Meinl E Dual role of inflammation in CNS disease. Neurology 68:S58 S63. Hohlfeld R, Wekerle H Autoimmune concepts of multiple sclerosis as a basis for selective immunotherapy: from pipe dreams to (therapeutic) pipelines. Proc Natl Acad Sci U S A 101 (Suppl. 2): Hubbs AF, Benkovic SA, Miller DB, and others Vacuolar leukoencephalopathy with widespread astrogliosis in mice lacking transcription factor Nrf2. Am J Pathol 170: Hussien Y, Sanna A, Soderstrom M, Link H, Huang YM Glatiramer acetate and IFN-beta act on dendritic cells in multiple sclerosis. J Neuroimmunol 121: Interferon Beta Study Group Neutralizing antibodies during treatment of multiple sclerosis with interferon beta-1b: experience during the first three years. The IFNB Multiple Sclerosis Study Group and the University of British Columbia MS/MRI Analysis Group. Neurology 47: Itoh K, Chiba T, Takahashi S, and others An Nrf2/small Maf heterodimer mediates the induction of phase II detoxifying enzyme genes through antioxidant response elements. Biochem Biophys Res Commun 236: Jacobs LD, Beck RW, Simon JH, and others Intramuscular interferon beta-1a therapy initiated during a first demyelinating event in multiple sclerosis. CHAMPS Study Group. N Engl J Med 343: Jacobs LD, Cookfair DL, Rudick RA, and others Intramuscular interferon beta-1a for disease progression in relapsing multiple sclerosis. The Multiple Sclerosis Collaborative Research Group (MSCRG). Ann Neurol 39: Jacobsen M, Cepok S, Quak E, and others Oligoclonal expansion of memory CD8þ T cells in cerebrospinal fluid from multiple sclerosis patients. Brain 125: Johnson KP, Brooks BR, Cohen JA, and others Copolymer 1 reduces relapse rate and improves disability in relapsingremitting multiple sclerosis: results of a phase III multicenter, double-blind placebo-controlled trial. The Copolymer 1 Multiple Sclerosis Study Group. Neurology 45: Jung CG, Kim HJ, Miron VE, and others Functional consequences of S1P receptor modulation in rat oligodendroglial lineage cells. Glia 55: Kappos L, Clanet M, Sandberg-Wollheim M, and others Neutralizing antibodies and efficacy of interferon beta-1a: a 4- year controlled study. Neurology 65: Kappos L, Freedman MS, Polman CH, and others Effect of early versus delayed interferon beta-1b treatment on disability

17 INTERFERON BETA IN MULTIPLE SCLEROSIS THERAPY 725 after a first clinical event suggestive of multiple sclerosis: a 3-year follow-up analysis of the BENEFIT study. Lancet 370: Kappos L, Gold R, Miller DH, and others Efficacy and safety of oral fumarate in patients with relapsing-remitting multiple sclerosis: a multicentre, randomised, double-blind, placebo-controlled phase IIb study. Lancet 372: Kappos L, Radue EW, O Connor P, and others A placebocontrolled trial of oral fingolimod in relapsing multiple sclerosis. N Engl J Med 362: Karussis DM, Meiner Z, Lehmann D, and others Treatment of secondary progressive multiple sclerosis with the immunomodulator linomide: a double-blind, placebo-controlled pilot study with monthly magnetic resonance imaging evaluation. Neurology 47: Kim HJ, Ifergan I, Antel JP, and others Type 2 monocyte and microglia differentiation mediated by glatiramer acetate therapy in patients with multiple sclerosis. J Immunol 172: Koch MW, Mostert JP, de Vries JJ, De KJ Treatment with interferon beta-1b delays conversion to clinically definite and McDonald MS in patients with clinically isolated syndromes. Neurology 68: Kolattukudy SJ, Kattan MW, Miller D, Fox R Risk tolerance in multiple sclerosis patients. AAN Abstract P Korn T, Magnus T, Toyka K, Jung S Modulation of effector cell functions in experimental autoimmune encephalomyelitis by leflunomide mechanisms independent of pyrimidine depletion. J Leukoc Biol 76: Korn T, Toyka K, Hartung HP, Jung S Suppression of experimental autoimmune neuritis by leflunomide. Brain 124: Kutzelnigg A, Faber-Rod JC, Bauer J, and others Widespread demyelination in the cerebellar cortex in multiple sclerosis. Brain Pathol 17: Kutzelnigg A, Lucchinetti CF, Stadelmann C, and others Cortical demyelination and diffuse white matter injury in multiple sclerosis. Brain 128: Lassmann H, Wekerle H The pathology of multiple sclerosis. In: Compston A, Confavreux C, Lassman H, McDonald I, et al. In: McAlpine s Multiple Sclerosis. London: Churchill Livingstone. pp Leary SM, Miller DH, Stevenson VL, Brex PA, Chard DT, Thompson AJ Interferon beta-1a in primary progressive MS: an exploratory, randomized, controlled trial. Neurology 60: Le Page E, Leray E, Taurin G, and others Mitoxantrone as induction treatment in aggressive relapsing remitting multiple sclerosis: treatment response factors in a 5 year follow-up observational study of 100 consecutive patients. J Neurol Neurosurg Psychiatry 79: Leppert D, Waubant E, Burk MR, Oksenberg JR, Hauser SL Interferon beta-1b inhibits gelatinase secretion and in vitro migration of human T cells: a possible mechanism for treatment efficacy in multiple sclerosis. Ann Neurol 40: Li J, Johnson D, Calkins M, Wright L, Svendsen C, Johnson J Stabilization of Nrf2 by tbhq confers protection against oxidative stress-induced cell death in human neural stem cells. Toxicol Sci 83: Liliemark J The clinical pharmacokinetics of cladribine. Clin Pharmacokinet 32: Lindsay KL, Trepo C, Heintges T, and others A randomized, double-blind trial comparing pegylated interferon alfa- 2b to interferon alfa-2b as initial treatment for chronic hepatitis C. Hepatology 34: Lublin FD, Lavasa M, Viti C, Knobler RL Suppression of acute and relapsing experimental allergic encephalomyelitis with mitoxantrone. Clin Immunol Immunopathol 45: Malucchi S, Sala A, Gilli F, and others Neutralizing antibodies reduce the efficacy of betaifn during treatment of multiple sclerosis. Neurology 62: Miller A, Shapiro S, Gershtein R, and others Treatment of multiple sclerosis with copolymer-1 (Copaxone): implicating mechanisms of Th1 to Th2/Th3 immune-deviation. J Neuroimmunol 92: Miller DH, Soon D, Fernando KT, and others MRI outcomes in a placebo-controlled trial of natalizumab in relapsing MS. Neurology 68: Miller DH, Thompson AJ, Filippi M Magnetic resonance studies of abnormalities in the normal appearing white matter and grey matter in multiple sclerosis. J Neurol 250: Mix E, Meyer-Rienecker H, Zettl UK Animal models of multiple sclerosis for the development and validation of novel therapies potential and limitations. J Neurol 255 (Suppl. 6):7 14. Montalban X Overview of European pilot study of interferon beta-ib in primary progressive multiple sclerosis. Mult Scler 10 (Suppl. 1):S62 S64. MS Study Group Interferon beta-1b is effective in relapsing-remitting multiple sclerosis. I. Clinical results of a multicenter, randomized, double-blind, placebo-controlled trial. The IFNB Multiple Sclerosis Study Group. Neurology 43: Munschauer FE III, Kinkel RP Managing side effects of interferon-beta in patients with relapsing-remitting multiple sclerosis. Clin Ther 19: O Connor P The effects of intramuscular interferon beta-ia in patients at high risk for development of multiple sclerosis: a post hoc analysis of data from CHAMPS. Clin Ther 25: O Connor PW, Li D, Freedman MS, and others A Phase II study of the safety and efficacy of teriflunomide in multiple sclerosis with relapses. Neurology 66: Panitch H, Goodin DS, Francis G, and others Randomized, comparative study of interferon beta-1a treatment regimens in MS: The EVIDENCE Trial. Neurology 59: Pender MP, Greer JM Immunology of multiple sclerosis. Curr Allergy Asthma Rep 7: Peterson JW, Bo L, Mork S, Chang A, Trapp BD Transected neurites, apoptotic neurons, and reduced inflammation in cortical multiple sclerosis lesions. Ann Neurol 50: Polman C, Barkhof F, Sandberg-Wollheim M, Linde A, Nordle O, Nederman T Treatment with laquinimod reduces development of active MRI lesions in relapsing MS. Neurology 64: Polman CH, O Connor PW, Havrdova E, and others A randomized, placebo-controlled trial of natalizumab for relapsing multiple sclerosis. N Engl J Med 354: PRISMS Randomised double-blind placebo-controlled study of interferon beta-1a in relapsing/remitting multiple sclerosis. PRISMS (Prevention of Relapses and Disability by Interferon beta-1a Subcutaneously in Multiple Sclerosis) Study Group. Lancet 352: PRISMS Study Group and the University of British Columbia MS/MRI Analysis Group PRISMS-4: long-term efficacy of interferon-beta-1a in relapsing MS. Neurology 56: Putnam TJ, Chiavacci LV Results of treatment of multiple sclerosis with dicoumarin. Arch Neurol Psychiatry 57:1 13. Ramtahal J, Jacob A, Das K, Boggild M Sequential maintenance treatment with glatiramer acetate after mitoxantrone

18 726 FARRELL AND GIOVANNONI is safe and can limit exposure to immunosuppression in very active, relapsing remitting multiple sclerosis. J Neurol 253: Ravnborg M, Bendtzen K, Christensen O, and others Treatment with azathioprine and cyclic methylprednisolone has little or no effect on bioactivity in anti-interferon beta antibody-positive patients with multiple sclerosis. Mult Scler 15: Rice GP, Filippi M, Comi G Cladribine and progressive MS: clinical and MRI outcomes of a multicenter controlled trial. Cladribine MRI Study Group. Neurology 54: Rose AS, Kuzma JW, Kurtzke JF, Namerow NS, Sibley WA, Tourtellotte WW Cooperative study in the evaluation of therapy in multiple sclerosis. ACTH vs. placebo final report. Neurology 20:1 59. Rose AS, Kuzma JW, Kurtzke JF, Sibley WA, Tourtellotte WW Cooperative study in the evaluation of therapy in multiple sclerosis; ACTH vs placebo in acute exacerbations. Preliminary report. Neurology 18: Suppl-10. Rose AS, Kuzma JW, Kurtzke JF, Sibley WA, Tourtellotte WW Cooperative study in the evaluation of therapy in multiple sclerosis: ACTH vs placebo in acute exacerbation. Trans Am Neurol Assoc 94: Rudick RA, Stuart WH, Calabresi PA, and others Natalizumab plus interferon beta-1a for relapsing multiple sclerosis. N Engl J Med 354: Savitsky N, Karliner W Electroshock therapy and multiple sclerosis. N Y State J Med 51:788. Sayre LM, Perry G, Smith MA Oxidative stress and neurotoxicity. Chem Res Toxicol 21: Schilling S, Goelz S, Linker R, Luehder F, Gold R Fumaric acid esters are effective in chronic experimental autoimmune encephalomyelitis and suppress macrophage infiltration. Clin Exp Immunol 145: Sharief MK, Semra YK, Seidi OA, Zoukos Y Interferonbeta therapy downregulates the anti-apoptosis protein FLIP in T cells from patients with multiple sclerosis. J Neuroimmunol 120: Simarro-Puig J, Llor CJ Treatment of multiple sclerosis with nitrogen mustards. Rev Clin Esp 40: Smolen JS, Emery P, Kalden JR, and others The efficacy of leflunomide monotherapy in rheumatoid arthritis: towards the goals of disease modifying antirheumatic drug therapy. J Rheumatol Suppl 71: Sorensen PS, Deisenhammer F, Duda P, and others Guidelines on use of anti-ifn-beta antibody measurements in multiple sclerosis: report of an EFNS Task Force on IFNbeta antibodies in multiple sclerosis. Eur J Neurol 12: Sorensen PS, Mellgren SI, Svenningsson A, and others NORdic trial of oral Methylprednisolone as add-on therapy to Interferon beta-1a for treatment of relapsing-remitting Multiple Sclerosis (NORMIMS study): a randomised, placebocontrolled trial. Lancet Neurol 8: Sospedra M, Martin R Immunology of multiple sclerosis. Annu Rev Immunol 23: Stadelmann C, Kerschensteiner M, Misgeld T, Bruck W, Hohlfeld R, Lassmann H BDNF and gp145trkb in multiple sclerosis brain lesions: neuroprotective interactions between immune and neuronal cells? Brain 125: Teitelbaum D, Aharoni R, Klinger E, and others Oral glatiramer acetate in experimental autoimmune encephalomyelitis: clinical and immunological studies. Ann N Y Acad Sci 1029: Trapp BD, Peterson J, Ransohoff RM, Rudick R, Mork S, Bo L Axonal transection in the lesions of multiple sclerosis. N Engl J Med 338: van Sechel AC, Bajramovic JJ, van Stipdonk MJ, Persoon-Deen C, Geutskens SB, van Noort JM EBV-induced expression and HLA-DR-restricted presentation by human B cells of alpha B-crystallin, a candidate autoantigen in multiple sclerosis. J Immunol 162: Waxman SG. 2006a. Axonal conduction and injury in multiple sclerosis: the role of sodium channels. Nat Rev Neurosci 7: Waxman SG. 2006b. Ions, energy and axonal injury: towards a molecular neurology of multiple sclerosis. Trends Mol Med 12: Wierinckx A, Breve J, Mercier D, Schultzberg M, Drukarch B, Van Dam AM Detoxication enzyme inducers modify cytokine production in rat mixed glial cells. J Neuroimmunol 166: Wolinsky JS The PROMiSe trial: baseline data review and progress report. Mult Scler 10 (Suppl. 1):S65 S71. Wolinsky JS, Narayana PA, Noseworthy JH, and others Linomide in relapsing and secondary progressive MS: part II: MRI results. MRI Analysis Center of the University of Texas- Houston, Health Science Center, and the North American Linomide Investigators. Neurology 54: Wynn D, Kaufman M, Montalban X, and others Daclizumab in active relapsing multiple sclerosis (CHOICE study): a phase 2, randomised, double-blind, placebo-controlled, addon trial with interferon beta. Lancet Neurol 9: Xu X, Blinder L, Shen J, and others In vivo mechanism by which leflunomide controls lymphoproliferative and autoimmune disease in MRL/MpJ-lpr/lpr mice. J Immunol 159: Yang JS, Xu LY, Xiao BG, Hedlund G, Link H Laquinimod (ABR ) suppresses the development of experimental autoimmune encephalomyelitis, modulates the Th1/Th2 balance and induces the Th3 cytokine TGF-beta in Lewis rats. J Neuroimmunol 156:3 9. Yong VW Differential mechanisms of action of interferonbeta and glatiramer aetate in MS. Neurology 59: Yong VW, Chabot S, Stuve O, Williams G Interferon beta in the treatment of multiple sclerosis: mechanisms of action. Neurology 51: Zeis T, Graumann U, Reynolds R, Schaeren-Wiemers N Normal-appearing white matter in multiple sclerosis is in a subtle balance between inflammation and neuroprotection. Brain 131: Zhao J, Moore AN, Redell JB, Dash PK Enhancing expression of Nrf2-driven genes protects the blood brain barrier after brain injury. J Neurosci 27: Ziemssen T, Kumpfel T, Klinkert WE, Neuhaus O, Hohlfeld R Glatiramer acetate-specific T-helper 1- and 2-type cell lines produce BDNF: implications for multiple sclerosis therapy. Brain-derived neurotrophic factor. Brain 125: Address correspondence to: Dr. Rachel A. Farrell Institute of Neurology University College London Queen Square London WC1N 3BG United Kingdom rfarrell@ion.ucl.ac.uk Received 15 July 2010/Accepted 15 July 2010

19 JOURNAL OF INTERFERON & CYTOKINE RESEARCH Volume 30, Number 10, 2010 ª Mary Ann Liebert, Inc. DOI: /jir Single-Nucleotide Polymorphisms in Response to Interferon-Beta Therapy in Multiple Sclerosis Koen Vandenbroeck 1,2 and Manuel Comabella 3 Interferon-beta (IFN-b) is one of the main first-line disease-modifying drugs indicated for the treatment of multiple sclerosis ( MS). The drug exhibits only limited effectiveness, and does not produce clinical benefits in around 20% 50% of patients. The availability of biomarkers would be beneficial for identification of patients at high risk of treatment failure, before initiation of therapy. Over the last 5 years, the search for such biomarkers has intensified and various promising candidates have been uncovered. Here, we review the main attempts undertaken to identify polymorphic variants associated with response to IFN-b therapy in MS by means of candidate gene approaches and whole-genome association scans. Despite substantial progress made in the field, there is still a long way to go before biomarker discoveries can be incorporated into clinical practice to predict IFN-b-responder status in MS patients. Introduction Multiple sclerosis (MS) is considered a T cellmediated autoimmune disease of the central nervous system that mainly affects young people and leads to important neurological disability. Although the etiology of MS is unknown, disease phenotype is possibly the result of the interaction between a complex genetic background and environmental triggers such as viral infections (as reviewed by Sospedra and Martin 2005). There are 3 main clinical forms of MS, ie, relapsing-remitting MS, secondary progressive MS, and primary progressive MS (Lublin and Reingold 1996). Patients who have a first relapse but are not yet found to have MS are referred to as clinically isolated syndromes, and this is considered the earliest manifestation of MS. Interferon-beta (IFN-b) was the first disease-modifying therapy approved by the Food and Drug Administration for MS treatment. In clinically isolated syndrome patients, IFN-b is effective in delaying conversion to clinically definite MS ( Jacobs and others 2000; Comi and others 2001; Kappos and others 2006). In relapsing-remitting MS, IFN-b has demonstrated beneficial effects on decreasing the relapse rate, delaying the time to sustained disability progression, and reducing brain disease activity as assessed by magnetic resonance imaging (MRI) (The Interferon b Multiple Sclerosis Study Group 1993; Jacobs and others 1996; PRISMS Study Group 1998). In secondary progressive MS, although inconsistent findings in terms of effect on disability were observed, IFN-b demonstrated positive effects on relapses and brain MRI activity (European Study Group on Interferon Beta-1b in Secondary Progressive MS 1998; Cohen and others 2002; Panitch and others 2004). Despite the beneficial effects of IFN-b in treatment of MS, the drug is only partially effective, and its long-term impact on disease progression remains unknown. Further, there is a relatively large proportion of patients who do not respond to IFN-b, estimated at 20% 55% of treated patients depending on the clinical and radiological criteria used to evaluate treatment failure (Río and others 2002). Unfortunately, response criteria to IFN-b are usually applied after 1 or 2 years of follow-up, a period of time during which MS patients will be treated without any benefit and at high socioeconomic cost. Thus, candidate response biomarkers should allow, either separately or in combination with clinical and/or radiological data, the early identification of treatment failure or ideally even predict nonresponder status. Although several molecules have recently been proposed as response biomarkers (Wandinger and others 2003; Baranzini and others 2005; Soilu-Hanninen and others 2005; Minagar and others 2007; Krumbholz and others 2008), up to now, it is impossible to predict which patients will respond to IFN-b in MS. Markers of response to IFN-b can be detected at the gene, mrna, and protein levels. The present review will focus on the candidate gene studies and whole-genome single-nucleotide polymorphism (SNP) screens that have pursued the discovery of allelic variants associated with the response to IFN-b 1 Neurogenomiks Group, Universidad del País Vasco (UPV/EHU), Leioa, Spain. 2 IKERBASQUE, Basque Foundation for Science, Bilbao, Spain. 3 Centre d Esclerosi Múltiple de Catalunya, CEM-Cat, Unitat de Neuroimmunologia Clínica, H. Universitari Vall d Hebron (HUVH), Barcelona, Spain. 727

20 728 VANDENBROECK AND COMABELLA treatment in MS patients (Table 1). Recently, 3 independent whole-genome association screens have converged toward the identification of interleukin-28b (IL28B) as main gene involved in the response to IFN-a ribavirin-mediated clearance of viral hepatitis C RNA from serum (Ge and others 2009; Suppiah and others 2009; Tanaka and others 2009). These studies unequivocally illustrate the power and validity of model-independent SNP screens as tools toward identification of type I IFN therapy response modifiers. While our knowledge on IFN-b response genes in MS is still fragmentary, as discussed below, there is an increasing consensus on a polygenic mechanism of response involving a series of brain-specific genes, including genes belonging to the glutamatergic system, as well as type I IFN signature genes. Candidate Gene Studies The first step in signal transduction consists of the binding of IFN-b to its unique heterodimeric cell surface receptor complex, composed of IFNAR1 and IFNAR2 subunits (Fig. 1). Ensuing activation of the JAK-STAT pathway leads to assembly of an IFN-stimulated gene factor 3 complex consisting of STAT1, STAT2, and IRF9 that is capable of translocating to the cell nucleus where it binds to IFN-stimulated response elements (ISREs), most frequently located in the 5 0 regulatory regions of IFN-inducible genes, and activates gene transcription. In addition, STAT homo- or heterodimers may be formed, which can regulate gene transcription through binding to IFNactivated site (GAS) elements, also located in 5 0 regulatory sequences. Thus, any polymorphisms located in genes coding for components of either of both canonical signaling pathways or in more downstream type I IFN-responsive genes may constitute potential response-modifier genetic biomarkers (O Doherty and others 2007) and, as such, merit scrutiny. Sriram and others (2003) analyzed a series of polymorphisms in IFNAR1 and IFNAR2 in a group of IFN-b-treated MS patients but did not find evidence for association with response, exception made for an intronic SNP in IFNAR1, which displayed a trend toward relapse-free status. In the study by Leyva and others (2005), SNPs in IFNAR1 or IF- NAR2 were found not to be associated with response to IFNb treatment. In a group of Northern Irish and Irish MS patients treated with IFN-b, a polymorphic GT n repeat element located in the promoter of the IFNAR1 gene showed a weak association with response (Cunningham and others 2005), while a SNP in the 3 0 region of IFNAR2 was not associated (O Doherty and others 2009). As described in more detail below, an intronic SNP in IFNAR2 emerged from the genome-wide 500K SNP screen by Comabella and others (2009a) as potential response modifier, even though the Reference Table 1. Summary of Polymorphic Variants Associated with Response to Interferon-Beta in Multiple Sclerosis at a Significance Level of P < 0.05 (Uncorrected) Genes or gene combinations associated rs numbers Observation Cunningham and others (2005) IFNAR1 GT n repeat in promoter Associated with response CTSS rs PSMB8 rs MX1 rs , rs Wergeland and others (2005) IL10 rs , rs , rs Trend toward fewer MRI lesions in non-gcc haplotypes Martínez and others (2006) IFNG CA n repeat in first intron Associated with response Byun and others (2008) a HAPLN1 rs Associated with response GPC5 rs , rs COL25A1 rs NPAS3 rs ERC2 (CAST) rs FAM19A1 rs LOC rs Comabella and others (2009a) a GRIA3 rs Associated with response CIT rs ADAR rs ZFAT rs STARD13 rs ZFHX4 rs IFNAR2 rs Cénit and others (2009) GPC5 rs Associated with response O Doherty and others (2009) b Combination 1 { JAK2 rs Associated with response IL10RB rs GBP1 rs PIAS1 rs Combination 2 JAK2 rs IL10 rs { CASP3 rs a Intergenic single-nucleotide polymorphisms not included. b Only the two most significant allele combinations are presented. Abbreviation: MRI, magnetic resonance imaging.

21 SNPS AND IFN-BETA TREATMENT IN MS 729 FIG. 1. Type I interferon (IFN) signaling pathways. After interaction with its heterodimeric receptor, type I IFNs activate the JAK-STAT signaling pathway. As a consequence, IFN-stimulated gene factor 3 (ISGF3) complexes are formed that move into the nucleus where they bind to IFN-stimulated response elements (ISREs) in the promoter of IFN-responsive genes, and initiate transcription. In addition, STAT homo- or heterodimers are generated that display intrinsic ability to bind to IFN-activated site (GAS) elements in gene promoters. Other signaling cascades, not indicated in the figure, can be activated as well. These include the mitogen-activated protein kinase and the phosphoinositide 3-kinase signaling pathways. Cytoplasmic suppressors of cytokine signaling and nuclear protein inhibitors of activated STAT are known to modulate the amplitude of the signal induced by IFN. earlier 300K SNP screen by Byun and others (2008) did not uncover any associations with type I IFN receptor genes. Taken together, current evidence regarding a putative role of IFNAR1 and/or IFNAR2 polymorphisms in determining individual response of MS patients to IFN-b is conflicting and therefore inconclusive. A more final verdict could perhaps be reached upon targeted high-density screening of SNPs spanning the IFNAR1 and IFNAR2 loci in a more highly powered and clinically well-characterized cohort. SNPs in 34 other candidate response genes, many of which belonged to IFN-related signaling pathways, including JAK1, STAT1, TYK2, JAK2, IRF4, and IRF9, were not found to be associated with response (O Doherty and others 2009). Still, in the same study, combinations of allelic variants imputed via a Markov chain Monte Carlo based method differed significantly between responders and nonresponders. Overall examination of the top scoring allele combinations revealed that JAK2, JAK1, IL10RB, GBP1, and PNPT1 were the most frequently encountered genes. This raises the possibility that rather than individual polymorphic variants, gene interaction may constitute the driving force in bringing about phenotypic differences in response of MS patients to IFN-b (O Doherty and others 2009). As IFN-b exerts its effect at least partially through interaction of IFN-stimulated gene factor 3 with ISREs, Cunningham and others (2005) sequenced the promoter regions encompassing ISREs in 100 type I IFN responsive genes to identify polymorphisms potentially altering the ISRE consensus motif and, hence, theoretically type I IFN inducibility. In total 51 SNPs, 2 repeat elements and 1 novel 3-nucleotide duplication were identified in 32 genes. Four genes were identified containing polymorphisms associated with response to treatment: MX1, cathepsin S (CTSS), proteasome subunit, b type, 8 (PSMB8/LMP7), and, as indicated above, IFNAR1. In the study by Weinstock-Guttman and others (2007), 2 SNPs in the MX1 promoter were found not to be associated with response to IFN-b. Further candidate gene studies have focused principally on IL10, IFNG, and the HLA region. In a Norwegian study, haplotype distribution of 3 SNPs in the promoter region of IL10 was analyzed for initial IFN treatment response (Wergeland and others 2005). Although the authors did not find any evidence for effects on clinical disease activity, MS patients with non-gcc haplotypes experienced fewer MRI lesions. Martínez and others (2006) found evidence for allelic association of an intronic CA n repeat in the first intron of IFNG with response of MS patients to IFN-b. In a group of Spanish MS patients, the HLA-DR2 (DRB1*1501, DQB1*0602) haplotype did not display differential distribution between responders and nonresponders to therapy (Villoslada and others 2002). Fernández and others (2005) were also unable to demonstrate association of HLA genotype with clinical response to IFN-b. Finally, the influence of HLA class I and II genes on response to IFN-b in MS was investigated by Comabella and others (2009b). HLA-A, -B, -C and -DRB1, -DQA1, and -DQB1 alleles were not distributed differentially between responders and nonresponders selected on the basis of stringent clinical criteria. Box Genome-wide association studies (GWAS) are experimental approaches that scan high numbers of polymorphic variants, mostly single-nucleotide polymorphisms (SNP) spread over the genome, in 2 different groups of patients and control subjects, or as is frequently the case in pharmacogenomics studies, in 2 groups of patients that respond differentially to a therapy. GWAS typically include <100K to >1M SNPs, either distributed evenly over the genome or enriched in gene regions. GWAS are model independent, as no assumptions are made in advance on specific selection of genes. Candidate gene approaches specifically assess polymorphic variants in genes thought to be involved either in the etiopathogenesis of a disease, or, as in pharmacogenomics studies, in drug mechanisms. For instance, genes belonging to type I IFN signaling pathways as well as type I IFN response genes are credible candidate genes for investigation in IFN-b therapy of MS. The main drawback of the candidate gene approach resides in its dependence on specific knowledge of the biological mode of action of the drug under investigation. The increased accessibility to and decreasing costs of GWAS are driving a gradual replacement of the candidate gene approach as primary gene-hunting tool by GWAS.

22 730 VANDENBROECK AND COMABELLA In conclusion, the candidate gene approach applied to the pharmacogenomics of IFN-b in MS has yielded at best a few weak associations, none of which have been replicated. As described below, whole-genome association studies have revealed that clinical benefits in response to IFN-b appear to be dependent on genetic variation in multiple gene loci. As such, low penetrance, gene gene and/or gene environment interactions, in addition to limited statistical power, are likely to jeopardize robust identification of treatment response genes, especially so in the small cohorts (< included response-classified MS patients) typically employed in the aforementioned MS pharmacogenomics studies (see also Vandenbroeck and Matute 2008). At any rate, structured network-based attempts are being undertaken to produce the larger sample sizes needed to identify and validate IFN-b response-modifying polymorphisms with greater level of confidence (Vandenbroeck and others 2009). Whole-Genome SNP Screens To date, only 2 IFN-b pharmacogenomics whole-genome association studies have been reported in MS (Byun and others 2008; Comabella and others 2009a). Using a poolingbased strategy, both studies aimed to identify allele variants associated with the response to IFN-b. In the study by Byun and colleagues (2008), 206 IFN-b-treated MS patients from 4 collaborative centers were genotyped by means of the Affymetrix 100K SNP arrays. One of the main and most intriguing findings of the study was the enrichment for genes encoding g-aminobutyric or glutamate receptors among the SNPs that best discriminated between responders and no-responders to IFN-b. This result opened for the first time an attractive link between neuronal excitation and response to IFN-b. Among the genes that distinguished between responders and nonresponders and were validated by individual genotyping, it was interesting to observe genes such as glypican 5 (GPC5) and neuronal PAS domain protein 3 (NPAS3), which are expressed in neurons and play roles in neurogenesis and neuroprotection. Of note, the importance of GPC5 as an IFN-b response gene is further underscored by the recent finding of an association between GPC5 and response to treatment in an independent study (Cénit and others 2009). In the study by Comabella and colleagues (2009a), DNA pools from 106 MS patients treated with IFN-b were genotyped by means of the higher density Affymetrix 500K SNP arrays. In a second phase of the study, top-scoring SNPs arising from the arrays were genotyped in an independent cohort of 94 responders and nonresponders to IFN-b. Surprisingly, the strongest association signal corresponded to an intronic polymorphism located in the glutamate receptor ionotropic AMPA 3 (GRIA3), a gene that codes for a glutamate receptor. This finding supports the potential connection between genes that code for neurotransmitter-gated channels and response to IFN-b suggested in the previous genome-wide association study by Byun and colleagues (2008). Another interesting result that emerged from gene ontology analysis was the finding of the immune response category related with the antiviral actions of IFNs as the most represented among top scoring SNPs that best differentiated between responders and nonresponders. In fact, 2 of the genes that were validated in the replication cohort, adenosine deaminase, RNA-specific (ADAR) and IFN (alpha, beta, and omega) receptor 2 (IFNAR2), are type I IFN-responsive genes. The 3 main conclusions that could be drawn from both studies are as follows. (1) Similar to the MS genetic susceptibility, the response to IFN-b is complex and polygenic in nature. The final good or poor outcome of therapy is most likely mediated by the interaction of several genes playing different roles and related with the pleiotropic actions of IFN-b. (2) The glutamatergic system seems to be involved in the response to IFN-b. It should be taken into account that glutamate receptors are expressed in oligodendrocytes and mediate most of the excitatory synaptic transmission in the central nervous system (Seeburg 1993; Steinhäuser and Gallo 1996; Mayer and Armstrong 2004). Further, accumulation of glutamate in the extracellular space is involved in the excitotoxic injury of axons and oligodendrocytes due to overactivation of glutamate receptors (Sánchez-Gómez and Matute 1999; Pitt and others 2000; Werner and others 2001). Although in vitro studies are necessary, the implication of the glutamatergic system in the response to IFN-b may be related with yet unknown neuroprotective effects of the treatment modulating synaptic transmission and neuronal excitability. (3) Genes belonging to the type I IFN pathway through which IFN-b acts may also be implicated in treatment response. This hypothesis is supported by recent findings from a gene expression study using microarrays that points to a dysregulation of the type I IFN pathway in nonresponders to IFN-b (Comabella and others 2009c). Conclusion Despite progress made in identifying genes that may be associated with the response to IFN-b, validation of top candidates in large cohorts of treatment responders and nonresponders is still needed. However, big efforts should be made in parallel to better define the clinical and/or radiological criteria of response and treatment failure to IFN-b. Although IFN-b is one of the most commonly used treatments for MS, consensus on the definition of the response criteria has not been reached yet. The use of different criteria to define response to treatment may well result in inconsistencies in the replication of the main findings emerging from pharmacogenomics whole-genome and candidate gene association studies. The emergence of neurotransmitter-gated channels as potential class of IFN-b response-modifying genes should give a new impetus to the glutamatergic theory of neuronal degeneration in MS, while the appearance of the type I IFNinducible ADAR locus from both the Byun and others (2008) and Comabella and others (2009a) whole-genome screens may implicate a previously unanticipated role for its gene product in IFN-b-induced clinical benefits or lack thereof. While IFNAR2 was among the first genes to be studied in the context of IFN-b pharmacogenomics, its re-emergence from a genome scan (Comabella and others 2009a) reinforces the need for both pharmacogenomics and functional re-analysis. It is important to bear in mind that while the focus of this review was specifically on DNA-based polymorphic variants, it remains to be seen whether a combination of SNPs, transcriptomic, metabolomic, and/or proteomic markers used in conjunction with clinical and radiological measures may be suited better to classify patients according to response status. In this regard, emerging technologies such as next-generation DNA sequencing may also be incorporated in the design of IFN-b pharmacogenomics studies to perform

23 SNPS AND IFN-BETA TREATMENT IN MS 731 combined large-scale genetic and gene expression profiling studies. A more definitive whole-genome association screens applied to IFN-b in MS should be based upon much higher patient numbers than those included in recent studies (Byun and others 2008; Comabella and others 2009a), so as to ensure sufficient power for identification of genes with moderate effects withstanding genome-wide correction. In conclusion, significant progress has been made toward the deciphering of the pharmacogenomics of IFN-b in MS, but much more work needs to be done for these discoveries to be translated into clinical applications. While increasing numbers of disease-modifying therapies for MS are entering the market, critical evaluation of the conceptual and practical drawbacks and strengths of completed and ongoing IFN-b pharmacogenomics studies may help to define a consensual template on which to base future pharmacogenomics studies of novel MS therapies. Acknowledgments Research on the pharmacogenomics of MS treatments in the laboratories of K.V. and M.C. is supported with funding from the European Community s Seventh Framework Programme (FP7/ ) under grant agreement no (UEPHA*MS; Author Disclosure Statement No competing financial interests exist. References Baranzini SE, Mousavi P, Rio J, Caillier SJ, Stillman A, Villoslada P, Wyatt MM, Comabella M, Greller LD, Somogyi R, Montalban X, Oksenberg JR Transcription-based prediction of response to IFNbeta using supervised computational methods. PLoS Biol 3:e2. Byun E, Caillier SJ, Montalban X, Villoslada P, Fernández O, Brassat D, Comabella M, Wang J, Barcellos LF, Baranzini SE, Oksenberg JR Genome-wide pharmacogenomic analysis of the response to interferon beta therapy in multiple sclerosis. Arch Neurol 65: Cénit MD, Blanco-Kelly F, de las Heras V, Bartolomé M, de la Concha EG, Urcelay E, Arroyo R, Martínez A Glypican 5 is an interferon-beta response gene: a replication study. Mult Scler 15: Cohen JA, Cutter GR, Fischer JS, Goodman AD, Heidenreich FR, Kooijmans MF, Sandrock AW, Rudick RA, Simon JH, Simonian NA, Tsao EC, Whitaker JN; IMPACT Investigators Benefit of interferon beta-1a on MSFC progression in secondary progressive MS. Neurology 59: Comabella M, Craig DW, Morcillo-Suárez C, Río J, Navarro A, Fernández M, Martin R, Montalban X. 2009a. Genome-wide scan of 500,000 single nucleotide polymorphisms in responders and non-responders to interferon-beta in multiple sclerosis. Arch Neurol 66: Comabella M, Fernández-Arquero M, Río J, Guinea A, Fernández M, Cenit MC, de la Concha EG, Montalban X. 2009b. HLA class I and II alleles and response to treatment with interferonbeta in relapsing-remitting multiple sclerosis. J Neuroimmunol 210: Comabella M, Lünemann JD, Río J,Sánchez A, López C, Julià E, Fernández M, Nonell L, Camiña-Tato M, Deisenhammer F, Caballero E, Tortola MT, Prinz M, Montalban X, Martin R. 2009c. A type I interferon signature in monocytes is associated with poor response to interferon-beta in multiple sclerosis. Brain 132(Pt 12): Comi G, Filippi M, Barkhof F, Durelli L, Edan G, Fernández O, Hartung H, Seeldrayers P, Sørensen PS, Rovaris M, Martinelli V, Hommes OR; Early Treatment of Multiple Sclerosis Study Group Effect of early interferon treatment on the conversion to definite multiple sclerosis; a randomized study. Lancet 357: Cunningham S, Graham C, Hutchinson M, Droogan A, O Rourke K, Patterson C, McDonnell G, Hawkins S, Vandenbroeck K Pharmacogenomics of responsiveness to interferon IFN-beta treatment in multiple sclerosis: a genetic screen of 100 type I interferon-inducible genes. Clin Pharmacol Ther 78: European Study Group on Interferon Beta-1b in Secondary Progressive MS Placebo-controlled multicentre randomised trial of interferon beta-1b in treatment of secondary progressive multiple sclerosis. Lancet 352: Fernández O, Fernández V, Mayorga C, Guerrero M, León A, Tamayo JA, Alonso A, Romero F, Leyva L, Alonso A, Luque G, de Ramon E HLA class II and response to interferonbeta in multiple sclerosis. Acta Neurol Scand 112: Ge D, Fellay J, Thompson AJ, Simon JS, Shianna KV, Urban TJ, Heinzen EL, Qiu P, Bertelsen AH, Muir AJ, Sulkowski M, Hutchinson JG, Goldstein DB Genetic variation in IL28B predicts hepatitis C treatment-induced viral clearance. Nature 461: Jacobs LD, Beck RW, Simon JH, Kinkel RP, Brownscheidle CM, Murray TJ, Simonian NA, Slasor PJ, Sandrock AW Intramuscular interferon beta 1a therapy initiated during a first demyelinating vent in multiple sclerosis. CHAMPS Study Group. N Engl J Med 343: Jacobs LD, Cookfair DL, Rudick RA, Herndon RM, Richert JR, Salazar AM, Fischer JS, Goodkin DE, Granger CV, Simon JH, Alam JJ, Bartoszak DM, Bourdette DN, Braiman J, Brownscheidle CM, Coats ME, Cohan SL, Dougherty DS, Kinkel RP, Mass MK, Munschauer FE 3rd, Priore RL, Pullicino PM, Scherokman BJ, Whitham RH, and others Intramuscular interferon beta-1a for disease progression in relapsing multiple sclerosis. The Multiple Sclerosis Collaborative Research Group (MSCRG). Ann Neurol 39: Kappos L, Polman CH, Freedman MS, Edan G, Hartung HP, Miller DH, Montalban X, Barkhof F, Bauer L, Jakobs P, Pohl C, Sandbrink R Treatment with interferon beta-1b delays conversion to clinically definite and McDonald MS in patients with clinically isolated syndromes. Neurology 67: Krumbholz M, Faber H, Steinmeyer F, Hoffmann LA, Kümpfel T, Pellkofer H, Derfuss T, Ionescu C, Starck M, Hafner C, Hohlfeld R, Meinl E Interferon-beta increases BAFF levels in multiple sclerosis: implications for B cell autoimmunity. Brain 131(Pt 6): Leyva L, Fernández O, Fedetz M, Blanco E, Fernandez VE, Oliver B, Leon A, Pinto-Medel MJ, Mayorga C, Guerrero M, Luque G, Alcina A, Matesanz F IFNAR1 and IFNAR2 polymorphisms confer susceptibility to multiple sclerosis but not to interferon-beta treatment response. J Neuroimmunol 163: Lublin FD, Reingold SC Defining the clinical course of multiple sclerosis: results of an international survey. National Multiple Sclerosis Society (USA) Advisory Committee on Clinical Trials of New Agents in Multiple Sclerosis. Neurology 46: Martínez A, de las Heras V, Mas Fontao A, Bartolomé M, de la Concha EG, Urcelay E, Arroyo R An IFNG polymorphism

24 732 VANDENBROECK AND COMABELLA is associated with interferon-beta response in Spanish MS patients. J Neuroimmunol 173: Mayer ML, Armstrong N Structure and function of glutamate receptor ion channels. Annu Rev Physiol 66: Minagar A, Adamashvili I, Kelley RE, Gonzalez-Toledo E, McLarty J, Smith SJ Saliva soluble HLA as a potential marker of response to interferon-beta1a in multiple sclerosis: a preliminary study. J Neuroinflamm 4:16. O Doherty C, Favorov A, Heggarty S, Graham C, Favorova O, Ochs M, Hawkins S, Hutchinson M, Vandenbroeck K Genetic polymorphisms, their allele combinations and IFNbeta treatment response in Irish multiple sclerosis patients. Pharmacogenomics 10: O Doherty C, Villoslada P, Vandenbroeck K Pharmacogenomics of type I interferon therapy: a survey of responsemodifying genes. Cytokine Growth Factor Rev 18: Panitch H, Miller A, Paty D, Weinshenker B; North American Study Group on Interferon Beta-1b in Secondary Progressive MS Interferon beta-1b in secondary progressive MS: results from a 3-year controlled study. Neurology 63: Pitt D, Werner P, Raine CS Glutamate excitotoxicity in a model of multiple sclerosis. Nat Med 6: Prevention of Relapses and Disability by Interferon Beta-1a Subcutaneously in Multiple Sclerosis (PRISMS) Study Group Randomised double-blind placebo-controlled study of interferon beta-1a in relapsing/remitting multiple sclerosis. Lancet 352: Río J, Nos C, Tintoré M, Borrás C, Galán I, Comabella M, Montalban X Assessment of different treatment failure criteria in a cohort of relapsing-remitting multiple sclerosis patients treated with interferon beta: implications for clinical trials. Ann Neurol 52: Sánchez-Gómez MV, Matute C AMPA and kainate receptors each mediate excitotoxicity in oligodendroglial cultures. Neurobiol Dis 6: Seeburg PH The TINS/TiPS Lecture. The molecular biology of mammalian glutamate receptor channels. Trends Neurosci 16: Soilu-Hanninen M, Laaksonen M, Hanninen A, Eralinna JP, Panelius M Downregulation of VLA-4 on T cells as a marker of long term treatment response to interferon beta-1a in MS. J Neuroimmunol 167: Sospedra M, Martin R Immunology of multiple sclerosis. Annu Rev Immunol 23: Sriram U, Barcellos LF, Villoslada P, Rio J, Baranzini SE, Caillier S, Stillman A, Hauser SL, Montalban X, Oksenberg JR Pharmacogenomic analysis of interferon receptor polymorphisms in multiple sclerosis. Genes Immun 4: Steinhäuser C, Gallo V News on glutamate receptors in glial cells. Trends Neurosci 19: Suppiah V, Moldovan M, Ahlenstiel G, Berg T, Weltman M, Abate ML, Bassendine M, Spengler U, Dore GJ, Powell E, Riordan S, Sheridan D, Smedile A, Fragomeli V, Müller T, Bahlo M, Stewart GJ, Booth DR, George J IL28B is associated with response to chronic hepatitis Cinterferon-alpha and ribavirin therapy. Nat Genet 41: Tanaka Y, Nishida N, Sugiyama M, Kurosaki M, Matsuura K, Sakamoto N, Nakagawa M, Korenaga M, Hino K, Hige S, Ito Y, Mita E, Tanaka E, Mochida S, Murawaki Y, Honda M, Sakai A, Hiasa Y, Nishiguchi S, Koike A, Sakaida I, Imamura M, Ito K, Yano K, Masaki N, Sugauchi F, Izumi N, Tokunaga K, Mizokami M Genome-wide association of IL28B with response to pegylated interferon-alpha and ribavirin therapy for chronic hepatitis C. Nat Genet 41: The Interferon b Multiple Sclerosis Study Group Interferon beta-1b is effective in relapsing-remitting multiple sclerosis. I. Clinical results of a multicenter, randomized, double-blind, placebo-controlled trial. Neurology 43: Vandenbroeck K, Comabella M, Tolosa E, Goertsches R, Brassat D, Hintzen R, Infante-Duarte C, Favorov A, Escorza S, Palacios R, Oksenberg JR, Villoslada P United Europeans for development of pharmacogenomics in multiple sclerosis network. Pharmacogenomics 10: Vandenbroeck K, Matute C Pharmacogenomics of the response to IFN-b in multiple sclerosis: ramifications from the first genome-wide screen. Pharmacogenomics 9: Villoslada P, Barcellos LF, Rio J, Begovich AB, Tintore M, Sastre- Garriga J, Baranzini SE, Casquero P, Hauser SL, Montalban X, Oksenberg JR The HLA locus and multiple sclerosis in Spain. Role in disease susceptibility, clinical course and response to interferon-beta. J Neuroimmunol 130: Wandinger KP, Lunemann JD, Wengert O, Bellmann-Strobl J, Aktas O, Weber A, Grundström E, Ehrlich S, Wernecke KD, Volk HD, Zipp F TNF-related apoptosis inducing ligand (TRAIL) as a potential response marker for interferon-beta treatment in multiple sclerosis. Lancet 361: Weinstock-Guttman B, Tamano-Blanco M, Bhasi K, Zivadinov R, Ramanathan M Pharmacogenetics of MXA SNPs in interferon-b treated multiple sclerosis patients. J Neuroimmunol 182: Wergeland S, Beiske A, Nyland H, Hovdal H, Jensen D, Larsen JP, Maroy TH, Smievoll AI, Vedeler CA, Myhr KM IL- 10 promoter haplotype influence on interferon treatment response in multiple sclerosis. Eur J Neurol 12: Werner P, Pitt D, Raine CS Multiple sclerosis: altered glutamate homeostasis in lesions correlates with oligodendrocyte and axonal damage. Ann Neurol 50: Address correspondence to: Prof. Koen Vandenbroeck Neurogenomiks Group Universidad del País Vasco (UPV/EHU) Edificio 205, planta -1 Parque Tecnológico de Bizkaia Zamudio Spain k.vandenbroeck@ikerbasque.org Received 15 July 2010/Accepted 15 July 2010

25 JOURNAL OF INTERFERON & CYTOKINE RESEARCH Volume 30, Number 10, 2010 ª Mary Ann Liebert, Inc. DOI: /jir Role of Differential Expression of Interferon Receptor Isoforms on the Response of Multiple Sclerosis Patients to Therapy with Interferon Beta Francesca Gilli The cytokine interferon (IFN)-b is successfully used in the treatment of multiple sclerosis. However, some patients fail to respond to therapy, probably due to different biological patterns that are of importance in influencing clinical response. A common mechanism involved in the modulation of responsiveness to cytokine is represented by regulation of their receptor expression through autocrine-ligand-mediated loops. Mechanistically, IFN-b exerts its biological effects through interaction with the IFN-a/-b-receptor (IFNAR), which then activates several transcription factors. IFNAR is composed of 2 chains, IFNAR-1 and IFNAR-2, which associate with IFN-b to form a ternary complex. The major ligand-binding subunit is IFNAR-2 and it exists in 3 mrna splice variants, resulting in 2 transmembrane (IFNAR-2b and IFNAR-2c) isoforms and a soluble (IFNAR-2a) one. On the contrary, from normal cells only one IFNAR-1 isoform, with transcriptional capacity, was identified. In the past decades, considerable information has accumulated pertaining to the downregulation of the IFNAR complex in IFN-treated patients, but only a few studies have investigated the molecular events involved in this phenomenon. The intent of the present review is to place this receptor downregulation in the context of IFN-b therapy and of its clinical and biological outcomes in IFN-b-treated patients. Introduction Interferons (IFNs) are inducible cytokines with potent antiviral, antiproliferative, and immunomodulatory effects. On the basis of the type of receptor through which they signal, human IFNs have been classified into 3 major types: type I IFNs (IFN-a, IFN-b, and IFN-o), which bind to a specific cell surface receptor complex known as the IFN-a/-b receptor (IFNAR); type II IFN (IFN-g), which binds to the IFN-g receptor; and type III IFNs, which signal through a receptor complex consisting of IL10R2 (also called CRF2-4) and IFNLR1 (also called CRF2-12). The immune effects of type I IFNs have been exploited to treat several diseases (Borden and others 2007). Particularly, among the type I IFNs, the cytokine IFN-b is successfully used in the treatment of multiple sclerosis (MS), as has been shown to decrease clinical relapses, reduce MS-associated brain lesions as detected by MRI, and slow progression of disability (The IFNB Multiple Sclerosis Study Group 1993; Jacobs and others 1996; The PRISM Study Group 2001). Unfortunately, there is a high degree of variability in the response, and about 30% of patients fail to respond, or respond suboptimally, to the therapy (Río and others 2006). This might be due, among other clinical and biological factors, to the existence of different biological cell patterns, which are of great importance for the biological and clinical response to therapy. The responsiveness of target cells to a particular cytokine is, indeed, dependent to the specific cytokines receptor repertoire and the number of receptors expressed on the cell surface (Uings and Farrow 2000). The IFNAR Human IFN-b is a naturally occurring glycoprotein that is 166 amino acids in length with a molecular weight of 22.5 kda, whose activity is mediated by its high affinity binding to specific cellular receptor (ie, IFNAR), which is common for all type I IFNs. The IFNAR is composed, as other cytokines receptors, of multiple components, in this case designated IFNAR-1 and IFNAR-2. These 2 components associate with type I IFNs to form a ternary complex that activates multiple intracellular signaling cascades, finally leading to the synthesis of proteins that mediate antiviral, growth inhibitory, and immunomodulatory responses (Uzé and others 1995; Brierley and Fish 2002). SCDO Neurology 2 Regional Reference Centre for Multiple Sclerosis (CReSM), Neuroscience Institute of the Cavalieri Ottolenghi Foundation, University Hospital S. Luigi Gonzaga, Ottolenghi, Orbassano (Torino), Italy. 733

26 734 GILLI Structure function When IFN-b binds to its receptor, a series of tyrosine phosphorylation events is initiated, including activation of Janus-associated tyrosine kinases Jak-1 and Tyk-2, which are found associated with the IFNAR complex and are shown to be necessary for the subsequent biological effects of the cytokine (Uzé and others 1995, 2007; Brierley and Fish 2002; Platanias 2005). The subunits of IFNAR make distinct contributions to ligand binding. IFNAR-1 has low but varied intrinsic affinity for the various type I IFNs, whereas IFNAR-2 shows moderate to high (nanomolar) affinity (Cutrone and Langer 2001; Cajean-Feroldi and others 2004; de Weerd and others 2007; Jaks and others 2007; Pan and others 2008). Receptor binding appears to be a sequential process, with IFN-b first binding to the higher affinity IFNAR-2, followed by recruitment of IFNAR-1, to form the ternary complex, with consequent receptor and cellular activation (Lamken and others 2004; Gavutis and others 2005; Jaitin and others 2006). Thus, while IFNAR-2 plays the major role in affinity determination and differential recognition of type I IFNs, IFNAR-1 modulates both the ligand affinity and selectivity of the IFNAR-1/IFNAR-2 receptor complex, leading to enhanced complex stability and binding affinity (Lamken and others 2004; Jaitin and others 2006). IFNAR-1 is also functionally involved in signal transduction because of its association with the cytoplasmic Janus family kinases, Jak-1 and Tyk-2 (de Weerd and others 2007): IFNAR-1 is preassociated with Tyk-2 (Yan and others 1996), which positively influences ligand binding to the receptor complex, and also stabilizes IFNAR-1 cell surface expression levels (Marijanovic and others 2007). On the other hand, Jak-1 associates with the IFNAR-2 and, after receptor engagement with type I IFNs, phosphorylates 2 members of the signal transducers and activators of transcription family, ie, STAT-1 and STAT-2 (Fig. 1). As a result, an IFNstimulated gene factor 3 (ISGF3) complex forms (this contains STAT-1, STAT-2, and a third transcription factor called IFN response factor 9) and moves into the cell nucleus (Fig. 1). After nuclear import, the ISGF3 complex binds with high affinity to specific nucleotide sequences called IFN-stimulated response elements (ISRE) in the promoters of certain genes, inducing the transcription of those genes (Brierley and Fish 2002). In detail, binding of ISGF3 to ISRE mediates the activation of a specific group of genes called IFN-stimulated genes (ISGs), including the double-stranded RNA-activated protein kinase, oligoadenylate synthetase, myxovirus resistant protein A (MxA), and guanylate-binding protein. Type I IFNs also induce the formation of other STATcontaining complexes. These heterodimers bind palindromic sequences designated gamma-activated sequences, located in the promoters of different ISGs. Hence, type I IFNs can induce expression of genes with either ISRE or gamma-activated sequence elements (Brierley and Fish 2002). In addition to the Jak-STAT pathway, type I IFNs can activate several other signaling cascades (Brierley and Fish 2002). For example, type I IFNs activate a member of the CRK family of adaptor proteins called CRKL, a nuclear adaptor for STAT-5 that also regulates signaling through the C3G/Rap1 pathway, the p38 mitogen-activated protein kinase, and the phosphatidylinositol 3-kinase signaling pathway. IFNAR-2c IFNb IFNAR-1 Cell Membrane Classical IFN signaling JAK1 Tyk2 P P STAT2 P P IRF9 STAT1 ISGF3 IRF9 Nucleus ISRE ISGs FIG. 1. Receptor complexes and IFN signaling. Conventional signaling occurs when IFN-b binds to the full-length transmembrane IFNAR-2c and IFNAR-1, resulting in cross-phosphorylation of receptors and associated Janus kinases (Tyk-2 and Jak-1). This provides docking sites on the receptor complex for STAT proteins (most likely STAT-1 and STAT-2, which combine with IRF9 to form the ISGF3 complex). STAT proteins are in turn phosphorylated and form homo- and heterodimeric complexes, which dissociate from the receptor and then translocate to the nucleus and bind to an ISRE element within the promoters of ISGs, leading to their transcription. IFN, interferon; IFNAR, IFN-a/-b receptor; IRF9, IFN response factor 9; ISGs, IFN-stimulated genes; ISRE, IFN-stimulated response elements; ISGF3, IFN-stimulated gene factor 3.

27 REGULATING IFNAR ISOFORMS 735 IFNAR isoforms IFNAR genes encode multiple isoforms that contribute to the potential complexity of the functional receptor (Novick and others 1994; Uzé and others 2007). Two splice variants of IFNAR-1 were identified in different cell lines (Abramovich and others 1994; Cook and others 1996). However, subsequent bioinformatic analyses of splice variants in expressed sequence tag (EST) databases and rapid amplification of cdna ends (3 0 RACE) analyses from normal cells identified only one isoform, suggesting that the former are either artefacts or aberrant transcripts found only in particular tumor cell lines (de Weerd and others 2007). In contrast, 4 IFNAR-2 transcripts encoding 3 isoforms are generated from the same gene by exon skipping, alternative splicing, and differential usage of polyadenylation sites (Fig. 2) (Lutfalla and others 1995). These transcripts encode a long transmembrane form (IFNAR-2c), a short transmembrane form (IFNAR-2b), and a soluble form (sifnar-2a) (Fig. 2). Among the 3 IFNAR-2 isoforms, the full-length form IF- NAR-2c is recognized as the functional one (Lutfalla and others 1995), whereas the short form IFNAR-2b shows a truncated cytoplasmatic tail and is therefore unable to perform signal transduction (Domanski and Colamonici 1996). Consistent with the latter observation, transfection of human IFNAR-1 and IFNAR-2c, but not IFNAR-2b, reconstitutes the antiviral IFN activity in NIH 3T3 cells (Cohen and others 1995). On the other hand, IFNAR-2b may acts as a dominant negative regulator of IFN responses (Gazziola and others 2005). The third isoform, sifnar-2a, lacks the transmembrane and intracytoplasmic domain and is regarded as a soluble receptor subunit that can be found in different body fluids (Novick and others 1994; Hardy and others 2001). This isoform is still capable of IFNs binding, and therefore it may represent an important regulatory factor of type I IFNs bioactivity (McKenna and others 2004; Gilli and others 2007). Soluble IFNAR-2a receptors can act either as agonist or antagonist based on the concentration fluctuations: sifnar-2a can neutralize the bioactivity of type I IFNs at high concentrations, and at lower concentrations, it causes an enhancement of type I IFN-mediated bioactivity (McKenna and others 2004). Soluble IFNAR-2a receptors bind type I IFNs with an affinity similar to the transmembrane IFNAR-2c isoform, thereby prolonging the IFN half-life (McKenna and others 2004). Further, sifnar-2a receptors can mediate type I IFNs biological effects by a mechanism known as trans-signaling. In general, trans-signaling occurs when a soluble receptorbound ligand interacts with a complementary transmembrane receptor chain of the receptor complex ( Jones and others 2005; Rubinstein and others 2006). With respect to the IFN-a/-b system, it has been demonstrated in vitro that sifnar-2a can bind either IFN-a or IFN-b and transduce a signal through IFNAR-1 (Hardy and others 2001). Further experiments using mice overexpressing sifnar-2a suggest that high levels of sifnar-2a may act as trans-signaling molecules in vivo (Platanias 2005; de Weerd and others 2007). Importantly, these data demonstrate that a signal can be transduced in the absence of the intracellular domain of IF- NAR-2, possibly through IFNAR-1. The mechanism for this remains to be elucidated, but could involve ligand-mediated dimerization of IFNAR-1, or recruitment of another, unknown component of the receptor complex. It remains, however, to explain why a soluble IFNAR-2a is able to trans-signal, whereas a truncated (membrane-associated) 0 1 kb 2 kb Stop IFNR-2a Alu AAA Stop IFNR-2b Alu AAA Stop IFNR-2c Alu AAA FIG. 2. Structure of the transcripts encoded by the IFNAR-2 gene. The IFNAR-2 gene is alternatively spliced to generate transcripts encoding a long transmembrane isoform (IFNAR-2c), a short transmembrane isoform (IFNAR-2b), and soluble (sifnar2a) isoform by exon skipping. Soluble IFNAR-2a is generated by splicing at exon 7 into splice acceptor site within exon 9. The long IFNAR-2c uses exons 7 and 8. Short IFNAR-2b uses exons 7 and 8. The leader peptide and transmembrane domain are boxed in black. Stop codons are indicated by vertical bars. The gray box indicates the presence of Alu sequences in the 3 0 non coding region.

28 736 GILLI IFNAR-2b is unable to do it. One explanation may lie in the different dimer subunit interactions. Clearly, IFNAR-2b can bind ligand and form a complex with IFNAR-1 (Domanski and Colamonici 1996; Hardy and others 2001), but exactly the IFNAR-2b/IFNAR-1 dimer formation, in turn, could block the ligand-mediated dimerization of IFNAR-1, thus preventing the generation of the activation signal. On the other hand, specific experiments aimed at testing whether sifnar-2a may block the biologic actions of type I IFNs have shown an antagonist effect of the activity of the ligand. In murine L929 cells, sifnar-2a inhibited the induction of an IFN-sensitive reporter in a dose-dependent manner, and sifnar-2a was an equally potent inhibitor of both IFN-a and IFN-b (Hardy and others 2001). These latter observations in mice are in line with a study analyzing the impact of circulating sifnar-2a on exogenously administered IFN-b in humans (Gilli and others 2007). In this study, it has been demonstrated that an sifnar-2a-mediated neutralization develops in 4% of patients chronically treated with IFN-b. Notably, such sifnar-2a-mediated neutralization is observed mainly in patients treated with high-dose regimens of the cytokine, suggesting that the phenomenon might be related to the frequency of administration of the drug. Since several cytokine receptors (eg, TNFR, IL4R, and IL6R) were shown to be released as a feedback regulation mechanism from the cell surface upon exposure to either their cognate ligands or agonists (Kiessling and Gordon 1998), it can be hypothesized that a similar change induced by the high frequency of administration may also involve the IFNAR. Regulation of the IFNAR Complex During IFN-b Therapy A common mechanism involved in the modulation of cytokine and chemokine responsiveness is represented by the regulation of their receptor expression through autocrineligand-mediated loops. With respect to the IFN-a/-b system, it has been shown that IFNAR complex is internalized and degraded after ligand binding (Branca and others 1982). In addition to receptor internalization, upon prolonged IFN-a exposure, IFNAR also undergo desensitization and downregulation; this phenomenon of ligand-induced downregulation of IF- NAR has been observed in Daudi (lymphoblastoid), WISH (human amnion), and other cultured cell lines (Lau and others 1986; Cleveland and others 1987; Zoon and others 1989). Concurrently, desensitization of IFNAR has been also described in a human cell line, after treatment with IFN-g (Hannigan and others 1984). The latter clearly results from an indirect mechanism since IFN-a and IFN-g do not share the same receptor sites. On the basis of these observations, it is argued that IFN-b can modulate expression of its own receptor in target cells, as well as both IFN-a and IFN-g. The regulation of IFNAR by IFNs themselves has important implications for the clinical use of these cytokines. Indeed, in vivo studies reporting on patients receiving chronic IFN-a therapy reveal a similar phenomenon of reduction in IFNAR expression. Concurrent with the progression of therapy, the IFN-a binding activities in the patients peripheral blood mononuclear cells were shown to be progressively reduced. This reduction in binding most likely results from IFN-induced downregulation of IFNAR expression, in vivo during IFN-a therapy (Lau and others 1986). From a therapeutic perspective, it has been also observed that a lack of IFNAR downregulation correlates with failure of therapy in non-hodgkin s lymphomas, suggesting that in those patients there is not enough receptor stimulation (Billard and others 1991). Likewise, examination of surface expression of IFNAR-1 and IFNAR-2c subunits on CD34- positive cells in bone marrow from chronic myelogenous leukemia patients showed that IFNAR-2c downregulation during IFN-a therapy is commonly observed in good responders, whereas correlation between IFNAR-1 downregulation and clinical response is not evident at all (Ito and others 2004). A well-known phenomenon linked to receptor stimulation is tachyphylaxis, a reduction in biological responsiveness resulting from the continued occupation of a particular receptor by an agonistic ligand (Sherwood 1997). There is evidence of tachyphylaxis during IFN-b therapy both from the clinical (with respect to most side-effects) and biological perspectives (Sandberg-Wollheim and others 2005; Karussis and others 2006). For example, it has been reported that chronic and prolonged (3 months) treatment with IFN-b in MS patients significantly decrease expression of the MxA gene, an ISG that is commonly used as a biomarker of IFN-b bioactivity (Vallittu and others 2002, 2006; Gilli and others 2004). Since the most important mechanism contributing to the reduced biological efficacy observed with prolonged treatment (ie, tachyphylaxis) appears to be a reduction in receptor density, it was proposed that there is a significant downregulation in IFN-b-treated MS patients similar to that previously observed in IFN-a-treated patients. On these grounds, in the last years a number of reports have addressed the biological and clinical significance of the regulation of the IFNAR components in patients with MS, chronically treated with the cytokine; all the studies suggest both IFNAR-1 and IFNAR-2 as possible contributors to IFNb treatment response in MS (Owczarek and others 1997; Oliver and others 2007; Gilli and others 2008). Particularly, during treatment with IFN-b, there is a downregulation of the IFNAR complex and, according to the hypothesis that a receptor downregulation could predict for a major clinical response to treatment, decreased IFNAR expression is mainly observed in patients who are clinically responders to IFNb therapy (Oliver and others 2007). Expression of the IFNAR complex may thus represent an important mechanism of regulation of cell responsiveness to IFN-b. Clearly, at this state of knowledge, further studies are necessary to clarify whether the response to treatment is influenced by changes in the level of the IFNAR depending on the pathological alteration of gene expression, or on its ligand receptor interaction (affinity, specificity, etc.). In this respect, the study of expression of the different IFNAR isoforms appears to be extremely useful. Most of the previous studies have focused on a gene expression analysis performed by using primer sequences present in all isoforms, thus preventing the authors from discriminating which was the consistently regulated IFNAR isoform during therapy (Oliver and others 2007). More recently, a different molecular approach has been used, allowing the comparison of the level of expression of any subunit/isoform between different groups of IFN-treated patients (Ito and others 2004; Gilli and others 2008). Thanks to this latter approach, the IFNAR isoforms analysis made it

29 REGULATING IFNAR ISOFORMS 737 clear that biologic response to type I IFNs may vary considerably between patients who express different levels of the IFNAR on the cell surface (Ito and others 2004; Gilli and others 2008). Consistently, high levels of expression of the full-length (and thus functionally active) IFNAR-2c isoform were correlated with greater biologic response to type IFNs, whereas such correlation was not true for both the other IFNAR-2 isoforms (ie, IFNAR-2b and sifnar-2a) (Gilli and others 2008). The correlation observed between the levels of expression of full-length IFNAR-2c and the induction responsiveness to type I IFNs supports the hypothesis that the biologic response of a tissue type to a specific ligand is dependent on the receptor expression profile in that tissue. In particular, the binding capacity (ie, the IFNAR-2c subunit as having a major role in binding type I IFN ligand, but requiring the cooperation of IFNAR-1 for signal transduction) (Owczarek and others 1997) of the receptor rather than the solely transcriptional capacity (ie, the IFNAR-1 subunit) seems to be important for the modulation of biologic responsiveness. Similar results have been obtained in certain other studies that have evaluated the biological response to other cytokines (eg, IL-12 and IL-10) (Presky and others 1996; Mathurin and others 2002) and growth factors (eg, epidermal growth factor) (Sundaresan and others 1997). Along with the observation of the prominent role played by the IFNAR-2c component, it has been found that among the 3 IFNAR-2 isoforms only IFNAR-2c expression is downregulated in long-term IFN-b-treated patients. Notably, such receptor downregulation was concurrent with an attenuation of the MxA gene upregulation being, therefore, a likely reason for it. Once again, the phenomena of tachyphylaxis and receptor downregulation are strictly linked. Interestingly, a slight increase over time in soluble IFNAR-2a expression was also observed. As already stated above, recent studies showed that sifnar-2a has both agonistic and antagonistic proprieties on type I IFNs activity (Hardy and others 2001; Gilli and others 2007), suggesting that this isoform could be a potential key regulator for the actions of type I IFNs. Because of the kinetics of IFNAR-2 induction, it is reasonable to suggest that the induction of sifnar-2a acts as a classic negative feedback loop, whereby IFN-b induces expression of an inhibitor of itself (Hardy and others 2001; Gilli and others 2007, 2008). Unlike IFNAR-2, IFNAR-1 expression only shows a slight downregulation along the treatment period, highlighting that expression of the IFNAR-2c might be more highly regulated by type I IFNs than expression of IFNAR-1. This hypothesis agrees with the results of a study where a differential regulation of expression of IFNAR-1 and IFNAR-2 subunits was shown in dendritic cells during their maturation (Gauzzi and others 2002). On the other hand, this hypothesis seems to partly contradict earlier work where IFNAR-1 was shown to play an important role for IFN-b sensitivity, as well as IFNAR-2 (Colamonici and others 1994; Dupont and others 2002). However, it is noteworthy that IFNAR-1 expression seems to be regulated by posttranscriptional rather than transcriptional mechanisms, as observed for IFNAR-2 (Eantuzzi and others 1997). Therefore, mrna measurements, as used in the above-mentioned studies (Oliver and others 2007; Gilli and others 2008), may prevent a proper evaluation of the IFNAR-1 regulation. Another possible explanation for the discrepancy between these studies may be that the IFNAR-1 component is expressed in various heterologous cell backgrounds that may process IFNAR-1 differently. To define the function of the IFNAR components clearly, it is an advantage to study the receptor function in a homologous background and without interference from other heterologous receptor components. In this regard, because IFNs play an important role in T- lymphocyte homeostasis, Durelli and others (2009) have recently studied the sensitivity of T-lymphocyte subsets to IFN-b. In T cell receptor-stimulated peripheral blood mononuclear cells (PBMCs) from MS patients, IFN-b was shown to decrease the Th17 cell number, without affecting the number of Th1 cells. A subsequent analysis of the level of IFNAR-1 on the cell surface of these T cell subsets showed that Th17 cells express a greater level of IFNAR-1 than Th1 cells, and this expression correlates with the higher sensitivity of Th17 cells to IFN-b. Since IFNAR-1 is needed for type I signal transduction, authors hypothesize that the levels of IFNAR-1 on Th17 cell surface might account for their higher sensitivity to IFN-b (Durelli and others 2009). Evidence of the different contribution given by distinct cell subsets to IFNAR expression is also derived from a recent study showing significantly elevated surface expression of IFNAR-1 on monocytes of nonresponder MS patients, as well as a lack of desensitization to further IFN-b stimulation (Comabella and others 2009). As a whole, all but one (Serana and others 2008) of the studies (Dupont and others 2002; Gauzzi and others 2002; Oliver and others 2007; Gilli and others 2008; Comabella and others 2009) shows a significant decrease in IFNAR expression upon chronic stimulation with IFN-b. The exception derives from a study that shows an increased IFNAR-1 expression in patients under long-term IFN-b treatment and in whom IFN-b was certainly bioactive (as witnessed by increased MxA mrna levels) (Serana and others 2008). Being aware that chronic cell stimulation by IFN should presumably induce IFNAR downregulation, Serana and others (2008) hypothesized that the increase of IFNAR-1 gene expression might serve as a mechanism for counterbalancing the receptor loss. The net effect of mrna increase might thus be a restoration of IFNAR surface protein levels and, consequently, cell responsiveness to IFN-b. However, there remains the fact that this is the only study showing a therapy-induced upregulation of the IFNAR-1 subunit. Regulation of the IFNAR Isoforms and Anti-IFN-b Antibodies Besides receptor downregulation, prolonged therapy with IFN-b often leads to the development of anti-ifn-b antibodies [binding antibodies (BABs)]. A subset of the BABs is of a neutralizing nature [neutralizing antibodies (NABs)] and has been associated with reduced biological (Deisenhammer and others 1999; Bertolotto and others 2003) and clinical efficacy of therapy (Malucchi and others 2004; Francis and others 2005; Kappos and others 2005). Because NABs block IFNAR stimulation (Sorensen and others 2005), it is argued that in NAB-positive patients there is a lack of regulation of IFNAR expression, as a consequence of the almost total lack of receptor stimulation. Considering long-term treated patients without anti-ifn-b antibodies (both BABs and NABs), ie, patients with a prolonged and fully active stimulation of the IFNAR, IFNAR-2c expression is shown to be downregulated over time, whereas IFNAR-1 expression only shows a slight downregulation

30 738 GILLI along the treatment period. Patients also show a concurrent increase in sifnar-2a expression. On the contrary, a different modulation is observed in NAB-positive patients, in whom the presence of NABs reverse both those effects, ie, increases full-length IFNAR-2c isoform without modifying both IFNAR-1 and sifnar-2a expression. NABs interfere with the interaction between IFNb and its receptor, which in turn blocks downstream IFN-b signaling and expression of ISGs products (Deisenhammer and others 1999; Bertolotto and others 2003; Gilli and others 2004; 2008). In NAB-positive patients, there is no stimulation of the receptor, which in turn increases its binding capacity. Thus, for the formation of an effective IFN-mediated response, the antagonistic effects of NABs need to be neutralized by a concomitant increase in full-length IFNAR-2c isoform expression. These results from the analysis performed by separating NAB-negative and NAB-positive patients confirm the pivotal role of the full-length IFNAR-2c subunit (ie, the binding capacity of the receptor) in regulating cell responsiveness. A second interesting result from the IFNAR isoforms analysis is that pretreatment expression levels of the fulllength IFNAR-2c isoform in MS patients are overall lower than in normal individuals (Oliver and others 2007; Gilli and others 2008). Nevertheless, as the expression levels in MS patients are rather widely distributed, it might be more accurate to say that there is a fraction of patients whose peripheral blood mononuclear cells have lower IFNAR-2c expression levels than controls or other patients with MS. This statement seems to be confirmed by the fact that initial low IFNAR-2c expression was found in a group of patients showing a significantly higher risk of developing NABs (Gilli and others 2008), also suggesting that IFN-b immunogenicity may be, at least in part, related to IFNAR-2 expression. A possible explanation for the higher risk of developing NABs observed in patients with a lower pretreatment IF- NAR-2c expression might be the longer circulation time of the injected IFN-b molecules. Lower expression of the IF- NAR-2c subunit on the cell surface could significantly decrease the binding of the IFN-b molecules, increasing their circulating time. This longer circulation time could then lead to an increase of immunogenicity, particularly in patients treated with high-dose and high-frequency regimens, in whom there are greater circulating concentrations of IFN-b. Another fascinating explanation might be the existence of a negative feedback acting through the production of specific auto-antibodies. In patients with a low pretreatment IFNAR- 2c expression, cells could be physiologically unable to respond to high concentrations of IFN-b. As a consequence, the immune system could mount a beneficial auto-antibody response to IFN-b, to counteracting, to a certain extent, the hyperstimulation of the receptor. This natural counteraction is illustrated in animal models of autoimmunity (Youssef and others 2000; Salomon and others 2002; Wildbaum and others 2002), and evidence is provided that it occurs in humans too (Wildbaum and others 2003; Meager and others 2006). In both instances, findings indicate that the regulation of IFNAR-2 expression is an important way of modulating the responsiveness to endogenous and systemically administered IFN-b. IFNAR-2 isoforms show a dual action, agonistic and antagonistic, that influences both the magnitude and the nature of the biologic response to IFN-b. Importantly, these data suggest that the levels of IFNAR-2 are regulated with the aim of keeping the body in a state of equilibrium, even when nonphysiologic stimuli are present. Nevertheless, it is noteworthy that both host-related and product-related factors have impact on immunogenicity, because part of the anti-ifn-b antibodies are surely due to the foreign nature of the drug, ie, either impurities or aggregates that break B-cell tolerance (Schellekens 2002). This statement is consistent with the previous observation that by improved recombinant IFN-a/-b formulations, immunogenicity decreases, but does not disappear completely ( Jacobs and others 1996; Giovannoni and others 2007; McKeage and Wagstaff 2007). Conclusions Tachyphylaxis is a term describing a decrease in the biological response to a drug after repeated doses over a short period. This specific phenomenon reflects the actions of the cells to maintain homeostasis, a constant degree of cell activity in spite of major changes in receptor stimulation. The most important mechanism contributing to the reduced biological efficacy observed with prolonged treatment appears to be a reduction in receptor density. Consistently, several studies have demonstrated that IFNAR is rapidly downregulated by chronically administered type I IFN therapy. In particular, the binding capacity (ie, IFNAR-2c, having a major role in binding type I IFN ligand) (Owczarek and others 1997) of the receptor rather than the solely transcriptional capacity (ie, IFNAR-1) seems to be important for the modulation of biological responsiveness in treated patients. Overall data indicate that the regulation of IFNAR-2 expression is an important way of modulating the responsiveness to endogenous and systemically administered IFNb. The 3 IFNAR-2 isoforms show a dual action, agonistic and antagonistic, that influences both the magnitude and nature of the biological response to IFN-b. Importantly, data suggest that the levels of all IFNAR-2 isoforms are concurrently regulated with the aim of keeping the body in a state of equilibrium, even when nonphysiological stimuli are present. On the contrary, IFNAR-1, which is differently processed in the various cell backgrounds and is not modulated during IFN-b therapy, seems to be more important for the early prediction of responsiveness to the cytokine. Higher IFNAR- 1 levels (particularly in pathogenic cells such as monocytes and Th17 cells) were indeed correlated to poor response to IFN-b therapy, as a cause of hyperactivity of the type I IFNs pathway. Author Disclosure Statement No competing financial interests exist. References Abramovich C, Ratovitski E, Lundgren E, Revel M Identification of mrnas encoding two different soluble forms of the human interferon-alpha receptor. FEBS Lett 338(3): Bertolotto A, Gilli F, Sala A, Capobianco M, Malucchi S, Milano E, Melis F, Marnetto F, Lindberg RLP, Bottero R, Di Sapio A, Giordana MT Persistent neutralizing antibodies abolish the interferon beta bioavailability in MS patients. Neurology 60:

31 REGULATING IFNAR ISOFORMS 739 Billard C, Ferbus D, Diez RA, Kolb JP, Mathiot C, Belanger C, Auzanneau G, Varet B, Falcoff E, Dumont J Correlation between the biological and therapeutic effects of interferonalpha in low-grade nodular non-hodgkin s lymphoma: lack of in vivo down-regulation and reduced affinity of IFN-alpha receptors in unresponsive patients. Leuk Res 15: Borden, EC, Sen GC, Uzè G, Silverman RH, Ransohoff RM, Foster GR, Stark GR Interferons at age 50: past, current and future impact on biomedicine [Review]. Nat Rev Drug Discov 6(12): Branca AA, Faltynek CR, D Alessandro SB, Baglioni C Interaction of interferon with cellular receptors: internalization and degradation of cell-bound interferon. J Biol Chem 257: Brierley MM, Fish EN IFNa/b receptor interactions to biologic outcomes: understanding the circuitry. J Interferon Cytokine Res 22: Cajean-Feroldi C, Nosal F, Nardeux PC, Gallet X, Guymarho J, Baychelier F, Sempe P, Tovey MG, Escary JL, Eid P Identification of residues of the IFNAR1 chain of the type I human interferon receptor critical for ligand binding and biological activity. Biochemistry 43: Cleveland MG, Lane RG, Klimpel GRR Enhanced IFN-a/ b and defective IFN-b production in chronic GVHD: a potential mechanism for immunosuppression. Cell Immunol 110: Cohen B, Novick D, Barak S, Rubinstein M Ligandinduced association of the type I interferon receptor components. Mol Cell Biol 15(8): Colamonici O, Yan H, Domanski P Direct binding to and tyrosine phosphorylation of the alpha subunit of the type I interferon receptor by p135tyk2 tyrosine kinase. Mol Cell Biol 14: Comabella M, Lünemann JD, Rio J, Sànchez A, Lopez C, Julia E, Fernandez M, Nonell L L, Camina-Tato M, Deisenhammer F, Caballero E, Tortola MT, Prinz M, Montalban X, Martin R A type I interferon signature in monocytes is associated with poor response to interferon-beta in multiple sclerosis. Brain 132(Pt12): Cook JR, Cleary CM, Mariano TM, Izotova L, Pestka S Differential responsiveness of a splice variant of the human type I interferon receptor to interferons. J Biol Chem 271(23): Cutrone EC, Langer JA Identification of critical residues in bovine IFNAR-1 responsible for interferon binding. J Biol Chem 276: Deisenhammer F, Reindel M, Harvey J, Gasse T, Dilitz E, Berger T Bioavailability of interferon-beta 1b in MS patients with and without neutralizing antibodies. Neurology 52: de Weerd NA, Samarajiwa SA, Hertzog PJ Type I interferon receptors: biochemistry and biological functions. J Biol Chem 282: Domanski P, Colamonici OR The type-i interferon receptor. The long and short of it [Review]. Cytokine Growth Factor Rev 7(2): Dupont SA, Goelz S, Goyal J, Green M Mechanisms for regulation of cellular responsiveness to human IFNb 1a. J Interferon Cytokine Res 22: Durelli L, Conti L, Clerico M, Borselli D, Contessa G, Ribellino P, Ferrero B, Eid P, Novelli F T-helper 17 cells expand in multiple sclerosis and are inhibited by interferon-b. Ann Neurol 65(5): Eantuzzi L, Eid P, Malorni W, Rainaldi G, Gauzzi MC, Pellegrini S, Belardelli F, Gessani S Post-translational up-regulation of the cell surface-associated component of the human type I interferon receptor during differentiation of peripheral blood monocytes: role in the biological response to type I interferon. Eur J Immunol 27: Francis GS, Rice GP, Alsop JC, PRISMS Study Group Interferon beta-1a in MS: results following development of neutralizing antibodies in PRISMS. Neurology 65: Gauzzi MG, Canini I, Eid P, Belardelli F, Gessani S Loss of type I IFN receptors and impaired IFN responsiveness during terminal maturation of monocyte-derived human dendritic cells. J Immunol 169: Gavutis M, Lata S, Lamken P, Muller P, Piehler J Lateral ligand-receptor interactions on membranes probed by simultaneous fluorescence-interference detection. Biophys J 88: Gazziola C, Cordani N, Carta S, De Lorenzo E, Colombatti A, Perris R The relative endogenous expression levels of the IFNAR2 isoforms influence the cytostatic and proapoptotic effect of IFN-alpha on pleomorphic sarcoma cells. Int J Oncol 26(1): Gilli F, Bertolotto A, Sala A, Hoffmann F, Capobianco M, Malucchi S, Glass T, Kappos L, Lindberg RLP, Leppert D Neutralizing antibodies against IFNb in multiple sclerosis: antagonization of IFNb mediated suppression of MMPs. Brain 127: Gilli F, Marnetto F, Caldano M, Valentino P, Granieri L, Di Sapio A, Capobianco M, Sala A, Malucchi S, Kappos L, Lindberg RLP, Bertolotto A Anti-interferon-b neutralising activity is not entirely mediated by antibodies. J Neuroimmunol 92: Gilli F, Valentino P, Caldano M, Granieri L, Capobianco M, Malucchi S, Sala A, Marnetto F, Bertolotto A Expression and regulation of IFNa/b receptor in IFNb-treated patients with multiple sclerosis. Neurology 71(24): Giovannoni G, Barbarash O, Casset-Semanaz F, Jaber A, King J, Metz L, Pardo G, Simsarian J, Sorensen PS, Stubinski B, the RNF Study Group Immunogenicity and tolerability of an investigational formulation of interferon-beta 1a: 24- and 48-week interim analyses of a 2-year, single-arm, historically controlled, phase IIIb study in adults with multiple sclerosis. Clin Ther 29: Hannigan GE, Fish EN, Williams BRG Modulation of human interferon-g receptor expression by human interferong. J Biol Chem 259: Hardy MP, Owczarek CM, Trajanovska S, Liu X, Kola I, Hertzog PJ The soluble murine type I interferon receptor Ifnar-2 is present in serum, is independently regulated, and has both agonistic and antagonistic properties. Blood 97(2): The IFNB Multiple Sclerosis Study Group Interferon beta- 1b is effective in relapsing-remitting multiple sclerosis. I. Clinical results of a multicenter, randomized, double-blind, placebo-controlled trial. Neurology 43: Ito K, Tanaka H, Ito T, Sultana TA, Kyo T, Imanaka F, Ohmoto Y, Kimura A Initial expression of interferon alpha receptor 2 (IFNAR2) on CD34-positive cells and its down-regulation correlate with clinical response to interferon therapy in chronic myelogenous leukemia. Eur J Haematol 73(3): Jacobs LD, Cookfair DL, Rudick RA, Herndon RM, Richert JR, Salazar AM, Fischer JS, Goodkin DE, Granger CV, Simon JH, Alam JJ, Bartoszak DM, Bourdette DN, Braiman J, Brownscheidle CM, Coats ME, Cohan SL, Dougherty DS, Kinkel RP, Mass MK, Munschauer FE 3 rd, Priore RL, Pullicino PM, Scherokman BJ, Whitham RH, et al Intramuscular interferon beta-1a for disease progression in relapsing multiple sclerosis. Ann Neurol 39:

32 740 GILLI Jaitin DA, Roisman, LC, Jaks E, Gavutis M, Piehler J, Van der, HJ, Uzè G, Schreiber G Inquiring into the differential action of interferons (IFNs): an IFN-alpha2 mutant with enhanced affinity to IFNAR1 is functionally similar to IFNbeta. Mol Cell Biol 26: Jaks E, Gavutis M, Uze G, Martal J, Piehler J Differential receptor subunit affinities of type I interferons govern differential signal activation. J Mol Biol 366: Jones SA, Richards PJ, Scheller J, Rose-John S IL-6 transsignalling: the in vivo consequences. J Interferon Cytokine Res 25(5): Kappos L, Clanet M, Sandberg-Wollheim M, Radue EW, Hartung HP, Hohlfeld R, Xu J, Bennet D, Sandrock A, Goelz S, the European Interferon Beta-1a IM Dose-Comparison Study Investigators Neutralising antibodies and efficacy of interferon beta-1a: a 4-year controlled study. Neurology 65: Karussis D, Biermann LD, Bohlega S, Boiko A, Chofflon M, Fazekas F, Freedman M, Gebeily S, Gouider R, Havrdova E, Jakab G, Karabudak R, Miller A, the International Working Group for Treatment Optimization in MS A recommended treatment algorithm in relapsing multiple sclerosis: report of an international consensus meeting [Review]. Eur J Neurol 13(1): Kiessling LL, Gordon EJ Transforming the cell surface through proteolysis [Review]. Chem Biol 5(3):R49 R62. Lamken P, Lata S, Gavutis M, Piehler J Ligand-induced assembling of the type I interferon receptor on supported lipid bilayers. J Mol Biol 341: Lau AS, Hannigan GE, Freedman MH, Williamo BRG Regulation of interferon receptor expression in human blood lymphocytes in vitro and during interferon therapy. J Clin Invest 77: Lutfalla G, Holland SJ, Cinato E, Monneron D, Reboul J, Rogers NC, Smith JM, Stark GR, Gardiner K, Mogensen KE, Kerr IM, Uzè G Mutant U5A cells are complemented by an interferon-alpha beta receptor subunit generated by alternative processing of a new member of a cytokine receptor gene cluster. EMBO J 14(20): Malucchi S, Sala A, Gilli F, Bottero R, Di Sapio A, Capobianco M, Bertolotto A Neutralizing antibodies reduce the efficacy of betaifn during treatment of multiple sclerosis. Neurology 62: Marijanovic Z, Ragimbeau J, van der Heyden J, Uzé G, Pellegrini S Comparable potency of IFN-alpha2 and IFN-beta on immediate JAK/STAT activation but differential downregulation of IFNAR2. Biochem J 407(1): Mathurin P, Xiong S, Kharbanda KK, Veal N, Miyahara T, Motomura K, Rippe RA, Bachem MG, Tsukamoto H IL- 10 receptor and coreceptor expression in quiescent and activated hepatic stellate cells. Am J Physiol Gastrointest Liver Physiol 282:G981 G990. McKeage K, Wagstaff AJ Subcutaneous interferon-beta-1a: new formulation. CNS Drugs 21: McKenna SD, Vergilis K, Arulanandam AR, Weiser WY, Nabioullin R, Tepper MA Formation of human IFN-beta complex with the soluble type I interferon receptor IFNAR-2 leads to enhanced IFN stability, pharmacokinetics, and antitumor activity in xenografted SCID mice. J Interferon Cytokine Res 24(2): Perheentupa J, Husebye E, Kadota Y, Willcox N Antiinterferon autoantibodies in autoimmune polyendocrinopathy syndrome type 1. PLoS Med 3:e289. Novick D, Cohen B, Rubinstein M The human interferon alpha/beta receptor: characterization and molecular cloning. Cell 77(3): Oliver B, Mayorga C, Fernandez V, Leyva L, Leon A, Luque G, Lopez JC, Tamayo JA, Pinto-Medel MJ, de Ramon E, Blanco E, Alonso A, Fernandez O Interferon receptor expression in multiple sclerosis patients. J Neuroimmunol 183: Owczarek CM, Hwang SY, Holland KA, Gulluyan LM, Tavaria T, Weaver B, Reich NC, Kola I, Hertzog PJ Cloning and characterization of soluble and transmembrane isoforms of a novel component of the murine type I interferon receptor, IFNAR 2. J Biol Chem 272(38): Pan M, Kalie E, Scaglione BJ, Raveche ES, Schreiber G, Langer JA Mutation of the IFNAR-1 receptor binding site of human IFN-R2 generates type I IFN competitive antagonists. Biochemistry 47: Platanias LC Introduction: interferon signals: what is classical and what is non-classical? J Interferon Cytokine Res 25(12):732. Presky DH, Yang H, Minetti LJ, Chua AO, Nabavi N, Wu CY, Gately MK, Gubler U A functional interleukin 12 receptor complex is composed of two beta-type cytokine receptor subunits. Proc Natl Acad Sci USA 93(24): The PRISMS (Prevention of Relapses and Disability by Interferon-b-1a Subcutaneously in Multiple Sclerosis) Study Group and the University of British Columbia MS/MRI Analysis Group PRISMS-4: long-term efficacy of interferonb-1a in relapsing MS. Neurology 56: Río J, Nos C, Tintoré M, Téllez N, Galán I, Pelayo R, Comabella M, Montalban X Defining the response to interferonbeta in relapsing-remitting multiple sclerosis patients. Ann Neurol 59: Rubinstein MP, Kovar M, Purton JF, Cho JH, Boyman O, Surh CD, Sprent J Converting IL-15 to a superagonist by binding to soluble IL-15Ra. Proc Natl Acad Sci USA 103(24): Salomon I, Netzer N, Wildbaum G, Schif-Zuck S, Maor G, Karin N Targeting the function of IFN-gamma inducible protein 10 suppress ongoing adjuvant arthritis. J Immunol 169: Sandberg-Wollheim M, Bever C, Carter J, Farkkila M, Hurwitz B, Lapierre Y, Chang P, Francis GS, the EVIDENCE Study Group Comparative tolerance of IFNb 1a regimens in patients with relapsing remitting MS. The evidence study. J Neurol 252(1):8 13. Schellekens H Bioequivalence and the immunogenicity of biopharmaceuticals. Nat Rev Drug Discov 1: Serana F, Sottini A, Ghidini C, Zanotti C, Capra R, Cordioli C, Caimi L, Imberti L Modulation of IFNAR1 mrna expression in multiple sclerosis patients. J Neuroimmunol 197: Sherwood L Human Physiology: From Cells to Systems, 3rd ed. Minneapolis (MN): West Publishing Co. Sorensen PS, Deisenhammer F, Duda P, Hohlfeld R, Myhr KM, Palace J, Polman C, Pozzilli C, Ross C, for the EFNS Task Force on Anti-IFNb Antibodies in Multiple Sclerosis Guidelines on use of anti-ifnb antibody measurements in multiple sclerosis: report of an EFNS Task Force on IFNb antibodies in multiple sclerosis. Eur J Neurol 12: Sundaresan S, Roberts PE, King KL, Sliwkowski MX, Mather JP Biological response to ErbB ligands in non-transformed cell lines correlates with a specific pattern of receptor expression. Endocrinology 139: Uings IJ, Farrow SN Cell receptors and cell signalling. Mol Path 53:

33 REGULATING IFNAR ISOFORMS 741 Uzé G, Luftalla G, Morgensen KE a and b interferons and their receptors and their friends and relations. J Interferon Cytokine Res 15:3 26. Uzé G, Schreiber G, Piehler J, Pellegrini S The receptor of the type I interferon family [Review]. Curr Top Microbiol Immunol 316: Vallittu AM, Halminen M, Peltoniemi J, Ilonen J, Julkunen I, Salmi A, Eralinna JP, the Finnish Beta-Interferon Study Group Neutralizing antibodies reduce MxA protein induction in interferon beta-1a-treated MS patients. Neurology 58: Vallittu AM, Salmi AA, Eralinna JP, the Finnish Beta-Interferon Study Group MxA protein induction in MS patients treated with intramuscular IFN-beta-1a. Neurol Sci 26(6): Wildbaum G, Nahir MA, Karin N Beneficial autoimmunity to pro-inflammatory mediators restrains the consequences of self-destructive immunity. Immunity 19: Wildbaum G, Netzer N, Karin N Tr1 cell-dependent active tolerance blunts the pathogenic effects of determinant spreading. J Clin Invest 110: Yan H, Krishnan K, Lim JT, Contillo LG, Krolewski JJ Molecular characterization of an alpha interferon receptor 1 subunit (IFNAR1) domain required for TYK2 binding and signal transduction. Mol Cell Biol 16(5): Youssef S, Maor G, Wildbaum G, Grabie N, Gour-Lavie A, Karin N C-C chemokine-encoding DNA vaccines enhance breakdown of tolerance to their gene products and treat ongoing adjuvant arthritis. J Clin Invest 106: Zoon KC, Arnheiter H, Nedden DZ, Fitzgerald DJP, Willingham MC Human interferon alpha enters cells by receptormediated endocytosis. Virology 130: Address correspondence to: Dr. Francesca Gilli SCDO Neurology 2 Regional Reference Centre for Multiple Sclerosis (CReSM) Neuroscience Institute of the Cavalieri Ottolenghi Foundation University Hospital S. Luigi Gonzaga AOU S. Luigi Gonzaga Regione Gonzole 10, I Orbassano (Torino) Italy francesca.gilli@fastwebnet.it Received 6 August 2010/Accepted 6 August 2010

34

35 JOURNAL OF INTERFERON & CYTOKINE RESEARCH Volume 30, Number 10, 2010 ª Mary Ann Liebert, Inc. DOI: /jir Differential Gene Expression and Translational Approaches to Identify Biomarkers of Interferon Beta Activity in Multiple Sclerosis Ed Croze More than 16 years ago human interferon-b-1b (IFN-b-1b) was shown to be effective in the treatment of the relapsing-remitting form of multiple sclerosis ( MS). Over time, IFN-b has been demonstrated to be both a safe and effective treatment. However, the mechanism of action of IFN-b in MS remains unknown. To better understand the mechanism of action of IFN-b, considerable effort has been made in transcriptional profiling of peripheral blood mononuclear cells collected from MS patients. IFN-b is known to induce a large number of genes that play an important role in regulating responses to viral infection, immune modulation, and cell proliferation. Identifying differentially induced genes that are linked to the beneficial effects observed during treatment is under active investigation. IFN biomarkers in MS patients have been proposed but have not been clearly confirmed in independent studies or consistently correlated with clinical measures of disease progression. Organizing single genes or gene signatures grouped according to molecular mechanisms meaningful in MS may help to link IFN activity measurements to clinical outcomes. In this review, IFN activity measurements will be discussed with a specific emphasis on what is known about differential gene expression and treatment effects in MS. Introduction Multiple sclerosis Multiple sclerosis is defined broadly as an inflammatory disease of the central nervous system (CNS) characterized by a progression toward permanent neurological disability (Hafler and others 2005; Weiner 2009). Immune homoeostasis within the CNS becomes disrupted as activated immune cells enter the CNS. Disease progression results in neuronal damage and physical disability. Neurological damage in the CNS can range from mild to severe demyelination concurrent with the appearance of lesions in the brain and spinal cord that can disrupt neural transmission. Disability occurring during disease progression is permanent and results in a continuing decrease in quality of life measurements (Kurzke 1983; McDonald and others 2001). Multiple forms of multiple sclerosis (MS) are described, including primary progressive, secondary progressive, and relapsing-remitting (Keegan and Noseworthy 2002). The underlying cause of MS is unknown although immune dysfunction associated with T cell regulation, activation of proinflammatory cytokines, and production of myelinspecific antibodies coinciding with a breakdown of the blood brain barrier (BBB) have been observed in MS (Arnason and others 1996; Fujinami and others 2006). Interferonb-1b (IFN-b-1b) (Betaseron Ò /Betaferon Ò ; Bayer HealthCare Pharmaceuticals) (The IFN beta Multiple Sclerosis Study Group 1993) was the first approved IFN-b therapy for the treatment of the relapsing-remitting form of MS followed by IFN-b-1a (Avonex Ò ; Biogen, Inc.; Rebif Ò ; Ares-Serono, SA). Differential gene expression as a biomarker of IFN-b activity in MS Disease progression in MS is known to vary from individual to individual and across disease stages. Therefore, biomarker development will most likely require approaches limited not only to the profiling of a single gene but also the identification and monitoring of groups genes that change over time in response to treatment and disease progression. Monitoring changes in biology through measurement of differential gene expression represents a modern approach to biomarker development. Changes in biology occurring in MS patients as a result of treatment can be documented using gene expression profiling. The manner of investigating such effects is by definition an extension of single gene Translational Research, Global Medical Affairs, Neurology, Specialty Medicine, Bayer HealthCare Pharmaceuticals, Inc., Richmond, California. 743

36 744 CROZE transcriptomics (Lee 2005; Ruan and others 2004; Wang and others 2009) that involves the grouping of families or networks of interrelated genes sharing a common biological function (Subramanian and others 2005). In MS, it has been proposed that activated immune cells enter the CNS and stimulate resident asctocytes and microglia cells through the secretion of proinflammatory cytokines. Once in the CNS, activated immune cells can influence processes that result in tissue damage and neuronal degradation (Arnason and others 1996; Hafler and others 2005; Weiner 2009). One approach to measure the cause and effect of treatment is to determine the extent of differential gene expression in peripheral immune cells collected from IFN-b treated or treatment naive MS patients. However, relating the measurement of gene expression activity associated with peripheral blood mononucleocytes (PBMCs) located in the periphery to effects occurring within the CNS must be performed with some caution. Nevertheless, the ability of activated immune cells, located in the periphery, to enter the CNS and influence local inflammatory processes is well documented (Floris and others 2002; Agrawal and Yong 2007; Owens and others 2008). Type I IFNs were first identified as low molecular weight proteins primarily involved in establishing a protective response to viral infection. IFN-bs have since been demonstrated to elicit much broader effects, including the complex regulation of cytokine responses (Lengyel 1982; Ketlinskiy and Kalinina 2008), immune cell activation (Weinstock- Guttman and others 1995; Stark and others 1998; Keegan and Noseworthy 2002), cell migration (Calabresi and others 1997; Chatzimanolis and others 2004; Muraro and others 2004), BBB protection (Leppert and Waubant 1996; Stüve and others 1997; Trojano and Avolio 1999), control of cell growth (Sharief and Semra 2002; Gniadek and others 2003), and regulation of oxidative stress (Croze and others 2009a, 2009b). The observation that IFN-b induces expression of a large number of genes may help to explain the various biological effects associated with IFN-b (Der and others 1998; Stark and others 1998; Cunningham and others 2005). Classic biomarkers of IFN-b activity include the antiviral cytopathic effect assay and the measurement of , -oligoadenylate synthetase 1 (OAS1) and MxA activity. The MxA assay measures expression of the myxovirus (influenza virus) resistance 1 (MxA) protein and is widely used to determine the serum titer of neutralizing antibodies to IFN-b (Pachner and others 2003). However, the induction, by IFNb, of a specific IFN-reporter represents an attractive alternative to the MxA assay for determining IFN activity and neutralizing antibody titer (Russell-Harde and others 1999; Lallemand and others 2008; Reder and others 2008). In addition, measuring increases in plasma levels of neopterin and b2-microglobulin has also been used to demonstrate IFN activity. PBMCs isolated from MS patients have been used as starting material for RNA collection in most differential gene expression studies although whole blood and purified immune cell populations, such as monocytes, have also been employed (Fig. 1). Drawing comparisons between various RNA-based pharmacokinetic studies has been difficult because of differences in methods of analysis, gene chips, and starting material. Regardless, some important observations have been made using gene expression profiling in MS. Measurement of the pharmacodynamics of differential gene FIG. 1. Steps used to identify an IFN biomarkers in relapsing-remitting multiple sclerosis (MS). Selection of human study samples from 3 different populations begins with initial observations and biomarker selection made using samples collected directly from the disease setting (eg, MS). induction in PBMCs obtained from MS patients after a single administration of IFN-b-1b shows a rapid, short-term induction of gene expression. Reder and others (2008) demonstrated considerable patient-to-patient variation in differential expression of IFN-b-1b inducible genes in MS patients. Over 100 differentially regulated genes were identified and peak gene expression was observed 4 18 h after IFN-b-1b administration followed by a return to baseline at 42 h. Differentially expressed genes associated with particular gene ontology included a group of strongly induced chemokines, including CCL8, CCL2, CXCL10, and CXCL11. Other genes grouped according to function included those involved in protection from viral infection ( OAS1, OAS2, MxA, IFIT1, and OASL), immune response (IL-1RA), neuronal preservation (MT2A and MT1X), and cytokine signaling (IRF7, STAT1, STAT2, and SOCS1). Among the patients studied, considerable variation was observed in both differential gene expression and magnitude of response, most notably for MxA, CCL8, and IRF7. Variations in the degree of gene expression in response to IFN-b treatment were also observed in healthy individuals treated with IFNb-1b in which pharmacodynamic changes in over 200 differentially regulated genes were noted (Hilpert and others 2008). In addition, genes responding to IFN-b treatment appear to be involved in the regulation of both pro- and antiinflammatory responses (Stürzebecher and others 1999; Weinstock-Guttman and others 2003; Iglesias and others 2004; Hilpert and others 2008; Reder and others 2008; van Baarsen and others 2008; Comabella and others 2009). Interestingly, differential gene expression profiles obtained from MS patients or healthy individuals do not completely overlap, suggesting a difference in gene expression results between these 2 groups. Further, Rani and others (2009) observed an interindividual heterogeneity of response in patients treated with IFN-b; however, responses within each individual subject were stable over a 6-month interval. Such results suggest that patient response to IFN-b therapy may, in part, be dependent on a precondition of responsiveness to IFN-b. A number of studies focused on both short- and longterm IFN-b (IFN-b-1b and IFN-b-1a) gene expression effects in MS reveal the common induction of many important

37 IFN-b BIOMARKERS IN MULTIPLE SCLEROSIS 745 immunoregulatory genes, including TRAIL, MxA, CCL8, CXCR3, CCR2, IP-10, MCP-1, IRF, b2-microglobulin, MMP- 9-2, TIMP-2, IL-1RA, E2F pathways and others during treatment (Byskosh and Reder 1996; Stürzebecher and others 1999; Koike and others 2003; Weinstock-Guttman and others 2003; Achiron and others 2004; Hong and others 2004; Iglesias and others 2004; Cunningham and others 2005; Reder and others 2008). Of note, Stürzebecher and colleagues demonstrated the downregulation of IL-8 and the Flt-3 ligand in addition to observing a difference between the magnitudes of gene expression ex vivo versus in vitro (Stürzebecher and others 2003). This draws attention to the complex and differential nature of in vivo gene expression observed in MS patients and gene expression data obtained from cell populations stimulated in cell culture (ex vivo) with IFN-b. Measurement of differential gene expression can provide general information on IFN activity. However, due to the fact that no single gene has been linked to the beneficial effect of IFN treatment, single gene measurements may be able to predict, to some extent, IFN activity but can fall short of being able to predict effectiveness of treatment over time. In addition, in correcting for hypothesis testing of any singlegene biomarker, biological differences can be relatively small compared to background calculations and noise associated with the measurement of a single differentially expressed gene. Such complications can reduce the likelihood of reaching statistical significance. Single-gene analysis also runs the risk of missing important pathway information and associations with disease-related biologies and treatment outcomes. The grouping of signatures of IFN-inducible genes as a measure of IFN-b treatment effects has recently begun to be evaluated with encouraging results (Robinson and others 2003; Baranzini and others 2005; Satoh and others 2006; Fernald and others 2007; Goertsches and others 2008; van Baarsen and others 2008; Vosslamber and others 2009). van Baarsen and others (2008) selected a set of 15 IFNinduced genes whose expression demonstrated a strong negative correlation between baseline and biological response. Using the average response of these 15 genes it was suggested that the biological activity of IFN-b is not dependent on treatment regimens or interfering serum proteins such as neutralizing antibodies. Comabella and others (2009) identified a group of IFN-b activity genes induced in monocytes. Five genes detectable at both baseline and 3 months after treatment included IFIT3, IFIT1, OASL, IFIT2, and IFI44. A reduction in these differentially expressed genes resulted in a loss of IFN-b activity and presumably treatment effects. Studies by Serrano-Fernandez and others (2010) determined the time course of differentially expressed genes at 4 time points over a 1-year period. Samples were taken from patients treated with IFN-b before onset of therapy, 2 days, 1 month, and 1 year after start of treatment. Fifteen IFNresponsive genes were differentially expressed at all time points: EIF2AK2, IFI6, IFI44, IFI44L, IFIH1, IFIT1, IFIT2, IFIT3, ISG15, MX1, OASL, RASD2, SN, XAF1 and the uncharacterized at gene product. In the majority of patients studied, gene expression was highest 30 days after beginning of treatment. It is of interest to note that patient samples were drawn 42 h after administration of IFN-b, which is after the expected short-term peak of gene expression (eg, 4 18 h postadministration). Measurements taken at such a time may reflect long-term changes in differential gene expression as a result of prolonged treatment. Differentiating between short- (eg, 42 h) and long-term treatment effects is important because it is possible that prolonged treatment with IFN-b leads, in part, to a re-setting or modification of a predicted immune response. It is encouraging to note that results from different studies, performed by various laboratories, applying nonidentical approaches and using samples obtained from different cohorts of MS patients have resulted in the identification of a group of commonly recognized differentially expressed genes. A short list of these genes includes IFIT1, IFIT2, and Mx1 in addition to OAS1, OASL, RSAD2, ISG15, and MT2A, which are detected in most but not all studies. It is important to note that expression of these genes is detectable when studying either short- or long-term treatment effects (Reder and others 2008; van Baarsen and others 2008; Comabella and others 2009; Rani and others 2009; Vosslamber and others 2009; Serrano-Fernandez and others 2010). Such a gene list provides a testable signature of IFN activity that can be further evaluated, preferably using a larger number of patients. Measuring IFN activity markers and monitoring biomarkers of response to treatment in MS are not necessarily analogous. The beneficial effects of IFN-b treatment in MS may be the result of more than one mechanism of action (MOA). It may be that determining IFN activity in MS is best performed by the simultaneous measurement of a signature or group of differentially expressed genes linked to a given pathophysiology or biology associated with MS. Further, measuring the differential expression of a preselected group of genes can also provide increased statistical power that may be lost during the tracking of a single gene (Bild and Febbo 2005; Subramanian and others 2005; Linkov and others 2008; Yamaguchi and others 2008). It is important to note that measuring IFN-b responses occurring in PBMCs or immune cell populations alone could fall short in identifying a genuine biomarker of disease and in explaining the over all positive clinical effects observed in MS patients treated with IFN-b (Wills 1990; Weinstock-Guttman and others 1995; Dhib-Jalbut and Marks 2010). Deregulation of regulatory T cells (Tregs), cytotoxic CD8 cells, B cell activity, increased permeability of the BBB, processes involving innate immunity, and CNS-based neuronal damage all appear to play a central role in MS pathophysiology (Weinstock-Guttman and others 2003; Weiner 2009; Dhib-Jalbut and Marks 2010). Although IFN-b can induce changes in MS pathophysiology that are beneficial to the patient, it may not be readily apparent how this takes place. However, some encouraging associations can be drawn from differential gene expression studies. The induction of a group of chemokines that mediate cell migration (CCL8, CCL2, CXCL10, CXCL11, MCP2, CXCR3, CCR2, and CCR7) (Satoh and others 2006; Vallittu and others 2007; Reder and others 2008) and molecules able to help maintain the integrity of the BBB (MMP-9, TIMP-1) (Hartrich and others 2003) have been noted in a number of studies. The ability of IFN-b to induce genes that influence both chemokine activity and BBB integrity points to an important role for IFN-b in regulating immune cell migration and entry of activated immune cells into the CNS. The regulation of a number of genes associated with the Janus family kinases ( JAKs) is also of interest. STAT1, STAT2, SOCS1, JAK2, IRF7,

38 746 CROZE and NMI (Fernald and others 2007; Reder and others 2008; van Baarsen and others 2008; Rani and others 2009) are differentially regulated by IFN-b treatment, suggesting a possible role for IFN-b in regulating JAK-STAT signaling (Ketlinskiy and Kalinina 2008). This intracellular signaling pathway is utilized by a number of cytokines that can regulate important aspects of immunity and inflammation known to influence the course of MS (O Sullivan and others 2007; Pesu and others 2008). IFN-b has also been shown to upregulate a family of proteins called metallothioneins (MT2A and MT1X) (Reder and others 2008; van Baarsen and others 2008; Croze and others 2009a; Rani and others 2009) that are implicated in regulation of oxidative stress, neuropreservation, and prevention of axonal loss (Penkowa and Hidalgo 2003). One long-standing proposed MOA of IFN-b describes a shift in the balance of the T helper cell, Th1/Th2 immune system (Arnason and others 1996; Hafler and others 2005; Weiner 2009). IFN-b does play an important role in the regulation of effector functions occurring within the immune system in MS; however, differential gene expression studies suggest that additional MOA s could be important. Therefore, a Th1/Th2 pattern shift or reorganization of pro- and anti-inflammatory responses does not appear to offer a complete explanation of the beneficial effects observed in MS patients treated with IFN-b (Wandinger and others 2001; Reder and others 2008). The involvement of IFN-b in the regulation of a third class of T helper cells, Th17, could also be important in that IL-17 has been found in both acute and chronic active MS plaques (Tzartos and others 2008) and sera from MS patients naive to treatment (Axtell and others 2010). IFN-b is known to inhibit Th17-mediated autoimmune inflammation in mice and restricts CD4 þ cell-mediated inflammation through inhibition of Th17 responses (Guo and others 2008; Martin-Saavedra and others 2008). Although differential gene expression studies using Th17 cells stimulated with IFN-b are lacking, IFN-b has been demonstrated to increase expression of the regulatory cytokines IL-27 (Guo and others 2008) and IL-10 (Waubant and other 2001). Clearly, this is an important area for investigation. Along this line of thinking, additional gene expression studies focused on understand the differential regulation of immune cell populations (Tregs, Th17, CD8 þ, and CD4 þ ) in MS patients treated with IFN-b will more than likely yield exciting and important information. As previously discussed, disease progression varies from individual to individual, and treatment response to IFN may be affected by a number of conditions, including dosing schedule, route of administration, and IFN receptor levels on circulating immune cells. It may be possible to overcome some of these variables by developing a standardize multiplex IFN activity assay that allows the simultaneous analysis of a signature of genes demonstrated to be present during both short- and long-term IFN-b treatment. In addition, such a signature of response would be strengthened if it could be associated with treatment outcomes, including patient demographics, disease status, accepted clinical end points, or surrogate markers of disease activity, including expanded disability status scale, relapse rate, brain atrophy, neuronal preservation, cognition, and magnetic resonance imaging parameters (Rio and others 2009). One way to efficiently organize such an approach would be to exploit recent advances in System Biology (Brazhnik and others 2002; Hood and others 2004; Lee 2005; Bauch and Superti-Furga 2006). Advances in System Biology (a computer-based interdisciplinary study field that focuses on organizing complex interactions associated with biological processes) can be merged with interactive network analysis and literature databases (eg, Aridane ResNet, IPA, Pathway Builder, and Pathway Studio) (Fig. 2) to combine what is known about IFN-b activity measurements and biological effects as related to disease outcomes measures (Calvan and others 2005; Hoops and others 2006). Initial results using such an approach have begun to expand our understanding of the effects of IFN-b treatment in MS and to identify differences in response to IFN-b between healthy individuals and MS patients (Yamaguchi and others 2008). In addition, network analysis approaches, using data collected from MS patients, have identified key changes in biology associated with MS and treatment (Baranzini and others 2005; Fernald and others 2007; Corvol and others 2008; Yamaguchi and others 2008). By taking advantage of this approach it may be possible to identify a group of genes or an IFN-b biomarker signature based on available MS scientific literature, genomic databases, gene ontology, and observations made directly within the disease setting. A remaining challenge will be how to organize IFN biomarkers along the line of disease characteristics and treatment effects over time. This will require stratification of a sufficient number of patients to achieve statistical significance and to perform hypothesis testing and validation of any proposed MS biomarker. Conclusions and Perspectives Validated biomarkers able to monitor IFN treatment effects in MS are not currently available. Measurement of IFN activity is possible but a strong linkage to specific diseaserelated outcomes will require additional studies. To date, variable RNA-based pharmacokinetic data together with the complex nature of MS have made it difficult to convincing link a given IFN biomarker to clinical measurements of FIG. 2. System and computational biology method to identify pathway components important in MS. Examples of the derivation of the regulated-by-ifn-b pathways components from the scientific literature. Proteins are depicted as nodes. Two proteins are connected if there is a relationship, such as regulates, between them (Yamaguchi and others 2008).

39 IFN-b BIOMARKERS IN MULTIPLE SCLEROSIS 747 disability and disease progression. The ability to group patients into clearly defined treatment response or nonresponsive groups is not standardized. However, the integration of proteomics, transcriptomics, system biology, curated literature databases, and patient samples collected from large, well-documented clinical studies provides a hopeful approach for future development of MS biomarkers able to predict and monitor benefit of treatment. Identification of clinically meaningful IFN activity measurements based on a single differentially expressed gene may not be possible; however, the simultaneous monitoring of a selected signature of differential expressed genes appears to represent a promising approach. Identifying IFN biomarkers associated with changes in MS pathophysiology occurring during disease progression and treatment can be aided using recent advances in system biology. Recent efforts, in this regard, have been encouraging, but additional hypothesis testing and validation of potential biomarkers using a larger number of patient samples followed by confirmation against independent studies and direct association with disease outcome measures is warranted. Acknowledgments This article was developed and drafted by the study author. No writing or editorial assistance was utilized in the production of this review. Author Disclosure Statement Dr. Croze is affiliated with Bayer HealthCare Pharmaceuticals Inc., Richmond, California, and has a financial involvement with this organization. References Achiron A, Gurevich M, Friedman N, Kaminski N, Mandel M Blood transcriptional signatures of multiple sclerosis: unique gene expression of disease activity. Ann Neurol 55: Agrawal SM, Yong WW Immunopathogenesis of MS. Int Rev Neurobiol 79: Arnason BGW, Dayal A, Qu ZX, Jensen M, Genç K, Reder AT Mechanisms of action of interferon beta in multiple sclerosis. Semin Immunopathol 18: Axtell RC, de Jong BA, Boniface K, van der Voort L, Bhat R, De Sarno P, Naves R, Han M, Zhong F, Castellanos JG, Mair R, Christakos A, Kolkowitz I, Katz L, Killestein J, Polman CH, de Waal Malefyt R, Steinman L, Raman C T helper type I and 17 cells determine efficacy of interferon-b in multiple sclerosis and experimental encephalomyelitis. Nat Med 16: Baranzini SE, Mousavi P, Rio J, Caillier SJ, Stillman A, Villoslada P, Wyatt MM, Comabella M, Greller LD, Somogyi R, Montalban X, Oksenberg JR Transcription-based prediction of response to IFN-b using supervised computational methods. PLoS Biol 3:E2. Bauch A, Superti-Furga G Charting protein complexes, signaling pathways, and networks in the immune system. Immunol Rev 210: Bild A, Febbo PG Application of a priori established gene sets to discover biologically important differential expression in microarray data. Proc Natl Acad Sci U S A 102: Brazhnik P, de la Fuente A, Mendes P Gene networks: how to put the function in genomics Trends Biotech 20: Byskosh PV, Reder AT Interferon beta-1b effects on cytokine mrna in peripheral mononuclear cells in multiple sclerosis. Mult Scler 1: Calabresi PA, Tranquill R, Dambrosia JM, Stone L, Maloni H, Bash CN, Frank J, McFarland HF Increases in soluble VCAM-1 correlate with a decrease in MRI lesions in multiple sclerosis treated with interferon beta-1b. J Neuroimmunol 41: Calvan ES, Xiao W, Richards DR, Felciano RM, Baker HV, Cho RJ, Chen RO, Brownstein HB, Cobb JP, Tschoeke SK, Miller- Graziano LC, Moldawe L, Mindrinos MN, Davis RW, Tompkins RG, Lowry SF A network-based analysis of systemic inflammation in humans. Nature 437: Chatzimanolis N, Kraus J, Bauer R, Engelhart B, Bregenzer T, Kuehne BS, Tofighi J, Laske C, Stolz F, Blaes F, Voigt K, Traupe H, Kaps M, Oschmann P CD45RAþ ICAM3þ lymphocytes in interferon beta-1b treated and untreated patients with relapsing-remitting multiple sclerosis. Acta Neurol Scand 110: Comabella M, Lünemann JD, Río J,Sánchez A, López C, Julià E, Fernández M, Nonell L, Camiña-Tato M, Deisenhammer F, Caballero E, Tortola MT, Prinz M, Montalban X, Martin R A type I interferon signature in monocytes is associated with poor response to interferon-beta in multiple sclerosis. Brain 132: Corvol J-C, Pelletier D, Henry RG, Cailler SJ, Wang J, Pappas D, Casazza S, Okuda DT, Hauser SL, Oksenberg JR, Baranzini SE Abrogation of T cell quiescence characterizes patients at high risk for multiple sclerosis after the initial neurological event. PNAS 105: Croze E, Beekman J, Knappertz, V, Sanbrink R, Pohl C, Paz P, Lindberg R, Kappos L, Reischl J. 2009a. Betaseron increases expression of metallothioneins in relapsing-remitting multiple sclerosis patients suggesting a role for betaseron in neuroprotection. Mult Scler 15:S138. Croze E, Velichko S, Tran T, Yamaguchi KD, Salamon H, Reder AT, Knappertz V, Paz P. 2009b. Betaseron induces a novel alternative start transcript in cells obtained from relapsingremitting multiple sclerosis patients and human brain that is associated with control of oxidative resistance. Mult Scler 15:S138. Cunningham S, Graham C, Hutchinson M, Droogan A, O Rourke K, Patterson C, McDonnell G, Hawkins S, Vandenbroeck K Pharmacogenomics of responsiveness to interferon IFN-b treatment in multiple sclerosis: A genetic screen of 100 type I interferon-inducible genes. Clin Pharmacol Ther 78: Der S, Zhou A, Williams B, Silverman RH Identification of genes differentially regulated by interferons a, b, or g using oligonucleotide arrays. Proc Natl Acad Sci USA 95: Dhib-Jalbut S, Marks S Interferon-b mechanisms of action in multiple sclerosis. Neurology 74:S17 S24. Fernald GH, Knott S, Pachner A, Caillier SJ, Narayan K, Oksenberg JR, Mousavi P, Baranzini SE Genome-wide network analysis reveals the global properties of IFN-beta immediate transcriptional effects in humans. J Immunol 178: Floris S, Ruuls SR, Wierinck A, van der Pol SM, Dopp E, van der Meide PH, Dijkstra CD, De Vries HE Interferon-b directly influences monocyte inflitration into the central nervous system. J Neuroimmunol 127:69 79.

40 748 CROZE Fujinami RS, von Herrath M, Christen U, Whitton JL Molecular mimicry, bystander activation, or viral persistence: infections and autoimmune disease. Clin Micro Rev 19: Gniadek P, Oktas O, Wandlinger K-P, Bellmann-Strobi J, Wengert O, Weber A, von Wussow P, Obert H-J, Zipp F Systemic IFN-b treatment induces apoptosis of peripheral immune cells in MS patients. J Neuroimmunol 137: Goertsches RH, Hecker M, Zerrl UK Monitoring of multiple sclerosis immunotherapy from single candidates to biomarker networks. J Neurol 255: Guo B, Chang EY, Cheng G The type I IFN induction pathway constrains Th17-mediated autoimmune inflammation in mice. J Clin Invest 118: Hafler DA, Slavik J, Anderson DE, O Connor K, De Jager P, Baecher-Allan C. Multiple Sclerosis Immunol Rev 204: Hartrich L, Weinstock-Guttman B, Hall D, Badgett D, Baier M, Patrick K, Feichter J, Hong J, Ramanathan M Dynamics of immune cell trafficking in interferon-beta treated multiple sclerosis patients. J Neuroimmunol 139: Hilpert J, Beekman JM, Schwenke S, Kowal K, Bauer D, Lampe J, Sandbrink R, Heubach JF, Stürzebecher S, Reischl J Biological response genes after single dose administration of interferon beta-1b to healthy male volunteers. J Neuroimmunol 199: Hong J, Zang Y, Hutton G, Rivera V, Zhang, J Gene expression profiling of relevant biomarkers for treatment evaluation in multiple sclerosis. J Neuroimmunol 152: Hood L, Heath JR, Phelps ME, Lin B Systems biology and new technologies enable predictive and preventative medicine. Science 306: Hoops S, Sahle S, Gauges R, Lee C, Pahle J, Simus N, Singhal M, Xu L, Nendes P, Kummer U COPASI a complex pathway simulator. Bioinformatics 22: Iglesias AH, Camelo S, Hwang D, Villanueva R, Stephanopoulos G, Dangond F Microarray detection of E2F pathway activation and other targets in multiple sclerosis peripheral blood mononuclear cells. J Neuroimmunol 150: Keegan BM, Noseworthy JH Multiple sclerosis. Ann Rev Med 53: Ketlinskiy SA, Kalinina NM Cytokines in demyelinating diseases. In: Phelps C, Korneva E, eds. Cytokines and the brain. Atlanta, GA, USA: Elsevier B.V., pp Koike F, Satoh J, Miyake S, Yamamoto T, Kawai M, Kikuchi S, Nomura K, Yokoyama K, Ota K, Kanda T, Fukazawa T, Yamamura T Microarray analysis identifies interferon b-regulated genes in multiple sclerosis. J Neuroimmunol 139: Kurzke JF Rating neurological impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 33: Lallemand C, Meritet JF, Erickson R, Grossberg SE, Roullet E, Lyon-Caen O, Lebon P, Tovey MG Quantification of neutralizing antibodies to human type I interferons using division-arrested frozen cells carrying an interferon-regulated reporter-gene. J Interferon Cytokine Res 28: Lee NH Genomic approaches for reconstructing gene networks. Pharmacogenomics 6: Lengyel P Biochemistry of interferon and their actions. Ann Rev Biochem 51: Leppert D, Waubant E Interferon beta-1b inhibits gelatinase secretion and in vitro migration of human T cells: a possible mechanism for treatment efficacy in multiple sclerosis. Ann Neurol 40: Linkov F, Ferris RL, Yurkovetsky Z, Marrangoni A, Velikokhatnaya L, Gooding W, Nolan B, Winans M, Siegel ER, Lokshin A, Stack BC Multiplex analysis of cytokines as biomarkers that differentiate benign and malignant thyroid diseases. Proteomics Clin Appl 10: Martin-Saavedra FM, Gonzalez-Garcia C, Bravo B, Ballester S b-interferon restricts the inflammatory potential of CD4 þ cells through the boost of the Th2 phenotype, the inhibition of Th17 response and the prevalence of naturally occurring T regulatory cells. Mol Immunol 45: McDonald WI, Compston A, Edan G, Goodin D, Hartung HP, Lublin FD, McFarland HF, Paty DW, Polman CH, Reingold SC, Sandberg-Wollheim M, Sibley W, Thompson A, van den Noort S, Weinshenker BY, Wolinsky JS Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis. Ann Neurol 50: Muraro PA, Liberati L, Bonanni L, Pantaline CM, Caporale C, Iarlori G, DeLuca D, Farina A, Lugaresi D, Gambi D Decreased integrin gene expression in patients with MS responding to interferon-b treatment. J Neuroimmunol 150: O Sullivan LA, Liongue C, Lewis RS, Stephenson SE, Ward AC Cytokine receptor signaling through the Jak-Stat-Socs pathway in disease. Mol Immunol 44: Owens T, Bechmann I, Engelhardt B Perivascular spaces and the two steps to neuroinflammation. J Neuropathol Exp Neurol 12: Pachner AR, Bertolotto A, Deisenhammer F Measurement of MxA mrna or protein as a biomarker of IFN beta bioactivity:detection of antibody-mediated decreased bioactivity (ADA). Neurology 61:S24 S26. Penkowa M, Hidalgo J Treatment with metallothionein prevents demyelination and axonal damage and increases oligodendrocyte precursors and tissue repair during experimential autoimmune encephalomyelitis. J Neurosci Res 72: Pesu M, Laurence A, Kishore N, Zwillich SH, Chan G, O Shea JJ Therapeutic targeting of janus kinases. Immunol Rev 223: Rani MR, Xu Y, Lee JC, Shrock J, Josyula A, Schlaak J, Chakraborthy S, Ja N, Ransohoff RM, Rudick RA Heterogeneous, longitudinally stable molecular signatures in response to interferon-beta. Ann N Y Acad Sci 1182: Reder AT, Velichko S, Yamaguchi KD, Hamamcioglu K, Ku K, Beekman J, Wagner TC, Perez, HD, Salamon H, Croze E IFN-b1b induces transient and variable gene expression in relapsing-remitting multiple sclerosis patients independent of neutralizing antibodies or changes in IFN receptor RNA expression. J Interferon Cytokine Res 28: Rio J, Comabella M, Montalban X Predicting responders to therapies for multiple sclerosis. Nat Rev Neurol 5: Robinson WH, Utz P, Steinman L Genomic and proteomic analysis of multiple sclerosis. Curr Opin Immunol 15: Russell-Harde D, Wagner TC, Perez HD, Croze E Formation of a uniquely stable type I interferon receptor complex by interferon beta is dependent upon particular interactions between interferon beta and its receptor and independent of tyrosine phosphorylation. Biochem Biophys Res Commun 255: Ruan Y, Le Bar P, Ng HH, Liu ET Interrogating the transcriptome. Trends Biotech 20: Satoh J, Nanri Y, Tabunoki H, Yamamura T Microarray analysis identifies a set of CXCR3 and CCR2 ligand chemokines as early IFN-b-response genes in peripheral blood

41 IFN-b BIOMARKERS IN MULTIPLE SCLEROSIS 749 lymphocytes in vitro: an implication for IFN-b-related adverse effects in multiple sclerosis. BMC Neurol 6: Serrano-Fernandez P, Moller S, Goertsches R, Fiedler H, Koczan D, Thiesen HJ, Zettl, UK Time course transcriptomic of IFN-b1b drug therapy in multiple sclerosis. Autoimmunity 43: Sharief MK, Semra YK Down-regulation of survivin expression in T-lymphocytes after interferon-b1a treatment in patients with multiple sclerosis. Arch Neurol 59: Stark GR, Kerr IM, Williams BR, Silverman RH, Schreiber RD How cells respond to interferons. Annu Rev Biochem 67: Stürzebecher S, Maibauer R, Heuner A, Beckmann K, Aufdembrinke B Pharmacodynamic comparison of single doses of IFN-b1a and IFN-b1b in healthy volunteers. J Interferon Cytokine Res 19: Stürzebecher S, Wandinger KP, Rosenwald A, Sathyamoorthy M, Tzou A, Mattar P, Frank JA, Staudt L, Martin R, McFarland HF Expression profiling identifies responder and nonresponder phenotypes to interferon beta in multiple sclerosis. Brain 126: Stüve O, Chabot S, Jung SS, Williams G, Yong VW Chemokine-enhanced migration of human peripheral blood mononuclear cells is antagonized by interferon beta-1b through an effect on matrix metalloproteinase-9. J Neuroimmunol 80: Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, Mesirov JP Gene set enrichment analysis: a knowledgebased approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 102: The IFN beta Multiple Sclerosis Study Group Interferon beta-1b is effective in relapsing-remitting multiple sclerosis. I Clinical results of a multicenter, randomized, double-blind, placebo-controlled trial. Neurology 43: Trojano M, Avolio C Changes of serum sicam-1 and MMP-9 induced by rifnb-1b treatment in relapsing-remitting MS. Neurology 53: Tzartos JS, Friese MA, Craner MJ, Palace J, Newcombe J, Esiri MM, Fugger L Interleukin-17 production in central nervous system-infiltrating T cells and glial cells is associated with active disease in multiple sclerosis. Am J Pathol 172: Vallittu AM, Saraste M, Airas L CCR7 expression on peripheral blood lymphocytes is up-regulated following treatment of multiple sclerosis with interferon-beta. Neurol Res 29: Van Baarsen LG, Vosslamber S, Tijssen M, Baggen JM, van der Voort LF, Killestein J, van der Pouw Kraan TC, Polman CH, Verweij CL Pharmacogenomics of interferon-b therapy in multiple sclerosis: baseline IFN signature determines pharmacological differences between patients. PLoS 3:E1927. Vosslamber S, van Baarsen L, Verweij CL Pharmacogenomics of IFN-b in multiple sclerosis: towards a personalized medicine approach. Pharmacogenomics 10: Wandinger KP, Stürzebecher CS, Bielekova B, Detore G, Rosenwald A, Staudt LM, McFarland HF, Martin R Complex immunomodulatory effects of interferon beta in multiple sclerosis including the upregulation of T helper 1-associated marker genes. Ann Neurol 59: Wang Z, Gerstein M, Snyder M RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet 10: Waubant E, Gee L, Bacchetti P, Sloan R, Cotleur A, Rudick R, Goodkin D Relationship between serum levels of IL-10, MRI activity and interferon beta-1a therapy in patients with relapsing remitting MS. J Neuroimmunol 112: Weiner HL The challenge of multiple sclerosis: how do we cure a chronic heterogenous disease? Ann Neurol 65: Weinstock-Guttman B, Badgett D, Patrick K, Hartrich L, Santos R, Hall D, Baier M, Feicher J, Ramanathan M Genomic effects of IFN-b in multiple sclerosis patients. J Immunol 171: Weinstock-Guttman B, Ransohoff RM, Kinkel P, Rudick RA The interferons: biologic effects, mechanism of action, and use in multiple sclerosis. Ann Neurol 37:7 15. Wills RJ Clinical pharmacokinetics of interferons. Clin Pharmacokinet 19: Yamaguchi KD, Ruderman DL, Croze E, Wagner TC, Velichko S, Reder AT, Salamon H IFN-beta-regulated genes show abnormal expression in therapy-naïve relapsingremitting MS mononuclear cells: gene expression analysis employing all reported protein-protein interactions. J Neurol 195: Address correspondence to: Dr. Ed Croze Translational Research Global Medical Affairs Neurology, Specialty Medicine Bayer HealthCare Pharmaceuticals, Inc Hilltop Drive Richmond, CA ed.croze@bayer.com Received 16 July 2010/Accepted 16 July 2010

42

43 JOURNAL OF INTERFERON & CYTOKINE RESEARCH Volume 30, Number 10, 2010 ª Mary Ann Liebert, Inc. DOI: /jir Regulatory Effects of Interferon-b on Osteopontin and Interleukin-17 Expression in Multiple Sclerosis Jian Hong and George J. Hutton Multiple sclerosis (MS) is a demyelinating disease characterized by autoimmune inflammation in the central nervous system. Despite over a decade of use of interferon-b (IFN-b) in the treatment of MS, its mechanisms of action are still not fully elucidated. New data now demonstrate that the 2 important proinflammatory cytokines involved in the pathogenesis of MS, osteopontin (OPN) and interleukin-17 (IL-17), are regulated by IFN-b. This review discusses the role of OPN and IL-17 in the development of MS and how the downregulation of the levels of OPN and interleukin-17 contributes to the therapeutic effects of IFN-b in MS. Introduction Interferon-b (IFN-b) is a major disease-modifying agent used for treatment of multiple sclerosis (MS). Most of the available data suggest that the therapeutic effects of IFN-b are related to its antiinflammatory properties, although the precise mechanisms for such beneficial actions remain unknown. Accumulated data suggest that the downregulation of T helper 1 cytokines and the inhibition of the migration of inflammatory T cells into the central nervous system (CNS) are important factors in the therapeutic effects of IFN-b. For a number of years Th1-related cytokines such as IFN-g have been considered to be the sole culprit of CNS inflammation in MS. As a matter of fact, a constellation of proinflammatory and inflammatory cytokines are implicated in the onset and relapse of MS. Recent discoveries suggest that 2 important proinflammatory cytokines, osteopontin (OPN) and interleukin-17 (IL-17), play significant roles in the pathogenesis of MS (Chabas and others 2001; Komiyama and others 2006; Hur and others 2007; Tzartos and others 2008). OPN (also known as early T cell activation gene 1 or ETA1 is encoded by secreted phosphoprotein 1 [SPP1]) is a T-bet-dependent proinflammatory cytokine produced by Th1 cells and dendritic cells (Weber and Cantor 1996; Ashkar and others 2000). OPN, as a proinflammatory cytokine, regulates expression of downstream inflammatory cytokines such as IL-10, IL-12, IL- 23, IL-27, and IL-17. IL-17 is expressed by a distinctive cell lineage named Th17 cells also recognized as a key mediator of MS. Th17 seems particularly important in the onset of MS. Recent findings indicate that IFN-b downregulates expression of OPN and the differentiation of IL-17-secreting Th17 cells in MS. This suggests that the regulation of OPN expression and Th17 cell differentiation constitute a significant part of the mechanisms of action IFN-b in the treatment for MS. OPN in MS OPN-producing cells and immunologic functions OPN is a pleiotropic protein that is expressed in a wide range of cells, including fibroblasts, preosteoblasts, osteoblasts, osteocytes, odontoblasts, osteoclasts, hypertrophic chondrocytes, some bone marrow cells, neutrophils, T cells, mast cells, dendritic cells, macrophages, smooth muscle, skeletal muscle myoblasts, endothelial cells, and other nonbone cells of the inner ear, brain, kidney, deciduum, and placenta. OPN is involved in a broad spectrum of biological responses, including bone remodeling, inflammation, cancer, and antimicrobial infection (Murugaiyan and others 2008). Of particular interest is the recent research into the modulation of immune responses by OPN. OPN is able to bind to several integrin receptors, including a4b1, a9b1, and a9b4. In inflammation, OPN serves as a proinflammatory cytokine and interacts with other cytokines. In OPN / mice, IL-12 expression is diminished, whereas IL- 10 expression is increased. A phosphorylation-dependent interaction between the amino-terminal portion of OPN and its integrin receptor stimulates IL-12 expression, whereas a phosphorylation-independent interaction with CD44 inhibits IL-10 expression (Ashkar and others 2000). Further, enhanced IFN-g expression by T cells was observed in OPN / mice (Chabas and others 2001; Jansson and others 2002). In a collagen-induced arthritis model of rheumatoid arthritis, OPN recruits inflammatory cells to arthritic joints (Gravallese 2003). OPN prevents macrophages from leaving the accumulation site in the brain of rhesus monkeys with neuroaids (Burdo and others 2007). OPN is also an important antiapoptotic factor (Gravallese 2003; Cao and others 2008). Anti-OPN antibody was effective in inhibiting the development of collagen-induced Department of Neurology and Baylor Multiple Sclerosis Center, Baylor College of Medicine, Houston, Texas. 751

44 752 HONG AND HUTTON arthritis and even reversed established disease in DBA/1J mice (Fan and others 2008). OPN in the pathogenesis of MS The discovery of a potential role of OPN in the pathogenesis of MS originated from a high throughput microarray screening of the CNS lesions of experimental autoimmune encephalomyelitis (EAE) rats. High level of OPN transcripts was found inthecnslesionsineae.further,eaewaslessseverein OPN-deficient mice (SPP1 / ) (Chabas and others 2001). The level of OPN is also elevated in the serum of patients with MS (Vogt and others 2003; Comabella and others 2005). The progression of MS and EAE are related to OPN, as it is found that OPN can promote the survival of activated T cells in the CNS (Hur and others 2007; Stromnes and Goverman 2007). Increased OPN expression has been found to amplify IL-17 production in CD4 þ T cells in both EAE and MS. Anti-OPN treatment, given before or after EAE induction, reduced the clinical severity of the disease (Murugaiyan and others 2008). In IFN-b-treated MS patients, the serum level of OPN was found to be decreased compared to that in untreated control patients (Chen and others 2009a). OPN expression is also markedly increased in CD11c þ DC of untreated MS patients compared to that of healthy individuals (Murugaiyan and others 2008). Collectively, it might be suspected that OPN could be a sensitive marker reflecting ongoing active Th1 response, and changes of OPN expression, in either soluble form or intracellular form, might correlate with disease progression. IL-17 in MS Th17 differentiation and lineage survival Differentiation of T cells toward the Th17 phenotype is driven principally by a specific cytokine milieu. Th17 cells develop from naive T cells in response to TGF-b, IL-6, and IL-21. IL-6 activates signal transducer and activator of transcription 3 (STAT3) and the lineage-determining transcription factors orphan nuclear receptor (RORgt) and RORa. This differentiation is enhanced in the presence of IL-1b and TNF-a. IL-1b could change the cytokine secretion of DC, which in turn inhibits the differentiation of Th17 cells (Zhang and others 2009). On the other hand, IL-1RI þ CD4 þ T cells were found to exhibit increased IL-17, RORc, and interferon regulatory factor 4 (IRF4) gene expression before activation, indicating that the effect of IL-1b is programmed in these cells via IL-1RI (Lee and others 2010). The differentiation of Th17 is strongly inhibited by IFN-g, IL-4,IL-27,IL-2,andIL-35.Afterthelineageisformed,IL-23 maintains the survival of differentiated Th17 cells. In IL-23 / mice, there is a deficiency of Th17 cells. Recently, IL-7 was found to participate in the maintenance of the Th17 cell lineage (Liu and others 2010; Zuvich and others 2010). In mice treated with multiple doses of anti-il-7 antibody, Th17 cells were depressed due to a block at a late stage of expansion. Differentiated Th17 cells are susceptible to apoptosis in the absence of IL-7. IL-27 derived from macrophages markedly inhibits the differentiation of Th17 cells. IL-27 was found to inhibit RORc and to break commitment of the Th17 lineage (Diveu and others 2009). IL-17 in the pathogenesis of MS Th17 cells produce an array of cytokines, including IL- 17A, IL-17F, and IL-22. IL-17A is commonly used as the IL-17 subtype indicative of the growth of Th17 cells. As the importance of Th17 cells in mediating tissue damage has been gradually recognized, the importance of the Th1/Th2 paradigm in the physiopathology of EAE and MS has been replaced to a certain extent by the Th17 theory. EAE induced in IL-17 / mice exhibited delayed onset, reduced maximum severity scores, ameliorated histological changes, and early recovery (Komiyama and others 2006). In inducible co-stimulator (ICOS)-deficient mice, the blockade of IL-17 strongly inhibited MOG induced EAE (Galicia and others 2009). Therapeutic neutralization of IL-17 with IL-17-receptor Fc-protein in acute EAE ameliorated clinical symptoms of EAE. Neutralization of IL-17 with a monoclonal antibody also ameliorated the disease course (Hofstetter and others 2005). However, 1 research group showed that IL-17 does not contribute to the development of EAE (Haak and others 2009). In their study, neither the T celldriven overexpression of IL-17A nor its complete loss had a major impact on the development of clinical disease. In IL- 17F-deficient mice, the mice were fully susceptible to EAE and displayed unaltered emergence and expansion of autoreactive T cells during disease. Treatment of IL-17F-deficient mice with antagonistic monoclonal antibodies specific for IL-17A again produced only a minimal beneficial impact on disease development. Compared to IL-17, IL-23 seems to be more important in the development of EAE (Cua and others 2003). IL-22, the other cytokine expressed by Th17 cells, was found to be involved in the physiopathology of other autoimmune diseases such as psoriasis and ulcerative colitis, rather than EAE and MS (Wolk and others 2006; Ma and others 2008; Sugimoto and others 2008). Despite the conflicting evidence, MS lesions still have a signature of IL-6, IFN-g, OPN, and IL-17(Chabas and others 2001; Boniface and others 2008). IL-17 likely remains a major proinflammatory cytokine in MS; however, its exact role remains to be elucidated. IL-17 may participate in a particular stage of EAE. Regulation of OPN and IL-17 by IFN-b Despite over a decade of use, the exact mechanism of IFNb in MS treatment is still unclear. There are several mechanisms of action believed to be involved in the therapeutic effect of IFN-b. IFN-b shifts Th1 cells to Th2 cells (Kozovska and others 1999), decreases the migration of T cells into the CNS by inhibition of matrix metalloproteinase-9 (Stuve and others 1996; Lou and others 1999; Yushchenko and others 2003; Boz and others 2006), reduces antigen presentation ( Jiang and others 1995), and inhibits proliferation of T cells (Malik and others 1998). In addition, IFN-b may be able to restore the function of suppressive cells that seem to be impaired in MS patients (Korporal and others 2008; Martin- Saavedra and others 2008). IFN-b may also improve the integrity of the blood brain barrier (Steinman 2001). Recent studies show that IFN-b downregulates expression of IL-17 and OPN in both humans and animals (Guo and others 2008; Martin-Saavedra and others 2008; Prinz and others 2008; Shinohara and others 2008; Chen and others 2009a), which may reveal a new mechanism of action of IFN-b (Table 1). It was discovered that IFN-b inhibited expression of OPN in CD4 þ T cells and DC (Guo and others 2008; Prinz and others 2008; Shinohara and others 2008; Chen and others 2009a). The depressed OPN in DC subsequently inhibits Th17 differentiation through derepressed IL-27 (Shinohara

45 REGULATION OF OPN AND IL-17 BY IFN-b IN MS 753 Table 1. Mechanisms of Action of Interferon-b in the Treatment of Multiple Sclerosis Mechanisms Evidence in favor Evidence against Th1/Th2 shift Block antigen presentation Block T cell migration Induction of Treg cells Modulate osteopontin expression Increased interleukin-10 and decreased IFN-g production (Noronha and others 1993; Rep and others 1996; Rudick and others 1996; Kozovska and others 1999; Martin-Saavedra and others 2008) Inhibits the induction of MHC class II molecules on human monocytes caused by IFN-g (Barna and others 1989). Reduces expression of MHC class II molecules on human microvessel endothelia cells or rodent microglia (Hall and others 1997). Reduces antigen presentation by B cells (Jiang and others 1995) Reduces matrix metalloproteinase-9 expression and transmigration (Leppert and others 1996; Stuve and others 1996, 1997) Observed expansion of Treg cells in multiple sclerosis patients treated with IFN-b after 6 months (de Andres and others 2007; Korporal and others 2008; Martin-Saavedra and others 2008; Namdar and others 2010) Inhibits intracellular and extracellular levels of osteopontin in humans and mice (Shinohara and others 2008; Chen and others 2009a) Regulation of Th17 cells Inhibits Th17 cell development (Guo and others 2008; Martin-Saavedra and others 2008; Prinz and others 2008; Shinohara and others 2008; Chen and others 2009a) Inhibits Th2 cell differentiation from naïve T cells (McRae and others 1997) Fails to inhibit basal or IFN-g-induced MHC class II molecules on human astrocytes and microglia (McLaurin and others 1995) Abbreviation: IFN, interferon. and others 2008). This finding seems important when considering the role of OPN and Th17 cells in the pathogenesis of MS. It has been shown that endogenous expression of IFNb in the CNS is significantly lower in patients with MS than in normal controls (Prinz and others 2008). One assumption is that injected IFN-b has to reach the CNS in order to be effective. However, the size of IFN-b may preclude it from crossing the normal blood brain barrier and being able to access lesions in the CNS. It is likely, despite some damage to the blood brain barrier in MS, that exogenous IFN-b acts principally in the periphery by diminishing active Th1 and Th17 responses. Downregulation by IFN-b. of OPN and other relevant molecules that affect transmigration across the blood brain barrier may, however, play an important role in the therapeutic action of IFN-b in MS. OPN is a critical molecule mediating the transmigration of many types of inflammatory cells across the blood brain barrier. In OPN / mice, regulated on activation, normal T cell expressed and secreted (RANTES)-induced T cell migration is significantly impaired (Chen and others 2009a). Enhanced CD8 þ T cell migration was observed in OPN transgenic mice (Higuchi and others 2004). In other cell types, OPN was found to increase migration and matrix metalloproteinase-9 upregulation via avb3 integrin, focal adhesion kinase (FAK), extracellular-regulated kinase (ERK), and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-kB)-dependent pathways (Chen and others 2009b). IFN-b may exert both direct and indirect effects on Th17 cells. Thus, IFN-b may bind directly to its receptor on Th17 cells, resulting in inhibition of the phosphorylation of STAT3, thereby blocking Th17 cell differentiation. On the other hand, IFN-b may decrease intracellular level of OPN in dendritic cells and subsequently upregulate IL-27 production by DC. Elevated IL-27 suppresses the differentiation of Th17 cells (Fig. 1). Although it is apparent that IFN-b inhibits Th17 cell differentiation, a recent study of the effectiveness of IFN-b treatment in EAE indicates that beneficial effects of IFN-b treatment may not be related to an effect on Th17 cells at least under some circumstances (Axtell and others 2010). It was shown in this study that IFN-b exacerbates Th17- mediated EAE but dampens Th1-mediated EAE. IFN-b requires IFN-g signaling to be effective in the treatment of EAE. It appears that the benefit of IFN-b treatment is related to the induction of IL-10 and the presence of IFN-g-responsive antigen-presenting cells. It was also found that IL-17F concentration is higher in IFN-b-nonresponsive MS patients (Axtell and others 2010). As reported by other investigators, Axtell and others (2010) show in vitro experiments that IFN-b decreases the early and late stages of Th17 differentiation. Therefore, the regulation of OPN by IFN-b may be of more importance in MS patients, or at certain stages of MS development, in which the Th17 response is dominant over the Th1 response. It is still unclear how IFN-b exacerbates Th17-mediated EAE. These results suggest, however, that the observed inhibition of Th17 cells by IFN-b may not account for its beneficial action in all cases of MS, in particular in those cases characterized by high concentrations of IL-17 in the serum. Conclusions and Future Prospects A better understanding of the mechanisms of action of IFN-b will help in the design of new therapeutic agents for the treatment of MS. Recent studies suggest that the downregulation of OPN and Th17 by IFN-b accounts for an important part of the mechanisms of action of IFN-b therapy in MS. The importance of the migration of inflammatory cells through the blood brain barrier in the pathogenesis of MS is highlighted by the role played by OPN in cell migration. This may be of particular importance if IFN-b exerts its action in

46 754 HONG AND HUTTON MMP-9 (+) DC IFN-b TH1 Differentiation TH1 Cytokines & Chemokines DC P STAT3 TH17? TH17 OPN IL-27 (+) Neutrophil OPN TH1 TH1 TH1 Apoptosis Periphery Blood Brain Barrier CNS FIG. 1. Role of interferon-b (IFN-b) in control of inflammatory Th1 and Th17 cells in multiple sclerosis. IFN-b modifies the immune responses largely in the periphery in multiple sclerosis patients due to the presence of the blood brain barrier. IFN-b affects the transmigration of Th1 cells across the blood brain barrier through inhibition of matrix metalloproteinase (MMP)-9 production by T cells and macrophages. The influence of IFN-b on Th17 cell migration is unclear. IFN-b may bind directly to its receptor on Th17 cells, resulting in inhibition of the phosphorylation of STAT3 and blockade of Th17 cell differentiation. IFN-b may also suppress expression of intracellular osteopontin (OPN) in DCs and subsequently upregulate interleukin-27 (IL-27) secretion by DCs. Elevated IL-27 in turn suppresses the differentiation of Th17 cells. Decreased OPN results in increased Th1 cell death through removal of the repression of apoptosis by OPN. both the CNS and peripheral compartments. The design of new type I interferon receptor (IFNAR) agonists that can permeate the blood brain barrier may be potentially beneficial. Indeed, in a recent study a mutation and screening approach has been used in an attempt to isolate improved IFNAR agonists (Langer 2007). It is conceivable that the identification of the critical amino acid residues and conformations of IFN-b may be of help in the development of a new generation of smaller sized IFN-b molecules that are able to achieve maximal benefit in the therapy of MS. The role of Th17 cells in the pathogenesis of MS has increasing been recognized. Many studies of immunomodulatory treatments for MS have begun to evaluate the effect of such treatments on the Th17 cellular response. It is plausible that IFN-b targets Th17 cells to block their differentiation or expansion, which is reflected by the downregulation of IL-17 expression. Current data indicate that suppression of IL-17 can be achieved through 2 pathways. IFN-b acts on its receptor on dendritic cells and inhibits intracellular OPN expression in DCs. Decreased expression of intracellular OPN in DCs subsequently upregulates expression of IL-27, which is an important Th17 inhibitor. On the other hand, IFN-b directly binds to its receptor on CD4 þ T cells and activates its signaling cascade. The activation of STAT1 by IFN-b reciprocally blocks the activation of STAT3 that is responsible for Th17 cell differentiation. Consequently, IFN-b could achieve its therapeutic effects through downregulating Th17 cells both directly and indirectly. Given the clinical effect of IFN-b, new agents that suppress the differentiation or expansion of Th17 could be valid options for MS therapy although a recent report suggested that IFN-b might not be effective in Th17-mediated EAE (Axtell and others 2010). The development of MS preferentially mediated by Th17 cells may not represent a majority of MS cases clinically since MS development usually involves both Th1 cells and Th17 cells. The anti-il-12 p40 antibody, ustekinumab, has been used for the treatment of MS. In a recent 37-week follow-up trial in relapsing-remitting-ms patients given ustekinumab or placebo, the study did not reach the primary endpoint of a significant reduction of the number of new gadoliniumenhancing lesions on a serial magnetic resonance imaging (Segal and others 2008). The IL-12 p40 chain is a common chain shared by IL-23 a cytokine that plays a critical role in the expansion of Th17 cells. The antibody blocks the expansion of Th17 cells. However, the failure of this trial does not exclude the notion that blockade of Th17 may be of use in

47 REGULATION OF OPN AND IL-17 BY IFN-b IN MS 755 the therapy of MS. Th17 or IL-17 gene knockout successfully delayed the onset of EAE in animal studies. Antibody is not usually able to attain access to the CNS. Therefore, Th17 cells in the CNS may not be affected by the antibody treatment. Taken together, the downregulation of OPN may be more significant than downregulation of IL-17 in IFN-b treatment of MS. Thus far, it appears that any new agent that could block the transmigration through the blood brain barrier will be beneficial. New agents that can simultaneously regulate Th1 and Th17 cytokines may provide to be of particular benefit. IFNAR is omnipresent in the body, and thus the overall biological effects of IFN-b are complex. It appears that IFN-b simultaneously exerts multiple beneficial effects in the treatment of MS. The design of new IFNAR agonists that can permeate the blood brain barrier could potentially be a useful approach to the therapy of MS. Author Disclosure Statement No competing financial interests exist. References Ashkar S, Weber GF, Panoutsakopoulou V, Sanchirico ME, Jansson M, Zawaideh S, Rittling SR, Denhardt DT, Glimcher MJ, Cantor H Eta-1 (osteopontin): An early component of type-1 (cell-mediated) immunity. Science 287(5454): Axtell RC, de Jong BA, Boniface K, van der Voort LF, Bhat R, De Sarno P, Naves R, Han M, Zhong F, Castellanos JG, Mair R, Christakos A, Kolkowitz I, Katz L, Killestein J, Polman CH, de Waal Malefyt R, Steinman L, Raman C T helper type 1 and 17 cells determine efficacy of interferon-beta in multiple sclerosis and experimental encephalomyelitis. Nat Med 16(4): Barna BP, Chou SM, Jacobs B, Yen-Lieberman B, Ransohoff RM Interferon-beta impairs induction of HLA-DR antigen expression in cultured adult human astrocytes. J Neuroimmunol 23(1): Boniface K, Blom B, Liu YJ, de Waal Malefyt R From interleukin-23 to T-helper 17 cells: Human T-helper cell differentiation revisited. Immunol Rev 226: Boz C, Ozmenoglu M, Velioglu S, Kilinc K, Orem A, Alioglu Z, Altunayoglu V Matrix metalloproteinase-9 (MMP-9) and tissue inhibitor of matrix metalloproteinase (TIMP-1) in patients with relapsing-remitting multiple sclerosis treated with interferon beta. Clin Neurol Neurosurg 108(2): Burdo TH, Wood MR, Fox HS Osteopontin prevents monocyte recirculation and apoptosis. J Leukoc Biol 81(6): Cao Z, Dai J, Fan K, Wang H, Ji G, Li B, Zhang D, Hou S, Qian W, Zhao J, Guo Y A novel functional motif of osteopontin for human lymphocyte migration and survival. Mol Immunol 45(14): Chabas D, Baranzini SE, Mitchell D, Bernard CC, Rittling SR, Denhardt DT, Sobel RA, Lock C, Karpuj M, Pedotti R, Heller R, Oksenberg JR, Steinman L The influence of the proinflammatory cytokine, osteopontin, on autoimmune demyelinating disease. Science 294(5547): Chen M, Chen G, Nie H, Zhang X, Niu X, Zang YC, Skinner SM, Zhang JZ, Killian JM, Hong J. 2009a. Regulatory effects of IFN-beta on production of osteopontin and IL-17 by CD4þ T Cells in MS. Eur J Immunol 39(9): Chen YJ, Wei YY, Chen HT, Fong YC, Hsu CJ, Tsai CH, Hsu HC, Liu SH, Tang CH. 2009b. Osteopontin increases migration and MMP-9 up-regulation via alphavbeta3 integrin, FAK, ERK, and NF-kappaB-dependent pathway in human chondrosarcoma cells. J Cell Physiol 221(1): Comabella M, Pericot I, Goertsches R, Nos C, Castillo M, Blas Navarro J, Rio J, Montalban X Plasma osteopontin levels in multiple sclerosis. J Neuroimmunol 158(1 2): Cua DJ, Sherlock J, Chen Y, Murphy CA, Joyce B, Seymour B, Lucian L, To W, Kwan S, Churakova T, Zurawski S, Wiekowski M, Lira SA, Gorman D, Kastelein RA, Sedgwick JD Interleukin-23 rather than interleukin-12 is the critical cytokine for autoimmune inflammation of the brain. Nature 421(6924): de Andres C, Aristimuno C, de Las Heras V, Martinez-Gines ML, Bartolome M, Arroyo R, Navarro J, Gimenez-Roldan S, Fernandez-Cruz E, Sanchez-Ramon S Interferon beta-1a therapy enhances CD4þ regulatory T-cell function: An ex vivo and in vitro longitudinal study in relapsingremitting multiple sclerosis. J Neuroimmunol 182(1 2): Diveu C, McGeachy MJ, Boniface K, Stumhofer JS, Sathe M, Joyce-Shaikh B, Chen Y, Tato CM, McClanahan TK, de Waal Malefyt R, Hunter CA, Cua DJ, Kastelein RA IL-27 blocks RORc expression to inhibit lineage commitment of Th17 cells. J Immunol 182(9): Fan K, Dai J, Wang H, Wei H, Cao Z, Hou S, Qian W, Li B, Zhao J, Xu H, Yang C, Guo Y Treatment of collagen-induced arthritis with an anti-osteopontin monoclonal antibody through promotion of apoptosis of both murine and human activated T cells. Arthritis Rheum 58(7): Galicia G, Kasran A, Uyttenhove C, De Swert K, Van Snick J, Ceuppens JL ICOS deficiency results in exacerbated IL- 17 mediated experimental autoimmune encephalomyelitis. J Clin Immunol 29(4): Gravallese EM Osteopontin: A bridge between bone and the immune system. J Clin Invest 112(2): Guo B, Chang EY, Cheng G The type I IFN induction pathway constrains Th17-mediated autoimmune inflammation in mice. J Clin Invest 118(5): Haak S, Croxford AL, Kreymborg K, Heppner FL, Pouly S, Becher B, Waisman A IL-17A and IL-17F do not contribute vitally to autoimmune neuro-inflammation in mice. J Clin Invest 119(1): Hall GL, Wing MG, Compston DA, Scolding NJ beta- Interferon regulates the immunomodulatory activity of neonatal rodent microglia. J Neuroimmunol 72(1): HiguchiY,TamuraY,UchidaT,MatsuuraK,HijiyaN,Yamamoto S The roles of soluble osteopontin using osteopontintransgenic mice in vivo: Proliferation of CD4þ T lymphocytes and the enhancement of cell-mediated immune responses. Pathobiology 71(1):1 11. Hofstetter HH, Ibrahim SM, Koczan D, Kruse N, Weishaupt A, Toyka KV, Gold R Therapeutic efficacy of IL-17 neutralization in murine experimental autoimmune encephalomyelitis. Cell Immunol 237(2): Hur EM, Youssef S, Haws ME, Zhang SY, Sobel RA, Steinman L Osteopontin-induced relapse and progression of autoimmune brain disease through enhanced survival of activated T cells. Nat Immunol 8(1): Jansson M, Panoutsakopoulou V, Baker J, Klein L, Cantor H Cutting edge: Attenuated experimental autoimmune encephalomyelitis in eta-1/osteopontin-deficient mice. J Immunol 168(5): Jiang H, Milo R, Swoveland P, Johnson KP, Panitch H, Dhib-Jalbut S Interferon beta-1b reduces interferon

48 756 HONG AND HUTTON gamma-induced antigen-presenting capacity of human glial and B cells. J Neuroimmunol 61(1): Komiyama Y, Nakae S, Matsuki T, Nambu A, Ishigame H, Kakuta S, Sudo K, Iwakura Y IL-17 plays an important role in the development of experimental autoimmune encephalomyelitis. J Immunol 177(1): Korporal M, Haas J, Balint B, Fritzsching B, Schwarz A, Moeller S, Fritz B, Suri-Payer E, Wildemann B Interferon betainduced restoration of regulatory T-cell function in multiple sclerosis is prompted by an increase in newly generated naive regulatory T cells. Arch Neurol 65(11): Kozovska ME, Hong J, Zang YC, Li S, Rivera VM, Killian JM, Zhang JZ Interferon beta induces T-helper 2 immune deviation in MS. Neurology 53(8): Langer JA Interferon at 50: New molecules, new potential, new (and old) questions. Sci STKE 2007(405):pe53. Lee WW, Kang SW, Choi J, Lee SH, Shah K, Eynon EE, Flavell RA, Kang I Regulating human Th17 cells via differential expression of IL-1 receptor. Blood 115(3): Leppert D, Waubant E, Burk MR, Oksenberg JR, Hauser SL Interferon beta-1b inhibits gelatinase secretion and in vitro migration of human T cells: A possible mechanism for treatment efficacy in multiple sclerosis. Ann Neurol 40(6): Liu X, Leung S, Wang C, Tan Z, Wang J, Guo TB, Fang L, Zhao Y, Wan B, Qin X, Lu L, Li R, Pan H, Song M, Liu A, Hong J, Lu H, Zhang JZ Crucial role of interleukin-7 in T helper type 17 survival and expansion in autoimmune disease. Nat Med 16(2): Lou J, Gasche Y, Zheng L, Giroud C, Morel P, Clements J, Ythier A, Grau GE Interferon-beta inhibits activated leukocyte migration through human brain microvascular endothelial cell monolayer. Lab Invest 79(8): Ma HL, Liang S, Li J, Napierata L, Brown T, Benoit S, Senices M, Gill D, Dunussi-Joannopoulos K, Collins M, Nickerson-Nutter C, Fouser LA, Young DA IL-22 is required for Th17 cellmediated pathology in a mouse model of psoriasis-like skin inflammation. J Clin Invest 118(2): Malik O, Compston DA, Scolding NJ Interferon-beta inhibits mitogen induced astrocyte proliferation in vitro. J Neuroimmunol 86(2): Martin-Saavedra FM, Gonzalez-Garcia C, Bravo B, Ballester S Beta interferon restricts the inflammatory potential of CD4(þ) cells through the boost of the Th2 phenotype, the inhibition of Th17 response and the prevalence of naturally occurring T regulatory cells. Mol Immunol 45(15): McLaurin J, Antel JP, Yong VW Immune and non-immune actions of interferon-beta-ib on primary human neural cells. Mult Scler 1(1): McRae BL, Picker LJ, van Seventer GA Human recombinant interferon-beta influences T helper subset differentiation by regulating cytokine secretion pattern and expression of homing receptors. Eur J Immunol 27(10): Murugaiyan G, Mittal A, Weiner HL Increased osteopontin expression in dendritic cells amplifies IL-17 production by CD4þ T cells in experimental autoimmune encephalomyelitis and in multiple sclerosis. J Immunol 181(11): Namdar A, Nikbin B, Ghabaee M, Bayati A, Izad M Effect of IFN-beta therapy on the frequency and function of CD4(þ)CD25(þ) regulatory T cells and Foxp3 gene expression in relapsing-remitting multiple sclerosis (RRMS): A preliminary study. J Neuroimmunol 218(1 2): Noronha A, Toscas A, Jensen MA Interferon beta decreases T cell activation and interferon gamma production in multiple sclerosis. J Neuroimmunol 46(1 2): Prinz M, Schmidt H, Mildner A, Knobeloch KP, Hanisch UK, Raasch J, Merkler D, Detje C, Gutcher I, Mages J, Lang R, Martin R, Gold R, Becher B, Bruck W, Kalinke U Distinct and nonredundant in vivo functions of IFNAR on myeloid cells limit autoimmunity in the central nervous system. Immunity 28(5): Rep MH, Hintzen RQ, Polman CH, van Lier RA Recombinant interferon-beta blocks proliferation but enhances interleukin-10 secretion by activated human T-cells. J Neuroimmunol 67(2): Rudick RA, Ransohoff RM, Peppler R, VanderBrug Medendorp S, Lehmann P, Alam J Interferon beta induces interleukin-10 expression: Relevance to multiple sclerosis. Ann Neurol 40(4): Segal BM, Constantinescu CS, Raychaudhuri A, Kim L, Fidelus- Gort R, Kasper LH Repeated subcutaneous injections of IL12/23 p40 neutralising antibody, ustekinumab, in patients with relapsing-remitting multiple sclerosis: A phase II, double-blind, placebo-controlled, randomised, dose-ranging study. Lancet Neurol 7(9): Shinohara ML, Kim JH, Garcia VA, Cantor H Engagement of the type I interferon receptor on dendritic cells inhibits T helper 17 cell development: Role of intracellular osteopontin. Immunity 29(1): Steinman L Multiple sclerosis: A two-stage disease. Nat Immunol 2(9): Stromnes IM, Goverman JM Osteopontin-induced survival of T cells. Nat Immunol 8(1): Stuve O, Chabot S, Jung SS, Williams G, Yong VW Chemokine-enhanced migration of human peripheral blood mononuclear cells is antagonized by interferon beta-1b through an effect on matrix metalloproteinase-9. J Neuroimmunol 80(1 2): Stuve O, Dooley NP, Uhm JH, Antel JP, Francis GS, Williams G, Yong VW Interferon beta-1b decreases the migration of T lymphocytes in vitro: Effects on matrix metalloproteinase-9. Ann Neurol 40(6): Sugimoto K, Ogawa A, Mizoguchi E, Shimomura Y, Andoh A, Bhan AK, Blumberg RS, Xavier RJ, Mizoguchi A IL-22 ameliorates intestinal inflammation in a mouse model of ulcerative colitis. J Clin Invest 118(2): Tzartos JS, Friese MA, Craner MJ, Palace J, Newcombe J, Esiri MM, Fugger L Interleukin-17 production in central nervous system-infiltrating T cells and glial cells is associated with active disease in multiple sclerosis. Am J Pathol 172(1): Vogt MH, Lopatinskaya L, Smits M, Polman CH, Nagelkerken L Elevated osteopontin levels in active relapsingremitting multiple sclerosis. Ann Neurol 53(6): Weber GF, Cantor H The immunology of Eta-1/osteopontin. Cytokine Growth Factor Rev 7(3): Wolk K, Witte E, Wallace E, Docke WD, Kunz S, Asadullah K, Volk HD, Sterry W, Sabat R IL-22 regulates the expression of genes responsible for antimicrobial defense, cellular differentiation, and mobility in keratinocytes: A potential role in psoriasis. Eur J Immunol 36(5): Yushchenko M, Mader M, Elitok E, Bitsch A, Dressel A, Tumani H, Bogumil T, Kitze B, Poser S, Weber F Interferon-beta- 1 b decreased matrix metalloproteinase-9 serum levels in primary progressive multiple sclerosis. J Neurol 250(10):

49 REGULATION OF OPN AND IL-17 BY IFN-b IN MS 757 Zhang X, Jin J, Tang Y, Speer D, Sujkowska D, Markovic-Plese S IFN-beta1a inhibits the secretion of Th17-polarizing cytokines in human dendritic cells via TLR7 up-regulation. J Immunol 182(6): Zuvich RL, McCauley JL, Oksenberg JR, Sawcer SJ, De Jager PL, Aubin C, Cross AH, Piccio L, Aggarwal NT, Evans D, Hafler DA, Compston A, Hauser SL, Pericak-Vance MA, Haines JL Genetic variation in the IL7RA/IL7 pathway increases multiple sclerosis susceptibility. Hum Genet 127(5): Address correspondence to: Dr. Jian Hong Department of Neurology Baylor Multiple Sclerosis Center Baylor College of Medicine Mail station NB302 One Baylor Plaza Houston, TX jhong@bcm.tmc.edu Received 9 July 2010/Accepted 9 July 2010

50

51 JOURNAL OF INTERFERON & CYTOKINE RESEARCH Volume 30, Number 10, 2010 ª Mary Ann Liebert, Inc. DOI: /jir Critical Review: Assessment of Interferon-b Immunogenicity in Multiple Sclerosis Klaus Bendtzen This review discusses type I interferon (IFN) immunogenicity with focus on methods of detection of anti-ifn antibodies in patients treated with human recombinant IFN-b. Pitfalls involved in the clinical use of various types of assays for binding antibodies and neutralizing antibodies against IFN-b are presented, and the widely held distinction between binding antibodies and neutralizing antibodies is questioned both in terms of detection and clinical importance. The article also addresses important bioavailability and pharmacokinetic issues occurring with prolonged use of protein drugs. The rationale for individualized or personalized medicine, ie, optimizing therapies according to individual needs rather than using standardized trial-and-error regimens to all patients, is highlighted. Interferons Natural interferons (IFNs) constitute a heterogeneous group of proteins (Pestka and others 2004). They may be divided into 3 main subgroups based on structural and functional differences: the type I IFNs (IFN-a/IFN-b), the type II IFN (IFN-g), and the type III IFN (IFN-l-1, -2, and -3, also named interleukin (IL)-29, IL-28A, and IL-28B). IFN-b is produced primarily by virus-infected fibroblasts and some epithelial cells. It consists of a subgroup of at least 2 members of kda glycoproteins called IFN-b1 and IFN-b3 (IFN-b2, also known as IL-6, is structurally different from this subgroup). In contrast to IFN-a, IFN-b is species specific in that human cells do not respond to IFN-b of other species. An essential function of IFN-b is its antiviral activity, which affects almost all cell types infected with a broad spectrum of viruses. Additionally, and perhaps more important for its effect in multiple sclerosis (MS), IFN-b has antiproliferative and immunomodulatory functions (Pestka and others 2004). MS is a chronic immunoinflammatory disease that causes gradual destruction of myelin sheaths around axons in the brain and spinal cord, leading to demyelination and scarring with plaque formation in patches throughout the central nervous system (Hafler and others 2005). Certain stages of MS, particularly the relapsing-remitting form of the disease, can be treated by the administration of human IFN-b (The IFNB Multiple Sclerosis Study Group 1993; Ann Marrie and Rudick 2006). Immunogenicity of IFN-b Naturally occurring autoantibodies (aabs) to type I IFNs, primarily IFN-a, have been reported in both patients and normal individuals (Bendtzen and others 1990, 1998; Meager 1997; Meager and others 2003; Bendtzen and Svenson 2007). These aabs occur with a frequency of up to 10% in apparently healthy individuals and in practically all therapeutic IgG preparations formulated from normal human donors, and in up to 70% of patients suffering from various diseases, eg, thymoma/myasthenia gravis (Bendtzen and others 2000; Meager and others 2003). Unlike the relatively common occurrence of aabs to IFN-a, aabs to IFN-b have been reported in less than 0.1% of healthy individuals, and their clinical relevance is unknown (Bendtzen and Svenson 2007). Type I IFNs have been used as therapies in patients suffering from many different diseases, with varying clinical success (Pestka and others 2004). Today, genetically engineered IFNs are used primarily in the treatment of patients with MS, hepatitis B and C, malignant melanoma, neuroendocrine tumors, and certain lymfoproliferative and hematological diseases, including diseases characterized by thrombocytosis. The use of IFN-b is almost exclusively restricted to patients with MS. Two human recombinant IFN-b species are currently registered for this use: IFN-b-1b expressed in Escherichia coli and IFN-b-1a expressed in Chinese hamster ovary cells. The exact pathomechanism behind the beneficial function of IFN-b in MS is unknown. Patients are not cured by this form of therapy, but both types of IFN-b halt MRI plaque formation and reduce attack frequency and severity. Consequently, treatment with IFN-b is long lasting, usually several years unless development of anti-ifn-b antibody (Ab) causes relapses or side effects. As all proteins are potentially immunogenic, especially if administered similarly to vaccines using repeated subcutaneous injections of more or less aggregated protein. Like Institute for Inflammation Research (IIR), Rigshospitalet University Hospital, Copenhagen, Denmark. 759

52 760 BENDTZEN vaccines, biopharmaceuticals may therefore lead to both T lymphocyte- and B lymphocyte-mediated reactions in the recipients. This was previously considered a minor issue by many investigators. Fortunately, this has changed and immunogenicity is now in focus from drug development to postmarketing studies, and monitoring immune responses to therapeutic proteins, primarily by detection and quantification of anti-drug Abs (ADA), are now slowly being implemented into everyday clinical practice (Koren and others 2008; Bendtzen and others 2009; Bertolotto 2009). Known consequences of drug immunogenicity range from transient appearance of low titers of ADA with little clinical impact to both acute and delayed immune reactions, some of which with life-threatening consequences. It is also increasingly clear that ADA are important factors underlying the dose-creep phenomenon, ie, the observed need for increased dosage and/or more frequent administration of protein therapeutics in the course of prolonged therapies. In MS and in other chronic diseases, this often culminates in a condition where the clinical effect of continued injection of protein therapeutics is abrogated (Neumann and Foote 2000; Casadevall and others 2002; Pipe 2005; Sorensen and others 2006; Hyrich and others 2007; Strand and others 2007; Rutgeerts and others 2010). MS patients treated with recombinant human IFN-b may develop anti-ifn-b ADA. However, the reported frequencies and titers of these Abs vary considerably depending upon the IFN-b species and how the drugs are administered, but also on the types of assays being used (Ross and others 2000, 2006; Bendtzen 2003; Hartung and others 2007; Bendtzen and Kromminga 2008). Even though the appearance of neutralizing antibodies (NAbs) against IFN-b-1b was noted in the first pivotal trial in a substantial fraction of patients (The IFNB Multiple Sclerosis Study Group 1993), it took 10 years before Ab-mediated decrease in bioactivity of IFN-b was universally recognized (Pachner 2003; Sorensen and others 2003; Francis and others 2005; Hartung and others 2007). Indeed, regular screening of patients for ADA against IFN-b and other biotechnology-derived therapeutics is now advocated by regulatory authorities such as the European Medicines Agency (EMEA) (EMEA 2007). Much of the delay in recognizing the importance of drug immunogenicity in relapsing-remitting MS patients can be ascribed to the variable nature of the disease, making it difficult to temporally associate ADA development with therapeutic failure. The widely held dogma that the immune system is tolerant to polypeptides with amino acid sequences identical or almost identical to those of naturally occurring counterparts has probably contributed as well, as has the findings that anti-ifn-b ADA may persist after cessation of therapy and that ADA may disappear in a minor proportion of MS patients despite continued long-term therapies (Sorensen and others 2005; Petersen and others 2006; Polman and others 2006). It is also increasingly clear that testings not only for immunogenicity but also for bioavailability of protein drugs may prevent prolonged use of inadequate therapeutic regimens. For example, patients may show large inter- and even intraindividual differences in bioavailability of locally injected biopharmaceuticals, particularly if self-administered. Drug pharmacokinetics may also differ considerably from patient to patient. Awareness of these problems is rising with the concept of individualized or personalized medicine, which in this context is directed at optimizing therapies according to individual needs rather than using standardized regimens deducted from a few pivotal trials. These investigations, usually carried out with one or only a few dose regimens and in large cohorts of patients, are not suited to reveal differences in therapeutic responses in individual patients, including the impact of sex and age differences, comorbidities, immunocompetence of recipients, and variations in concurrent therapies (Bendtzen and others 2009). We have shown that ADA detected by fluid-phase radioimmunoassay (RIA), ie, IFN-b-binding Abs (BAbs), develops within 6 months of therapy in up to 90% of MS patients treated with IFN-b-1b, and that 90% 100% of these BAbs also neutralize IFN-b bioactivity (NAbs), provided they are assessed by sufficiently sensitive bioassays (Ross and others 2000, 2006; Bendtzen 2003). The frequency and the clinical relevance of these ADA depend of course on the nature of the drug, but treatment characteristics such as dosage and mode of administration are also essential factors (Ross and others 2000, 2006). Not surprisingly, immunogenicity is greatest for E. coli derived IFN-b-1b, which is nonglycosylated and has 2 amino acid differences from wildtype IFN-b, and subcutaneous administration is more immunogenic than intramuscular delivery. ADA generated in patients treated with IFN-b-1b crossreact with IFN-b-1a and vice versa but not with IFN-a, and they crossreact as aabs with wild-type IFN-b. Methods for Detection of Anti-IFN-b Abs There are significant difficulties in obtaining reliable methods for monitoring patients on prolonged IFN-b therapies. The World Health Organization has recommended the use of antiviral neutralization bioassays in which neutralization of IFN-b-induced antiviral activity is expressed as a titer defined as the reciprocal of the serum dilution that reduces the IFN potency from 10 to 1 laboratory units (LU)/ ml, where 1 LU/mL is the level of IFN-b inducing 50% protection against challenge virus in a given assay (Grossberg 2003). EMEA recently recommended the use of an enzyme-linked immunosorbent assay (ELISA) for intracellular myxovirus resistance protein A (MxA), a surrogate marker of IFN-b activity (EMEA 2008) (see below). Assays for Abs Binding to IFN-b Currently available methods for detection of BAbs to IFNb include ELISAs, RIAs, and various surface-binding techniques (Ross and others 2000; Pachner 2003; Hartung and others 2007). The most commonly used type of BAb assay is ELISA in various modifications with or without the use of enhancement technologies, eg, chemiluminescence to increase sensitivity (Golgher and others 1999). The major advantage is the simplicity and relative low cost. However, since BAbs bind to both conformational and sequence-restricted (linear) epitopes on proteins, surface-binding technologies using more or less denatured proteins in the capture phase may hamper detection of BAbs restricted to the native conformation of the molecules (Fig. 1, upper panel) (Bendtzen 2003). Unless special precautions are made, these techniques may also reveal BAbs even in cases where Ab affinity is low and therefore of questionable clinical

53 ASSESSMENT OF IFN-b IMMUNOGENICITY IN MS 761 FIG. 1. Solid-phase assays for binding Abs against IFN-b. Basic principles of often-used ELISAs for detection of ADA, including potential causes of false-positive and false-negative test outcomes (red shaded areas). The upper panel depicts a sandwich ELISA, whereas the lower panel shows the principle underlying bridging or double-antigen ELISA, which is dependent upon functional bivalency of the ADA being measured. As IgG4 Abs are usually bispecific due to exchange of half molecules during synthesis, this subclass of anti-ifn-b ADA escapes detection. Ab, antibody; ADA, antidrug antibodies; ELISA, enzyme-linked immunosorbent assay; IFN, interferon. significance. As this binding is often nonsaturable and hence nonspecific, such Abs should not be categorized as true BAbs. Demonstration of ligand binding to the Fab fragments of serum immunoglobulins, combined with saturation binding analysis and fluid-phase demonstration of crossbinding to the drug, is therefore advocated to verify the presence of true specific BAbs against IFN-b. Other causes of false-positives include detection of Abs to neoepitopes caused by drug aggregation. ELISAs may also generate falsenegative results, eg, because of high background values and matrix effects which may mask epitopes that have triggered Ab production in vivo (Fig. 1). A relatively new approach to detect BAbs is the use of bridging- or double-antigen ELISAs. This format uses immobilized IFN-b as catching reagent and an enzyme-labeled IFN-b as detecting antigen (Fig. 1, lower panel). BAbs in patient samples are thought to recognize both immobilized and dissolved antigen due to their bivalent binding characteristics. A bridging ELISA should therefore combine the high-throughput need of a screening assay and the specificity of a (semi)fluid-phase assay. Unfortunately, the use of bridging ELISAs for BAb measurements is clinically unsound because this technology fails to detect monovalent BAbs against IFN-b, eg, Abs of the IgG4 isotype (Fig. 1). Such Abs predominate in NAb-positive MS patients after 12 months of immunization with IFN-b (Gibbs and Oger 2008). Since IgG4 Abs may impede drug absorption, increase drug clearance, prevent the drug from reaching affected tissues, and neutralize IFN-b binding to their cellular receptors (see Fig. 2), failure to detect this class of Abs makes bridging ELISAs ill-suited as a basis for therapeutic decisions in individual patients.

54 762 BENDTZEN Because of the abovementioned problems with solidphase assays, a fluid-phase RIA for direct binding to 125 I- labeled IFN-b may better reflect the in vivo situation. This test procedure has been detailed previously, and clinically validated in MS patients (Ross and others 2000, 2006). There are several reasons for this: The tracer preserves as closely as possible the natural configuration of IFN-b and allows detection of BAbs directed against conformational epitopes, it can be validated and tested for retained bioactivity, it allows high assay sensitivity, and the specificity, kinetics, avidity, and capacity of the IFN-b BAb binding may be assessed, as can crossbinding to non-ab factors in serum. However, denaturation may occur when iodinating proteins, and one should therefore purify the tracer and validate it for specific activity, stability, and immunoreactivity before use. It is of course a drawback with all isotope techniques that they require specialized facilities and personnel. Surface plasmon resonance biosensors are optical sensors that use electromagnetic waves to probe interactions between an analyte in solution, eg, a BAb, and a molecular recognition element immobilized on the sensor surface, eg, human recombinant IFN-b (Kaliyaperumal and Jing 2009). The use of biosensors is particularly useful for real-time determination of binding kinetics. Although the method of choice for characterization of the nature of the antigen Ab interaction, this method requires costly equipment, and low throughput and relatively low sensitivity also make it unsuited for routine clinical use. It is a general requirement for all BAb assays that steps should be included to control for saturability of binding and for factors other than BAbs that may affect binding to IFN-b. Indeed, it is important to note that many serum proteins may influence BAb assays, including ELISAs and RIAs, eg, soluble receptors and preexisting IFN-a and IFN-b. BAbs in immune complexes may also escape detection because of a low exchange rate between BAb-bound unlabeled and labeled IFN-b during incubation, or if BAbs form complexes with high dissociation rates, allowing them to dissociate before assay readout. Assays for Abs Neutralizing IFN-b Bioactivity These include antiviral neutralization bioassays measured in a cytopathic effect (CPE) setup, and tests for inhibition of production of second mediators of IFN-b, eg, the induction of MxA mrna or MxA protein. Recently, reporter-gene assays have been developed for IFN-b NAb determination, offering interesting possibilities both as high-throughput screening assays and as investigational tools for drug development. The CPE assay has been the most commonly used bioassay for IFN-b NAbs and has been clinically validated in MS patients (Ross and others 2000, 2006; Sorensen and others 2003; Pachner and others 2005). The test is usually carried out in one of 2 modifications: (1) neutralization capacity assay, sometimes termed constant Ab assay, and (2) an assay yielding a titer calculated as the serum dilution that reduces IFN potency from 10 to 1 LU/mL, sometimes termed constant antigen assay (Ross and others 2000, 2006; Grossberg and others 2001, 2009). Both assays measure the ability of NAb-containing serum samples to counteract the protective effect of IFN-b when an IFN-b-sensitive cell line is exposed to a cytopathic virus. The human lung cancer line A549 is the most frequently used cell line, in combination with vesicular stomatitis virus or encephalomyocarditis virus, and with the use of 3-(4,5 dimethyl thiazol-2-yl) 2,5-difenyl tetrazolium bromide or other colorimetric tests for cell viability. The neutralization capacity assay quantifies the capacity of a NAb-containing test sample to inhibit the antiviral activity of IFN-b (Ross and others 2006). There are several advantages of this test compared to titer determinations. A practical one is that it eliminates the need for measurements of serially diluted samples, making it cheaper and better suited for screening and routine use. More importantly, however, this test better mimics the in vivo situation in the blood, where a relatively fixed level of NAbs neutralize small amounts of IFN-b gradually released into the circulation from subcutaneous or intramuscular injection sites. The 10 to 1 LU/mL neutralization assay, often called Kawade assay, is also a semiquantitative test for NAbs against type I IFN, where serially diluted serum samples are tested for IFN-b neutralization in an otherwise similar CPE setup (Grossberg 2003; Ross and others 2006). The most important advantage of this test is the ability to quantify NAbs at very high levels and, thus, the ability to follow changes in (high) NAb titers over time that may have prognostic and therapeutic implications. A potential problem with titer determinations of NAbs using this assay is that antigen Ab complexes are formed in a scenario far from the in vivo situation. In patients with high-titered NAbs, immune complexes would form in excess of Ab in the circulation, whereas formation of immune complexes in the Kawade end-point test would take place in antigen excess.thisisbecausenabsaredilutedexcessivelyinthe assay, frequently several thousand times, whereas the IFNb level is fixed at 10 LU/mL. This artificial situation has an impact on both size and solubility of the IFN-b Ab complexes. To what extent this affects the interpretation of NAb levels in vivo, and hence the clinical relevance of this method, has not been addressed. MxA assays are carried out in 2 major modifications (Fig. 3) (Pachner and others 2005). One assay makes use of an IFNsensitive cell line, eg, A549 cells, and measures changes in IFN-b-induced MxA expression in the presence of NAbs, using either real-time polymerase chain-reaction for quantification of MxA transcripts or ELISA for assessment of intracellular MxA protein (Fig. 3 left). The latter modification has recently been advocated by EMEA as a standardized test for IFN-b NAbs in MS patients (EMEA 2008). The second format of the MxA assay analyzes patient whole blood for leucocyte MxA mrna or MxA protein (Fig. 3 right). The advantages of this assay are that it directly measures cellular consequences of IFN-b bioactivity in individual patients. Unfortunately, uncertainties concerning optimal sample collection, which may vary from patienttopatientandeveninthesamepatientatdifferent time points, and interference from intercurrent diseases, eg, virus infections, and other therapies, eg, glucocorticoids, which interfere with MxA transcription and translation, makes this type of assay problematic in the clinical setting. The principle underlying reporter-gene assays is to replace an IFN-sensitive gene with one that is relatively easy to measure and quantify. As shown in Fig. 2 (left panel), IFN-b signals through IFN-a receptors (IFNARs) resulting in acti-

55 ASSESSMENT OF IFN-b IMMUNOGENICITY IN MS 763 FIG. 2. Reporter-gene assay for neutralizing anti-ifn-b antibodies. The left panel shows the pathways activated by IFN-b in IFNAR-positive cells resulting in activation of ISRE in the cell nucleus and subsequent transcription of &130 genes, including MxA and ISG15. In this example, cells suitable as reporters of type I IFN activity are generated by gene technology in that ISRE; eg, ISG15 is cloned upstream of a reporter gene such as the luciferase gene Luc. After exposing the cells to IFN-b, Luc is activated and translated into luciferase, an enzyme that catalyses a reaction with luciferin to produce quantifiable light emission. The right panel illustrates the effect of IgG1 and IgG4 NAbs in a Luc reporter-gene assay. JAK1, Janus kinase 1; ISGF3, interferon-stimulated gene factor 3 (which is the type I IFN-dependent transcription factor that is assembled in the cytoplasm and subsequently translocates to the nucleus and binds to ISRE); ISRE, interferon-stimulated response element; ISG15, interferon-stimulated gene 15; Luc, luciferase gene; NAb, neutralizing antibody; TYK2, tyrosine kinase 2 (see also legends of Figs. 1 and 3). vation of the IFN-sensitive responsive element, which then activates a host of genes governing antiviral, immune, and proliferative responses, MxA being one of these genes. Using genetic engineering techniques, IFN-sensitive responsive element for MxA or other indigenous genes such as ISG15 may be cloned upstream of a gene, eg, the luciferase gene, which under control of a relevant promoter may be used as a reporter-gene yielding quantitative responses in IFNARpositive cell lines. This may be used for easy assessments of type I IFN activities, and since NAbs interfere with IFN-b binding to IFNAR, appropriately modified reporter-gene assays may also be used to detect IFN-b NAbs (Farrell and others 2008; Lallemand and others 2008; Lam and others 2008). Thus, a newly developed reporter-gene IFN-b NAb assay produced under quality certification (ISO13485) and using growth-arrested cells exhibit high reproducibility, accuracy, specificity, and insensitivity to interfering serum factors (toxic and antiviral), which may interfere with both CPE and MxA assays (Lallemand and others 2008; Ravnborg and others 2009). The most recent development is a one-step assay that allows both drug activity and drug NAbs to be quantified rapidly and with a high degree of precision simply be adding reporter cells to a sample (Lallemand and others, 2010). Assay Characteristics of Clinical Relevance There is an obvious need for clinically useful, inexpensive, and widely accepted/standardized screening assays for anti-ifn-b ADA in MS patients, but attempts in this direction have so far failed. Major reasons for this are the inconvenience and high cost of carrying out elaborate bioassays that require specialized laboratories and expert personnel plus the fact that assays must be clinically validated to provide useful information to therapists. Hence, many investigators favor the use of binding assays, at least as a screening test for BAbs, unfortunately often without due consideration to the potential to yield false-positive and false-negative results. This has contributed to the widely held assumption that there is reason to distinguish between BAbs and NAbs in patients treated with IFN-b, evenwhen such distinction is based on false data obtained from clinically inappropriate methodologies. A few examples might illustrate the issue: 1. Discrimination between BAbs and NAbs may reflect a quantitative rather than a qualitative difference in the Abs being tested. Development of NAbs in vivo is a dynamic process where initially produced NAbs may not reveal their neutralizing potential simply because of

56 764 BENDTZEN In vitro MxA assay In vivo MxA assay IFN -/+ NAb 1) Inj. of IFN-beta 2) Blood sample after 12 hours 10 LU/ml of IFN-beta Cells are lysed 4 hours IFNAR+ cell line Isolation of RNA Single strain cdna Testings: Quantitative RT-PCR for MxA mrna: - Sensitive measure of biological response to type 1 IFN - Reduced response in the presence of NAb Major caveats: - Specificity issues - Influenced by factors other than IFN-beta T- and B cell activators and inhibitors - Requires special equipment - Clinical validation in MS still lacking - Unsuited for mass screening a) Quantitative RT-PCR for MxA mrna or b) ELISA for MxA protein level - Individual variations in time of peak response - Unknown influence of diseases/infections - Unknown influence of concomitent therapies - All problems associated with the in vitro assay FIG. 3. MxA assays for neutralizing anti-ifn-b antibodies. The left panel illustrates the in vitro bioassay using type I IFNreceptor-positive cells. After cell lysis, the IFN-b-induced MxA transcripts are quantified using RT-PCR. The right panel shows the in vivo bioassay, where MxA mrna in patient leucocytes is extracted and quantified using RT-PCR (a). Alternatively, MxA protein may be extracted and quantified by ELISA specific for MxA protein (b). IFNAR, interferon-a receptor; LU, laboratory units; MxA, myxovirus resistance protein A; RT-PCR, real-time polymerase chain reaction (see also legend of Fig. 1). their low level. Although not yet detectable in NAb assays, they may be detected in a sensitive BAb assay giving the false assumption that they are nonneutralizing BAbs (Gneiss and others 2008). Although not yet having a major effect on therapeutic efficacy, such Abs may, nevertheless, neutralize portions of subsequently administered drug. These Abs also signal further NAb development if therapy continues, especially because Ab maturation and increased polyclonality contribute to drug neutralization (Pachner and others 2005; Gibbs and Oger 2008). 2. Determination of false-positive BAbs, occasionally experienced with the use of solid-phase assays, would be interpreted as nonneutralizing BAbs. The tendency to refer to such findings as if these Abs were true anti-ifn-b ADA and not the result of inaccurate methodologies is an unfortunate confounding factor. 3. If nonneutralizing BAbs do indeed exist in IFN-b-treated patients, they may still have clinically important pharmacological effects. Nonneutralizing BAbs may, eg, inhibit absorption of IFN-b from subcutaneous or intramuscular sites, thus decreasing bioavailability. Such BAbs may also reduce the circulating half-life of IFN-b and/or prevent the drug from reaching pathologically affected tissues, eg, in the brain and spinal cord. Immune complexes containing BAbs and complement may also cause side effects whether or not IFN-b bioactivity is neutralized, eg, Arthus reactions at injection sites and/or cytotoxicity through Ab-dependent cell-mediated injury or complement-mediated attack initiated by Abs binding to IFN-b associated to cells producing or responding to the cytokine. Hence, the trend to consider NAbs only of clinical importance in IFN-b-treated MS patients is not warranted (Bendtzen 2003). Finally, it is worth noticing that there is an unfortunate tendency to report drug immunogenicity as a percentage of ADA-positives in a given cohort of patients without critical approach to the methodology used for Ab determination. For example, the use of insensitive assays, or assays known to yield false-negative results would in this context signal low drug immunogenicity. It is also problematic that drug immunogenicity is often reported without specifying the time point where the prevalency of ADA-positives was determined. This is particularly relevant when comparing the immunogenicity of different drugs, because the number of ADA-positives increases over time at the early stages of immunization. Hence, longitudinal studies to assess immunogenicity are warranted instead of single-time point determinations of ADA-positive patients. Author Disclosure Statement The author owns stocks in Biomonitor A/S.

57 ASSESSMENT OF IFN-b IMMUNOGENICITY IN MS 765 References Ann Marrie R, Rudick RA Drug insight: interferon treatment in multiple sclerosis. Nat Clin Pract Neurol 2: Bendtzen K Anti-IFN BAb and NAb antibodies: a mini review. Neurology 61 (Suppl. 5):S6 S10. Bendtzen K, Ainsworth M, Steenholdt C, Thomsen OO, Brynskov J Individual medicine in inflammatory bowel disease: monitoring bioavailability, pharmacokinetics and immunogenicity of anti-tumour necrosis factor-alpha antibodies. Scand J Gastroenterol 44: Bendtzen K, Hansen MB, Ross C, Svenson M High-avidity autoantibodies to cytokines. Immunol Today 19: Bendtzen K, Hansen MB, Ross C, Svenson M Detection of autoantibodies to cytokines. Mol Biotechnol 14: Bendtzen K, Kromminga A Immunogenicity of interferonbeta. In: van de Weert M, Møller EH, eds. Immunogenicity of Biopharmaceuticals, Vol. viii. New York: Springer, pp Bendtzen K, Svenson M Cytokine autoantibodies. In: Shoenfeld Y, Meroni PL, Gershwin ME, eds. Autoantibodies. Elsevier Press, pp Bendtzen K, Svenson M, Jønsson V, Hippe E Autoantibodies to cytokines friends or foes? Immunol Today 11: Bertolotto A Implications of neutralising antibodies on therapeutic efficacy. J Neurol Sci 277 (Suppl. 1):S29 S32. Casadevall N, Nataf J, Viron B, Kolta A, Kiladjian JJ, Martin- Dupont P, Michaud P, Papo T, Ugo V, Teyssandier I, Varet B, Mayeux P Pure red-cell aplasia and antierythropoietin antibodies in patients treated with recombinant erythropoietin. N Engl J Med 346: European Medicines Agency (EMEA) Guideline on immunogenicity assessment of biotechnology-derived therapeutic proteins. Scientific_guideline/2009/09/WC pdf pp European Medicines Agency (EMEA) Biologics working party report to the CHMP. Beta-interferons and neutralising antibodies (in multiple sclerosis). en_gb/document_library/report/2009/11/wc pdf pp 1 3. Farrell R, Kapoor R, Leary S, Rudge P, Thompson A, Miller D, Giovannoni G Neutralizing anti-interferon beta antibodies are associated with reduced side effects and delayed impact on efficacy of interferon-beta. Mult Scler 14: Francis GS, Rice GP, Alsop JC Interferon beta-1a in MS: results following development of neutralizing antibodies in prisms. Neurology 65: Gibbs E, Oger J A biosensor-based characterization of the affinity maturation of the immune response against interferon-beta and correlations with neutralizing antibodies in treated multiple sclerosis patients. J Interferon Cytokine Res 28: Gneiss C, Brugger M, Millonig A, Fogdell-Hahn A, Rudzki D, Hillert J, Berger T, Reindl M, Deisenhammer F Comparative study of four different assays for the detection of binding antibodies against interferon-beta. Mult Scler 14: Golgher RR, Redlich PN, Totti DO, Grossberg SE Quantitative liquid-phase chemiluminescence ELISA: Detection of subtle epitope differences in huifn-beta. J Interferon Cytokine Res 19: Grossberg SE Perspectives on the neutralization of interferons by antibody. Neurology 61:S21 S23. Grossberg SE, Kawade Y, Grossberg LD The neutralization of interferons by antibody III. The constant antibody bioassay, a highly sensitive quantitative detector of low antibody levels. J Interferon Cytokine Res 29: Grossberg SE, Kawade Y, Kohase M, Klein JP The neutralization of interferons by antibody II. Neutralizing antibody unitage and its relationship to bioassay sensitivity: the tenfold reduction unit. J Interferon Cytokine Res 21: Hafler DA, Slavik JM, Anderson DE, O Connor KC, De Jager P, Baecher-Allan C Multiple sclerosis. Immunol Rev 204: Hartung H-P, Polman C, Bertolotto A, Deisenhammer F, Giovannoni G, Havrdova E, Hemmer B, Hillert J, Kappos L, Kieseier B, Killestein J, Malcus C, Comabella M, Pachner A, Schellekens H, Sellebjerg F, Selmaj K, Sorensen PS Neutralising antibodies to interferon beta in multiple sclerosis: expert panel report. J Neurol 254: Hyrich KL, Lunt M, Watson KD, Symmons DP, Silman AJ Outcomes after switching from one anti-tumor necrosis factor alpha agent to a second anti-tumor necrosis factor alpha agent in patients with rheumatoid arthritis: results from a large UK National Cohort Study. Arthritis Rheum 56: The IFNB Multiple Sclerosis Study Group Interferon beta- 1b is effective in relapsing-remitting multiple sclerosis. I. Clinical results of a multicenter, randomized, double- blind, placebo-controlled trial. Neurology 43: Kaliyaperumal A, Jing S Immunogenicity assessment of therapeutic proteins and peptides. Curr Pharm Biotechnol 10: Koren E, Smith HW, Shores E, Shankar G, Finco-Kent D, Rup B, Barrett YC, Devanarayan V, Gorovits B, Gupta S, Parish T, Quarmby V, Moxness M, Swanson SJ, Taniguchi G, Zuckerman LA, Stebbins CC, Mire-Sluis A Recommendations on risk-based strategies for detection and characterization of antibodies against biotechnology products. J Immunol Methods 333:1 9. Lallemand C, Meritet JF, Blanchard B, Lebon P, Tovey MG One-step assay for quantification of neutralizing antibodies to biopharmaceuticals. J Immunol Methods 356: Lallemand C, Meritet JF, Erickson R, Grossberg SE, Roullet E, Lyon-Caen O, Lebon P, Tovey MG Quantification of neutralizing antibodies to human type I interferons using division-arrested frozen cells carrying an interferon-regulated reporter-gene. J Interferon Cytokine Res 28: Lam R, Farrell R, Aziz T, Gibbs E, Giovannoni G, Grossberg S, Oger J Validating parameters of a luciferase reporter gene assay to measure neutralizing antibodies to IFNbeta in multiple sclerosis patients. J Immunol Methods 336: Meager A Natural autoantibodies to interferons. J Interferon Cytokine Res 17 (Suppl. 1):S51 S53. Meager A, Wadhwa M, Dilger P, Bird C, Thorpe R, Newsom- Davis J, Willcox N Anti-cytokine autoantibodies in autoimmunity: preponderance of neutralizing autoantibodies against interferon-alpha, interferon-omega and interleukin-12 in patients with thymoma and/or myasthenia gravis. Clin Exp Immunol 132: Neumann TA, Foote M Megakaryocyte growth and development factor (MGDF): an MPL ligand and cytokine that regulates thrombopoiesis. Cytokines Cell Mol Ther 6: Pachner AR Anti-IFNB antibodies in IFNB-treated MS patients. Neurology 61 (Suppl. 5):S1 S5. Pachner AR, Dail D, Pak E, Narayan K The importance of measuring IFNbeta bioactivity: monitoring in MS patients and the effect of anti-ifnbeta antibodies. J Neuroimmunol 166:

58 766 BENDTZEN Pestka S, Krause CD, Walter MR Interferons, interferonlike cytokines, and their receptors. Immunol Rev 202:8 32. Petersen B, Bendtzen K, Koch-Henriksen N, Ravnborg M, Ross C, Sorensen PS, The Danish Multiple Sclerosis Group Persistence of neutralizing antibodies after discontinuation of IFNbeta therapy in patients with relapsing-remitting multiple sclerosis. Mult Scler 12: Pipe SW The promise and challenges of bioengineered recombinant clotting factors. J Thromb Haemost 3: Polman CH, Schellekens H, Killestein J Neutralizing antibodies to interferon-beta may persist after cessation of therapy: what impact could they have? Mult Scler 12: Ravnborg M, Bendtzen K, Christensen O, Jensen PEH, Hesse D, Tovey MG, Sorensen PS Treatment with azathioprine and cyclic methylprednisolone has little or no effect on bioactivity in anti-interferon beta antibody-positive patients with multiple sclerosis. Mult Scler 15: Ross C, Clemmesen KM, Svenson M, Sorensen PS, Koch-Henriksen N, Skovgaard GL, Bendtzen K Immunogenicity of interferon-b in multiple sclerosis patients: influence of preparation, dosage, dose frequency, and route of administration. Danish Multiple Sclerosis Study Group. Ann Neurol 48: Ross C, Svenson M, Clemmesen KM, Sorensen PS, Koch-Henriksen N, Bendtzen K Measuring and evaluating interferon-beta-induced antibodies in patients with multiple sclerosis. Mult Scler 12: Rutgeerts P, Vermeire S, van Assche G Predicting the response to infliximab from trough serum levels (commentary). Gut 59:7 8. Sorensen PS, Koch-Henriksen N, Ross C, Clemmesen KM, Bendtzen K, The Danish Multiple Sclerosis Study Group Appearance and disappearance of neutralizing antibodies during interferon-beta therapy. Neurology 65: Sorensen PS, Ross C, Clemmesen KM, Bendtzen K, Frederiksen JL, Jensen K, Kristensen O, Petersen T, Rasmussen S, Ravnborg M, Stenager E, Koch-Henriksen N, Danish Multiple Sclerosis Study Group Clinical importance of neutralising antibodies against interferon beta in patients with relapsing-remitting multiple sclerosis. Lancet 362: Sorensen PS, Tscherning T, Mathiesen HK, Langkilde AR, Ross C, Ravnborg M, Bendtzen K Neutralizing antibodies hamper IFNbeta bioactivity and treatment effect on MRI in patients with MS. Neurology 67: Strand V, Kimberly R, Isaacs JD Biologic therapies in rheumatology: lessons learned, future directions. Nat Rev Drug Discov 6: Address correspondence to: Dr. Klaus Bendtzen Institute for Inflammation Research (IIR) Rigshospitalet University Hospital DK-2200 Copenhagen Denmark Received 19 July 2010/Accepted 19 July kben@dbmail.dk

59 JOURNAL OF INTERFERON & CYTOKINE RESEARCH Volume 30, Number 10, 2010 ª Mary Ann Liebert, Inc. DOI: /jir On the Role of Aggregates in the Immunogenicity of Recombinant Human Interferon Beta in Patients with Multiple Sclerosis Miranda M.C. van Beers, 1,2 Wim Jiskoot, 2 and Huub Schellekens 1 Like many other therapeutic proteins, recombinant human interferon beta (rhifn-b) elicits undesirable immune responses. rhifn-b-treated multiple sclerosis patients may form binding antibodies and neutralizing antibodies (NAbs), with the latter being responsible for inhibition of the therapeutic effect of the protein. The incidence of binding antibodies and NAbs against rhifn-b as well as the titer and persistence of NAbs differ among the marketed products. The proportion of patients forming antibodies against rhifn-b-1b is higher than that against rhifn-b-1a, which is likely explained by the differences in protein structure and aggregation behavior between the 2 types of rhifn-b. Here, we summarize the different factors influencing the immunogenicity of rhifn-b in patients with multiple sclerosis and discuss the role played by rhifn-b aggregates. Introduction Protein drugs have added a new dimension to the treatment of several diseases, eg, monoclonal antibodies in the treatment of rheumatoid arthritis (Tracey and others 2008), erythropoietin in anemia (Parnham and others 2007), interferon alpha (IFN-a) in hepatitis (Spiegel 1989), and IFN-b in multiple sclerosis (MS) (Runkel and others 1998). Therapeutic proteins can serve as valuable supplements or full substitutes for existing treatments. Although protein drugs fulfill a medical need, their application is often limited due to their ability to elicit an immune response in patients. Practically all therapeutic proteins show immunogenicity, including recombinant human IFN-b (rhifn-b) (Giovannoni and others 2002; Sorensen 2008b). If an immune response is triggered after (often repetitive) administration of the protein, the resulting antibodies can inhibit the efficacy of the drug (Hartung and others 2005). Binding antibodies (BAbs) alter the pharmacokinetics of the protein, whereas neutralizing antibodies (NAbs) interfere with target receptor binding or may even bind to the endogenous homologous protein with severe clinical consequences (Casadevall and others 2002). In principle, recombinant human proteins are self-proteins and should therefore be tolerated, so what causes them to elicit an immune response? There is no definite answer to this, as numerous factors are found to play a role (Schellekens 2002a). Factors affecting the immune response can roughly be divided into product-related and nonproductrelated factors. The interplay between the characteristics of a protein product and other, nonproduct-related factors ultimately determines its immunogenic potency. Nonproduct-related factors include route and frequency of administration, concomitant medication, patient features, disease state, and antibody assay technology. One example of a product-related factor is the degree of foreignness of the therapeutic versus the natural human protein (Schellekens 2002b; Hermeling and others 2004). This may include differences in amino acid composition as well as post-translational modification such as glycosylation. Further, it has been recognized that changes in the formulation or packaging can decrease the stability of the protein product and enhance the presence of aggregates, some of which are linked to immunogenicity (Hermeling and others 2004; Rosenberg 2006). Aggregates may consist of protein drug alone, or as a mixture of protein and excipients or impurities such as nanoand microparticles shed from pumps, pipes, vessels, filters, or primary packaging material (Carpenter and others 2009). To address why certain aggregates are immunogenic, various authors hypothesized on the underlying immunological mechanism (Schellekens 2002a, 2008; Rosenberg 2006; Sauerborn and others 2010). In general, it is believed that protein complexes forming a highly organized, repetitive array of native-like antigens are particularly immunogenic (Dintzis and others 1976; Bachmann and Zinkernagel 1997). High antigen repetitiveness allows very efficient cross-linking of 1 Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS), Utrecht University, Utrecht, The Netherlands. 2 Division of Drug Delivery Technology, Leiden/Amsterdam Centre for Drug Research (LACDR), Leiden University, Leiden, The Netherlands. 767

60 768 VAN BEERS, JISKOOT, AND SCHELLEKENS B cell receptors, which triggers B cell activation independent of T cell help and eventually breaks the immune tolerance for a self-protein (Chackerian and others 2002). Such breaking of tolerance is observed in MS patients treated with rhifn-b. The appearance of NAbs is a severe problem in rhifn-b treatment and, also in this case, aggregates are suggested to be involved (Runkel and others 1998; Hartung and others 2005; Hermeling and others 2005b). In this review, the clinical immunogenicity of rhifn-b and its consequences are described, followed by the factors that influence the immunogenicity. Moreover, we discuss the differences between the commercial rhifn-b products, thereby focusing on the role of aggregates in rhifn-b immunogenicity. Clinical Observations on rhifn-b Immunogenicity rhifn-b was introduced by Berlex in 1993 as a therapeutic protein produced in Escherichia coli under the U.S. brand name Betaseron Ò (FDA 1993). It is indicated for use in patients with relapsing-remitting MS to reduce the frequency of clinical exacerbations. The first reports on antibody formation against Betaseron in MS patients were published in the same year (Knobler and others 1993; The IFNB Multiple Sclerosis Study Group 1993). Not only BAbs but also NAbs developed in these patients. Titers were highly variable, peaked around 2 years of treatment and tended to decrease upon prolongation of therapy. The biological impact of NAbs against rhifn-b-1b was detected as a decrease in therapeutic effectiveness after about 18 months of treatment (The IFNB Multiple Sclerosis Study Group and the University of British Columbia MS/MRI Analysis Group 1996; Rudick 2003). The Betaseron protein and the recently marketed Extavia Ò (Novartis) protein are of the rhifn-b-1b type, which differs from the natural human protein at 3 points: (1) it is not glycosylated, (2) it lacks the N-terminal methionine, and (3) the cysteine residue normally found at position 17 has been replaced with a serine residue (Runkel and others 1998). These differences are not present in the later developed commercial products, Rebif Ò (Serono) and Avonex Ò (Biogen Idec), which are both of the rhifn-b-1a type and produced in Chinese hamster ovary (CHO) cells. Despite their high similarity with natural human IFN-b, the rhifn-b-1a products are, nevertheless, able to elicit immune responses in MS patients (Rudick and others 1998; Ross and others 2000; SPECTRIMS Study Group 2001). The incidence of BAbs and NAbs in MS patients is dependent on the rhifn-b product (Table 1). Clinical studies have demonstrated different percentages of patients developing BAbs and NAbs against specific products, depending on the techniques used to measure antibody levels and cutoff points for antibody positivity. However, overall, the incidence of antibodies in rhifn-b-1b is higher than in rhifn-b-1a-treated patients, with Rebif being more immunogenic than Avonex. NAbs, which are a specific population of BAbs, are present in patients that have relatively high immunoglobulin G (IgG) titers and high frequencies of the subclasses IgG2 and IgG4 (Deisenhammer and others 2001). The rhifn-b products also differ with respect to the levels and persistence of NAb titers (Table 1). The highest proportion of patients with very high titers is reported after treatment with Rebif and this product has also the highest rate of NAb-positivity among BAb-positive patients (Gneiss and others 2006; Sominanda and others 2007). Both ontreatment and after discontinuation of therapy, NAbs can be lost over the course of many years, but the probability of reversion is higher with lower titers (Petersen and others 2006; Hurwitz 2008). Low NAb levels are probably the reason why rhifn-b-1b-treated patients have a relative high chance of seroconversion to NAb-negativity (Gneiss and others 2004; Sorensen and others 2005). Patients treated with Betaferon Ò were reported to show a rapid increase in antirhifn-b-1b antibodies, which peaked 3 9 months after therapy initiation and then slowly but progressively declined in a high proportion of patients (Perini and others 2001). In general, NAbs develop within 3 18 months of therapy, while clinical effects on magnetic resonance imaging (MRI) and relapse rate become apparent the following year (Sorensen and others 2008). More marked and prolonged NAb levels are positively correlated with an increase in exacerbation rate as well as an increase in the numbers of new and enlarging MRI lesions (The IFNB Multiple Sclerosis Study Group and the University of British Columbia MS/MRI Analysis Group 1996; Rudick 2003). This is caused by NAbs inhibiting the binding of rhifn-b to its receptor, thus preventing biologic effects and abrogating the efficacy of the drug (The IFNB Multiple Sclerosis Study Group and the University of British Columbia MS/MRI Analysis Group 1996; Rudick and others 1998; Francis and others 2005; Kappos and others 2005). Antibodies against rhifn-b-1a and -1b are cross-reactive and potentially bind to the natural human protein (Shapiro and others 2006). Conventionally, patients are considered NAb positive if 2 consecutive measurements with a time interval of 3 months or more yield titers of at least 20 neutralizing units with the cytopathic effect assay (Hemmer and others 2005). NAb titers higher than 150 ten-fold reduction units per milliliter as measured with the Myxovirus protein A induction bioassay are suggested to result in loss of in vivo bioactivity (Sominanda and others 2009). Patients who are NAb negative or have very high NAb titers often remain stable, whereas moderate titers may lead to seroconversion with regained bioactivity (Gneiss and others 2004). Therefore, it is suggested to change the therapy of patients with high titers in particular. If the therapy is changed to another rhifn-b preparation, a washout period of 2 3 months is recommended to regain therapeutic effect before switching (Perini and others 2001). Besides switching to a less immunogenic rhifn-b product, other suggested strategies include tolerance induction by administering escalating doses of rhifn-b, switching to another immunomodulatory agent (eg, glatiramer acetate), or concomitant immunosuppression (eg, mitoxantrone) (Hemmer and others 2005). Factors Influencing rhifn-b Immunogenicity As with other therapeutic proteins, there are numerous factors that affect the immunogenicity of rhifn-b in patients. Studying the influence of each of these aspects is complicated by the high variability of antibody assays and methods of expressing antibody titers (Hurwitz 2008). Lack of assay standardization and difference in timing of antibody testing after the start of treatment hamper comparison of rhifn-b immunogenicity data from different clinical studies, and may also explain part of the variability in antibody incidence

61 AGGREGATES AND IMMUNOGENICITY OF RHIFN-b IN MS 769 Table 1. Immunogenicity of Commercial Recombinant Human Interferon Beta Products as Shown by Different Clinical Studies Using Various Techniques for Binding Antibody and Neutralizing Antibody Measurement % patients with Brand name a Type Administration BAbs b NAbs Relative NAb titer height c Relative NAb persistence d Betaferon -1b 1.6 or 8.0 MIU s.c. every e Low to moderate Low Betaseron other day Rebif -1a 22 or 44 mg s.c. 1 or 3 weekly f Moderate to high Moderate to high Avonex -1a 30 or 60 mg i.m. 1 weekly g Moderate to high Moderate to high a Extavia (recombinant human interferon beta-1b) was not included in the table since published data on its clinical immunogenicity are lacking due to its recent Food and Drug Administration approval in b Kivisäkk and others (2000); Ross and others (2000); Perini and others (2001); Gneiss and others (2006). c Bertolotto and others (2000); Kivisäkk and others (2000); Gneiss and others (2006); Sominanda and others (2007); Hurwitz (2008). d Sorensen and others (2005); Bellomi and others (2003); Gneiss and others (2004); Hurwitz (2008); Sominanda and others (2009). e The IFNB Multiple Sclerosis Study Group (1993); The IFNB Multiple Sclerosis Study Group and the University of British Columbia MS/ MRI Analysis Group (1996); European Study Group on interferon beta-1b in secondary progressive MS (1998); Sominanda and others (2007). f PRISMS Study Group (1998); OWIMS Study Group (1999); SPECTRIMS Study Group (2001); Francis and others (2005); Sominanda and others (2007). g Jacobs and others (2000); Clanet and others (2002); Sominanda and others (2007). Abbreviations: BAbs, binding antibodies; i.m., intramuscular; MIU, million international units; NAbs, neutralizing antibodies; s.c., subcutaneous. observed by different laboratories (Table 1) (Gneiss and others 2006). For instance, NAb titers depend on the rhifn-b product that is used in the assay; ie, using rhifn-b-1a results in *3-fold higher titers than using rhifn-b-1b (Bertolotto and others 2000; Files and others 2007). The human IFN-b protein and human anti-human-ifn-b antibody WHO reagents may be used for reference (Grossberg and others 2001). Further, on request of the European regulatory authorities a Myxovirus protein A mrna induction inhibition bioassay has been developed as a common standardized assay to quantify rhifn-b NAbs (Hemmer and others 2005). In 2006, this resulted in the establishment of the NABINMS ( Neutralizing Antibodies on Interferon beta in Multiple Sclerosis ) consortium, with research institutes, hospitals, and industrial partners collaborating to understand and prevent antibody formation against rhifn-b by first of all validating standard antibody assays. Efforts are made to compare inter- and intralaboratory consistency of BAb assays and the relation between BAbs and NAbs (Gilli and others 2006). Currently, the field of assay development is rapidly moving forward and uniform methods are established that facilitate ongoing studies on factors influencing immunogenicity (Lallemand and others 2010). One of those factors that can contribute to unwanted immunogenicity is the nature of the disease that is being treated. Although MS is an autoimmune disease, it is unclear if this has an effect on the perceived immunogenicity of rhifnb. Not only MS patients but also cancer patients treated with rhifn-b showed NAb formation (Hartung and others 2005). Patient features play a role in the immunogenicity of rhifn-b. The intrinsic sensitivity for rhifn-b varies among patients. Baseline expression of type I IFN and type I IFNinduced genes (van Baarsen and others 2008; Comabella and others 2009) as well as other genetic factors (Byun and others 2008; Vandenbroeck and Matute 2008) affect the therapeutic response, and possibly also the immune response to rhifn-b. To date, no correlation could be found between the time of appearance and titers of anti-rhifn-b antibodies, patient s age and gender, disease duration, and relapse rate before therapy (Perini and others 2001; Sorensen and others 2003). In 2008, however, researchers demonstrated that both the patient s human leukocyte antigen type and pretreatment levels of the type I IFN receptor (IFNAR) directly relate to rhifn-b immunogenicity (Gilli and others 2008; Hoffmann and others 2008; Sorensen 2008a). Other factors that modulate the immune response to an antigen are the route of application, dose, treatment duration, and administration of concomitant medication. A clear correlation has been shown between the proportion of NAbpositive patients and the weekly rhifn-b dose, by a dose comparison trial for Avonex (Clanet and others 2002), and the OWIMS trial reported a greater incidence of NAbs with 44 mg than with 22 mg Rebif (15% versus 5%) when administered once weekly (OWIMS Study Group 1999). Also, 58% of the patients treated subcutaneously with 22 mg Rebif once weekly developed BAbs within 1 year, compared with 89% of those treated 3 times weekly (Bertolotto and others 2004). More frequent rhifn-b injections generally result in higher NAb rates (Ross and others 2000; Perini and others 2001). Further, the route of administration can make a difference, as was demonstrated by the significant delay in antibody appearance and reduction in antibody levels against Betaferon if injected intramuscularly weekly instead of subcutaneously every other day (Perini and others 2001). For other therapeutic proteins it is known that subcutaneous administration is more immunogenic compared to intravenous administration (Peng and others 2009). Importantly, the protein itself has properties that determine the degree of immunogenicity. Human IFN-b is a cytokine with anti-inflammatory, antitumor, and antiviral functions that is mainly produced by macrophages and epithelial and fibroblast cells. rhifn-b is biologically active in the human immune system and its immunomodulating properties could influence antidrug antibody responses (Filipe and others 2010). Last but not least, immune responses can be enhanced by contaminants and impurities present in the protein product, such as impurities from host cells acting as adjuvants and non-native protein fractions (Ponce and others 2009). Due to major improvements in purifying therapeutic proteins, at present the likelihood of impurities from E. coli and the CHO cell lines used to produce rhifn-b-1b and rhifn-b-1a,

62 770 VAN BEERS, JISKOOT, AND SCHELLEKENS respectively, contributing to immunogenicity is low. Still, impurities may be present in the final preparation due to degradation of the protein during, eg, manufacture, fill-finish, storage, or handling (Hermeling and others 2004). Not much is known about the effect of chemical degradation of rhifn-b, such as oxidation and deamidation, on its immunogenicity. Chemical modifications may lead to the formation of new epitopes or the exposure of otherwise buried epitopes that are not recognized by the patient s immune system. For the patient it is critical whether or not the resulting antibodies cross-react with the native protein. Although a chemical modification itself might not induce cross-reacting antibodies, the resulting protein aggregates may do so, as was shown for metal catalyzed oxidized rhifn-a in transgenic, immune tolerant mice (Hermeling and others 2005a, 2006). Besides chemical modifications, the protein can undergo physical degradation by extensive exposure to light, heat, or stresses such as shear and shaking. Physical degradation of proteins also occurs through contact with foreign materials (ie, stainless steel pumps, glass containers, silicone oil syringe lubricant, etc.) (Bee and others 2009). This contact may induce unfolding and misfolding, and especially the subsequently formed aggregates can be highly immunogenic. Protein aggregates not only act as adjuvants by increasing the classical immune response against foreign proteins (Dresser 1962; White and others 2008), but can also break the immune tolerance for the specific protein (Moore and Leppert 1980; Hermeling and others 2005a). As for other proteins, the stability of rhifn-b against chemical and physical degradation and aggregation largely depends on its formulation and dosage form. These productrelated aspects, together with structural differences regarding amino acid sequence and glycosylation, may explain the differences in immunogenicity observed for the different rhifn-b preparations (Giovannoni and others 2002). The clinical use of rhifn-b products with distinct characteristics provide a unique opportunity to study the relation between protein structure and in vivo immunogenicity, in which aggregates seem to play a crucial role. Aggregation and Immunogenicity of rhifn-b FIG. 1. Silver-stained sodium dodecyl sulfate polyacrylamide gel electrophoresis gel (10% acrylamide) under nonreducing conditions of recombinant human interferon beta 1a (rhifn-b-1a) (1a), and Betaseron (1b). Note that the nonglycosylated Betaseron-rhIFN-b-1b migrates faster than the glycosylated rhifn-b-1a. Numbers on the left represent band positions of the molecular weight markers in kda. The highly immunogenic rhifn-b-1b molecule lacks Met-1 in comparison with natural human IFN-b, and its Cys-17 has been mutated to Ser-17. The primary amino acid sequence of rhifn-b-1a is identical to that of human IFN-b. The higher degree of foreignness with respect to primary protein structure probably does not explain the higher incidence of antibodies among rhifn-b-1b than rhifn-b-1a-treated patients (Table 1), since the antibodies are cross-reactive (Bertolotto and others 2000; Perini and others 2001). rhifn-b-1a possesses an N-linked glycan, which explains its slower migration during sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) as compared with rhifn-b-1b lacking the glycan (Fig. 1). The glycan structure of rhifn-b-1a may slightly differ from that of endogenous human IFN-b due to the production in CHO cells. Glycosylation is essential for preserving the structural stability of glycoproteins (Wang and others 1996). X-ray crystallography revealed that the rhifn-b-1a glycan formed hydrogen bonds with the peptide backbone and thus prevented solvent exposure of a hydrophobic area on the protein surface (Runkel and others 1998). Exposure of hydrophobic residues makes a protein more prone to chemical and physical degradation and aggregation. Indeed, deglycosylation of rhifn-b-1a was found to increase sensitivity to thermal denaturation and diminish biological activity, due to the formation of insoluble, disulfide-linked aggregates (Runkel and others 1998). Size exclusion chromatography showed that 98% of the glycosylated rhifn-b-1a protein eluted as a monomer, whereas *60% of the rhifn-b-1b protein consisted of large, soluble aggregates with an apparent molecular weight of over 600 kda (Runkel and others 1998). SDS-PAGE and an antiviral activity assay demonstrated that the rhifn-b-1b aggregates were noncovalent and had reduced bioactivity, respectively. The high aggregate percentage of Betaseron/Betaferon is likely to play a key role in its pronounced immunogenicity, which is consistent with several preclinical studies, indicating the importance of molecular structure and the presence of aggregates in IFN type I immunogenicity (Braun and others 1997; Jones and others 1998; Hermeling and others 2005a, 2006). In fact, dissociation of murine IFN-b aggregates with high hydrostatic pressure drastically reduced its immunogenicity in wildtype mice (Seefeldt and others 2008). Likewise, high-pressure-treated aggregate-free rhifn-b-1b did not induce antibodies in transgenic mice immune tolerant for human IFN-b, whereas Betaseron/Betaferon did (Cleland and others 2008). Despite the presence of the carbohydrate moiety, aggregates are also observed in rhifn-b-1a products. The relatively few aggregates (2%) in rhifn-b-1a detected with size exclusion chromatography were disulfide linked, probably due to involvement of Cys-17 in disulfide scrambling (Karpusas and others 1997, 1998). The glycan of rhifn-b-1a